Compare commits

...

233 Commits

Author SHA1 Message Date
Mark Qvist e95e9e6a89 Updated readme 2022-04-28 15:46:53 +02:00
Mark Qvist e8024e560f Updated manual 2022-04-28 15:31:38 +02:00
Mark Qvist 8cbbcb0fe9 Updated documentation 2022-04-28 15:12:37 +02:00
Mark Qvist 8e4bfbbd94 Updated documentation 2022-04-28 14:56:52 +02:00
Mark Qvist 600bd0e64d Updated documentation 2022-04-28 14:17:12 +02:00
Mark Qvist 123fd1de92 Configure IFAC size in bits 2022-04-28 11:51:04 +02:00
Mark Qvist 29df5950c8 Updated documentation 2022-04-28 11:50:46 +02:00
Mark Qvist b8ca89c2b6 Cleanup 2022-04-28 10:58:26 +02:00
Mark Qvist 79725a1637 Cleanup 2022-04-28 10:56:19 +02:00
Mark Qvist 1a2da0d7c7 Drop IFAC packets on non-IFAC interfaces 2022-04-28 10:38:34 +02:00
Mark Qvist fe065f8bdd Updated documentation 2022-04-28 10:19:43 +02:00
Mark Qvist 5d90ea565a Implemented interface authentication and virtual network segmentation 2022-04-27 19:00:09 +02:00
Mark Qvist b701cdd07f Generalised transport transmit handler 2022-04-27 13:50:56 +02:00
Mark Qvist 8e5b3b4e83 Cleanup 2022-04-27 13:31:44 +02:00
Mark Qvist 24b7cb777f Cleanup 2022-04-27 13:31:07 +02:00
Mark Qvist cf1ca01a3b Configuration support for interface access codes 2022-04-27 13:21:53 +02:00
Mark Qvist 7c70f9d865 Set IFAC parameters on spawned interfaces 2022-04-27 13:20:46 +02:00
Mark Qvist 6cf9288b11 Improved AutoInterface peering timeout 2022-04-27 13:19:48 +02:00
Mark Qvist 00816b55bb Updated docs 2022-04-27 13:19:03 +02:00
Mark Qvist 3856747e31 Updated docs and manual 2022-04-20 21:25:54 +02:00
Mark Qvist 50799bd2a6 Updated docs and manual 2022-04-20 21:24:01 +02:00
Mark Qvist ecffa1a7eb Cleanup 2022-04-20 20:28:14 +02:00
Mark Qvist 9fef53d083 Updated propagation timing 2022-04-20 20:06:44 +02:00
Mark Qvist 0db64610b1 Added reverse path detection 2022-04-20 19:29:25 +02:00
Mark Qvist 4af14a712c Added timeout to rnpath utility 2022-04-20 13:40:07 +02:00
Mark Qvist 402b5fc461 Random hash length from truncated hash length 2022-04-20 13:08:21 +02:00
Mark Qvist 38aeb1ab3b Improved logging 2022-04-20 12:58:00 +02:00
Mark Qvist b0a21b3aa9 Improved logging 2022-04-20 12:56:43 +02:00
Mark Qvist 5e6a1add6b Improved logging and rnpath output 2022-04-20 11:22:52 +02:00
Mark Qvist 104b186047 Added drop path ability to rnpath utility 2022-04-20 11:12:21 +02:00
Mark Qvist 6d23da360d Added path table output to rnpath utility 2022-04-20 10:40:51 +02:00
Mark Qvist 1be00a5c41 Improved announce logging 2022-04-20 10:08:55 +02:00
Mark Qvist 71e5eef8c1 Improved announce logging 2022-04-20 10:08:30 +02:00
Mark Qvist b3a439993d Improved rnstatus output 2022-04-20 09:59:58 +02:00
Mark Qvist 5606b64317 Improved transport logging 2022-04-20 09:24:48 +02:00
Mark Qvist 3d38ef27d4 Improved announce logging 2022-04-20 09:04:12 +02:00
Mark Qvist 93fa8e7240 Updated readme 2022-04-18 19:27:59 +02:00
Mark Qvist d53e8cf037 Bitrate guess on TCP and UDP interfaces 2022-04-18 18:09:31 +02:00
Mark Qvist be820b1965 Updated roadmap 2022-04-18 17:17:43 +02:00
Mark Qvist 425cf66cf7 Updated roadmap 2022-04-18 17:08:06 +02:00
Mark Qvist 8d294df3bb Updated readme formatting 2022-04-18 17:01:41 +02:00
Mark Qvist da297aeb64 Changed log level 2022-04-18 16:54:35 +02:00
Mark Qvist 282239fc57 Added interface announce queue stats to rnstatus 2022-04-18 16:41:38 +02:00
Mark Qvist 222437d851 Version bump 2022-04-18 16:26:08 +02:00
Mark Qvist c9de260e00 Cleanup 2022-04-18 16:25:11 +02:00
Mark Qvist 31104c6e9c Implemented bandwidth-based announce propagation calculation 2022-04-18 16:23:24 +02:00
Mark Qvist 64593e27be Added announce cap setting 2022-04-17 20:14:20 +02:00
Mark Qvist bac33d4e8b Updated documentation 2022-04-17 20:12:23 +02:00
Mark Qvist 124ec006b4 Auto-set sensible interface rate defaults 2022-04-17 19:35:31 +02:00
Mark Qvist dd55899775 Improved I2P interface display in rnstatus 2022-04-17 19:35:05 +02:00
Mark Qvist cc0c01661d Fixed a possible race condition in Transport startup when a local shared instance was restarted and client apps reconnect 2022-04-17 19:34:12 +02:00
Mark Qvist 5f36c8601f Improved rnstatus utility display 2022-04-17 19:08:48 +02:00
Mark Qvist 2f71296816 Basic interface rate estimation 2022-04-17 19:07:32 +02:00
Mark Qvist 7923322d92 Linearized announce propagation delays for supporting up to 128 hops 2022-04-17 18:27:35 +02:00
Mark Qvist fef5ed6bad Cleanup 2022-04-17 17:37:27 +02:00
Mark Qvist 059b0743ef Improved rnstatus util display 2022-04-17 17:30:39 +02:00
Mark Qvist 4d4d39651f Improved rnstatus util display 2022-04-16 23:26:57 +02:00
Mark Qvist 6a1e6417bb Updated logging 2022-04-16 16:44:10 +02:00
Mark Qvist ed20b27e9d Updated interface documentation 2022-04-16 16:43:41 +02:00
Mark Qvist 39f1258d0e Added interface mode config shorthands 2022-04-16 16:42:59 +02:00
Mark Qvist 03d3478b5e Added interface mode shorthand 2022-04-15 22:13:59 +02:00
Mark Qvist b35122a872 Fixed typo 2022-04-15 22:13:41 +02:00
Mark Qvist ae240f4697 Updated manual 2022-04-07 20:22:48 +02:00
Mark Qvist 4e1cdc638f Updated docs 2022-04-07 20:21:15 +02:00
Mark Qvist fc83c5b082 Updated documentation 2022-04-07 20:15:35 +02:00
Mark Qvist ee90605b30 Tweaked decay constant 2022-04-07 19:00:07 +02:00
Mark Qvist 3684fe502f Updated documentation 2022-04-07 18:59:42 +02:00
Mark Qvist d4aeb85191 Merge branch 'master' of github.com:markqvist/Reticulum 2022-04-06 15:51:31 +02:00
Mark Qvist 04540f6e48 Changed config description 2022-04-06 15:51:27 +02:00
markqvist 0db7eb1408 Update README.md 2022-04-06 15:49:39 +02:00
markqvist 5fe55243c6 Update README.md 2022-04-05 11:38:04 +02:00
Mark Qvist b56830b36e Moved docs section contributed by @4c3e 2022-04-02 10:05:46 +02:00
markqvist e3ea61c944 Merge pull request #33 from 4c3e/patch-1
[WIP] First draft of internet bridge section
2022-04-02 09:32:26 +02:00
4c3e 02f9c32da7 First draft of internet bridge section
Open to any and all criticism here, writing documentation isn't my strong suit. But here is a decent start at explaining the differences between the transports. Also I believe there should be a way to "compile" this markdown to html, not exactly sure how to accomplish that, will try to look into this further in the weekend.
2022-04-01 23:15:24 -04:00
Mark Qvist a4a9a1dd53 Updated readme 2022-04-01 18:22:17 +02:00
Mark Qvist d7f9b30638 Updated readme 2022-04-01 18:03:00 +02:00
Mark Qvist 02676d3b25 Updated readme 2022-04-01 18:01:32 +02:00
Mark Qvist 089612bfc1 Updated readme 2022-04-01 17:55:48 +02:00
Mark Qvist ca345b20ff Updated readme 2022-04-01 17:52:56 +02:00
Mark Qvist 3b5973085f Updated readme 2022-04-01 17:20:24 +02:00
Mark Qvist dc6877927e Updated license headers 2022-04-01 17:18:18 +02:00
Mark Qvist f01d838e17 Updated readme 2022-04-01 17:13:47 +02:00
Mark Qvist 9da6d39f64 Updated readme 2022-04-01 17:11:52 +02:00
Mark Qvist d17fbf1f34 Merge branch 'master' of github.com:markqvist/Reticulum 2022-03-28 15:20:35 +02:00
Mark Qvist 7398e312fc Updated version 2022-03-28 15:20:14 +02:00
markqvist 82fc8720ad Update README.md 2022-03-26 22:45:46 +01:00
markqvist 4b9686c31a Update README.md 2022-03-26 22:44:32 +01:00
Mark Qvist 86a5b3302a Updated readme 2022-03-25 20:15:11 +01:00
Mark Qvist c990aae648 Updated license 2022-03-25 20:07:09 +01:00
Mark Qvist 3051b6897d Updated filtering rules. Fixes #18. 2022-03-15 14:55:47 +01:00
Mark Qvist 550dfd44cb Improved cryptography API compatibility 2022-03-08 00:38:51 +01:00
Mark Qvist 95d3346da6 Fixed I2P interface missing attribute 2022-02-26 21:37:50 +01:00
Mark Qvist d4aabc8b89 Added I2P base32 address output to rnstatus utility 2022-02-26 21:04:54 +01:00
Mark Qvist d487609dcf Updated docs and manual 2022-02-26 20:46:14 +01:00
Mark Qvist c96c82f1d1 Updated manual 2022-02-26 19:35:37 +01:00
Mark Qvist cb023cde40 Fixed potential race condition in resource assembly 2022-02-26 18:27:11 +01:00
Mark Qvist 17be289f37 Updated documentation and manual 2022-02-25 22:34:41 +01:00
Mark Qvist b8105e23ff Fixed TCP interface mode reference 2022-02-25 22:10:55 +01:00
Mark Qvist f378d09cbe Updated documentation and manual 2022-02-25 21:50:03 +01:00
Mark Qvist 4dfa62833c Restructured default config 2022-02-25 21:48:25 +01:00
Mark Qvist 2ec6d3ba6c Updated I2P interface documentation 2022-02-25 21:41:43 +01:00
Mark Qvist 15d027e11e Restructured default config and added config example to rnsd 2022-02-25 21:41:24 +01:00
Mark Qvist 87a274d177 Added I2P interface documentation 2022-02-25 21:26:34 +01:00
Mark Qvist f8272793b4 Tuned AutoInterface timeouts 2022-02-25 20:29:47 +01:00
Mark Qvist 3a215be859 Interface mode defaults 2022-02-25 18:56:09 +01:00
Mark Qvist 0e1279d012 Added Access Point interface mode 2022-02-25 18:47:55 +01:00
Mark Qvist 8ec356a28e Interface outbound option enabled by default 2022-02-25 13:31:52 +01:00
Mark Qvist 49d7808835 i2plib license 2022-02-24 01:45:42 +01:00
Mark Qvist 48184134e4 Improved I2P Interface 2022-02-24 01:30:10 +01:00
Mark Qvist 987ff0658b Version bump 2022-02-23 22:53:16 +01:00
Mark Qvist 27dea7c524 Implemented I2PInterface 2022-02-23 22:43:08 +01:00
Mark Qvist 9c6fd132d4 Work on I2P Interface 2022-02-23 22:15:06 +01:00
Mark Qvist 8d58bb62ab Work on I2P Interface 2022-02-23 21:47:30 +01:00
Mark Qvist c357f7a94e Work on I2P Interface 2022-02-23 21:39:29 +01:00
Mark Qvist 4b3ead3db2 Work on I2P Interface 2022-02-23 21:29:18 +01:00
Mark Qvist b62e9af5d4 Work on I2P Interface 2022-02-23 21:19:43 +01:00
Mark Qvist fa82989a2e Preliminary I2P Interface support 2022-02-23 17:40:31 +01:00
Mark Qvist 07a65609b4 Updated documentation 2022-02-22 21:39:16 +01:00
Mark Qvist 257bd95da8 AutoInterface carrier loss detection 2022-02-22 20:16:02 +01:00
Mark Qvist 1ccfa9079c Work on AutoInterface recovery on WiFi carrier loss 2022-02-22 14:49:43 +01:00
Mark Qvist 57226201ff Fixed I2P tunneled config keyword 2022-02-22 14:45:36 +01:00
Mark Qvist d9419cd895 Merge branch 'master' of github.com:markqvist/Reticulum 2022-02-22 14:43:19 +01:00
Mark Qvist aae10ede72 Work on AutoInterface recovery on WiFi carrier loss 2022-02-22 14:43:14 +01:00
Mark Qvist 291b3056cd Updated docs 2022-02-01 23:07:18 +01:00
Mark Qvist 3f53c89d32 Added I2P-tunneled mode to TCP interfaces 2022-01-31 23:31:29 +01:00
Mark Qvist 05288d7c97 Updated documentation for release 2022-01-27 23:49:24 +01:00
Mark Qvist b403441074 Version updated 2022-01-27 23:30:44 +01:00
Mark Qvist d3a23e3b00 RNodeInterface firmware version check 2022-01-22 22:46:47 +01:00
markqvist 329d83587e Merge pull request #10 from 4c3e/4c3e-osx-nameerror-fix
OSX NameError fix
2022-01-22 21:39:01 +01:00
Mark Qvist 0a4dd64434 Improved support for ESP32-based RNodes 2022-01-22 21:36:49 +01:00
4c3e b96cbf1014 OSX NameError fix
Had the following error when trying to run Reticulum on OSX High Sierra:
line 115, in set_timeouts_osx
    sock.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
NameError: name 'sock' is not defined
This fix resolved the problem for me.
2022-01-21 01:34:55 +00:00
Mark Qvist 485558cd6b Updated documentation and manual 2022-01-14 22:21:13 +01:00
Mark Qvist 8d93867a22 Updated manual 2022-01-14 22:16:02 +01:00
markqvist 6b20a98adc Update README.md 2022-01-12 12:53:18 +01:00
Mark Qvist f3d04ba90f Improved AutoInterface handling on Android 2022-01-12 12:12:04 +01:00
Mark Qvist 1d2564cedb Interface import on Android 2022-01-12 12:02:00 +01:00
Mark Qvist bec8473695 Better Android detection 2022-01-12 11:50:03 +01:00
Mark Qvist 25620415a0 Updated platform utils 2022-01-12 11:18:24 +01:00
Mark Qvist b6df952995 Platform version check for Windows 2022-01-12 10:16:59 +01:00
Mark Qvist a72aaf12ca Platform version check for Windows 2022-01-12 10:07:44 +01:00
Mark Qvist b978a993b2 Version update 2022-01-11 03:07:34 +01:00
Mark Qvist 5ae00264e8 Preliminaly ESP32 support for RNodeInterface 2022-01-11 03:07:03 +01:00
Mark Qvist 5396b80e80 Updated example 2022-01-11 03:06:35 +01:00
Mark Qvist fdaa58a6fa Improved malformed packet detection 2022-01-11 03:06:16 +01:00
Mark Qvist 4253175627 Cleanup 2021-12-11 20:10:31 +01:00
Mark Qvist 81158c27e4 Cleanup 2021-12-11 18:41:28 +01:00
Mark Qvist eeb424ecee Link request debug 2021-12-11 18:33:09 +01:00
Mark Qvist 0273328b23 Link proof debug 2021-12-11 18:19:51 +01:00
Mark Qvist 20dfbcf0cc Link activation time 2021-12-11 17:26:45 +01:00
Mark Qvist c96e067839 Added proper requester interface detection for path requests for destinations behind local clients. 2021-12-11 16:50:03 +01:00
Mark Qvist 9ff37543f3 Adjusted request timeout 2021-12-11 16:42:57 +01:00
Mark Qvist 974ca48cb4 Adjusted peering timing 2021-12-11 16:42:15 +01:00
Mark Qvist 167d48c8ce Updated peering timeouts 2021-12-11 15:41:34 +01:00
Mark Qvist f253b08774 Updated documentation and manual 2021-12-10 20:10:11 +01:00
Mark Qvist 1c768e9219 Removed log statement 2021-12-10 18:55:17 +01:00
Mark Qvist df39cff520 Added recovery to local shared interfaces if master RNS instance is restarted 2021-12-10 18:32:24 +01:00
Mark Qvist e1e31692d7 UDP socket contructor and doc update 2021-12-10 16:23:35 +01:00
Mark Qvist 293a834c35 Log output and cleanup 2021-12-10 14:48:30 +01:00
Mark Qvist 1bbdd9b3f5 Ignore interfaces on Darwin 2021-12-10 11:10:09 +01:00
Mark Qvist d4b6b6ee59 Ignore interfaces on Darwin 2021-12-10 11:00:48 +01:00
Mark Qvist fca03bbdce Ignore AWDL interfaces on Darwin 2021-12-10 10:58:28 +01:00
Mark Qvist 29aa4f9315 Updated AutoInterface IPv6 bind address 2021-12-10 10:35:25 +01:00
Mark Qvist d5cac30a85 Log cleanup 2021-12-10 09:48:57 +01:00
Mark Qvist 6500bc7390 Updated docs 2021-12-09 18:53:28 +01:00
Mark Qvist 81fed10855 Updated readme 2021-12-09 18:35:38 +01:00
Mark Qvist a39876106b Updated readme 2021-12-09 18:35:01 +01:00
Mark Qvist 90b39774d1 Updated readme 2021-12-09 18:34:07 +01:00
Mark Qvist 006c70cd09 Updated documentation 2021-12-09 18:12:18 +01:00
Mark Qvist 02945f960d Updated timing 2021-12-09 17:44:19 +01:00
Mark Qvist e401ec870d Updated readme 2021-12-09 17:04:20 +01:00
Mark Qvist 90174fcc28 Cleanup 2021-12-09 17:02:13 +01:00
Mark Qvist c18ebed419 Added auto interface 2021-12-09 16:07:36 +01:00
Mark Qvist 1d180a96f6 Updated dependencies 2021-12-08 20:47:14 +01:00
Mark Qvist 4241990690 Implemented AutoInterface outbound traffic and multicast discovery listeners 2021-12-08 20:46:53 +01:00
Mark Qvist 3d49076602 Compatibility with IPv6 based interfaces 2021-12-08 20:45:41 +01:00
Mark Qvist 2e0dd278b6 Updated announce example 2021-12-08 20:45:03 +01:00
Mark Qvist b432a7c7de Updated documentation 2021-12-08 20:42:48 +01:00
Mark Qvist c0383fa2b0 Updated docs 2021-12-06 19:38:03 +01:00
Mark Qvist 98d66e2ba5 Updated documentation 2021-12-06 14:10:22 +01:00
Mark Qvist 2e4fcc659c Added KISS framing option to TCP client interface 2021-12-06 13:07:12 +01:00
Mark Qvist 8fe7c19c59 Updated documentation 2021-12-05 23:31:01 +01:00
Mark Qvist 27b46c9e89 Updated documentation 2021-12-05 23:28:15 +01:00
Mark Qvist 70a3637a98 Updated documentation 2021-12-05 23:26:52 +01:00
Mark Qvist 2e0476e6b9 Updated documentation 2021-12-05 23:24:30 +01:00
Mark Qvist 39911190aa Updated documentation 2021-12-05 16:07:53 +01:00
Mark Qvist 9e9606b8cf Systemd service support and documentation update 2021-12-05 16:05:43 +01:00
Mark Qvist 8be1acee0a Added auto reconnection for disconnected serial-based devices 2021-12-05 14:35:25 +01:00
Mark Qvist ba39a69175 Timeout default structure updated 2021-12-05 11:45:13 +01:00
Mark Qvist a692d29c90 Reconnect on serial port errors for KISS interface 2021-12-05 11:44:30 +01:00
Mark Qvist 7092589388 Updated documentation 2021-12-02 18:33:00 +01:00
Mark Qvist 2d3969aa3d Added makefile 2021-12-01 19:23:19 +01:00
Mark Qvist 1443f4c104 Updated umsgpack to 2.7.1 2021-12-01 19:20:24 +01:00
Mark Qvist d2232f19ba Removed pyserial dependency 2021-12-01 14:05:33 +01:00
Mark Qvist c44c6f9086 Conditional imports for serial-based interfaces 2021-12-01 13:57:40 +01:00
Mark Qvist 259c2aa397 Conditional imports for serial-based interfaces 2021-12-01 13:39:51 +01:00
Mark Qvist 10854bfdbc Added conditional import of netifaces 2021-12-01 11:46:19 +01:00
Mark Qvist f5236878b0 Added Android platform detection 2021-12-01 11:40:44 +01:00
Mark Qvist daf72f4237 Version updated 2021-12-01 11:40:31 +01:00
Mark Qvist 652b884d72 Added conditional import of netifaces 2021-12-01 11:39:40 +01:00
Mark Qvist ea3716f48e Added Android platform detection 2021-12-01 11:39:06 +01:00
Mark Qvist 165e620043 Improved shutdown handling on interrupt. Updated gitignore. 2021-11-04 17:15:58 +01:00
Mark Qvist 58f43b163e Updated docs 2021-10-15 19:26:53 +02:00
Mark Qvist 448ea8ceb5 Added try statements for various callbacks 2021-10-15 14:36:50 +02:00
Mark Qvist f7e8fc4719 Updated docs 2021-10-14 21:06:16 +02:00
Mark Qvist 1d6c877b4c Added RSSI and SNR to echo example 2021-10-12 18:31:46 +02:00
Mark Qvist c3dcd9366d Added RSSI and SNR to echo example 2021-10-12 18:09:02 +02:00
Mark Qvist 8d01586a5a Added RSSI and SNR to echo example 2021-10-12 18:04:55 +02:00
Mark Qvist 3e5f613f66 Fixed typo 2021-10-12 16:36:29 +02:00
Mark Qvist 614a139cd4 Merge branch 'master' of github.com:markqvist/Reticulum 2021-10-12 16:34:25 +02:00
Mark Qvist 1cf6570c2d Added RSSI and SNR reporting to packets on supported interfaces 2021-10-12 16:34:17 +02:00
Mark Qvist d207cbcd9c Update README.md 2021-10-11 15:26:21 +02:00
Mark Qvist 18b20f2d8d Update README.md 2021-10-11 15:26:07 +02:00
Mark Qvist c37533d2c7 Updated docs 2021-10-10 00:27:04 +02:00
Mark Qvist fd13e20165 Updated version 2021-10-09 23:23:44 +02:00
Mark Qvist 66ce58f0f4 Implemented path updating for moving nodes 2021-10-09 22:13:27 +02:00
Mark Qvist e8ee26f78d Emission timestamp in announce. 2021-10-09 21:36:01 +02:00
Mark Qvist c0fb419fe1 Fixed Resource string representation. Added emission timestamp in announce. 2021-10-09 21:30:34 +02:00
Mark Qvist 4ef369cdd8 Added logfile rotation 2021-10-08 19:23:10 +02:00
Mark Qvist a2f18b1daf Updated docs 2021-10-08 18:50:38 +02:00
Mark Qvist 2e411fa1de Updated docs 2021-10-08 18:49:13 +02:00
Mark Qvist 549dc40be6 Updated docs 2021-10-08 18:48:06 +02:00
Mark Qvist 1a99597f4d Updated documentation 2021-10-08 18:31:43 +02:00
Mark Qvist b21e0bee20 Updated documentation 2021-10-08 18:30:17 +02:00
Mark Qvist be8389a906 Updated readme 2021-10-08 17:54:17 +02:00
Mark Qvist 4ca00c6973 Added path expiry check to tunnel restoration 2021-10-08 17:09:31 +02:00
Mark Qvist 95f81cab7f Added path expiry check to tunnel restoration 2021-10-08 17:09:11 +02:00
Mark Qvist 60917f0eea Fixed interface detachment on TCP initiator interfaces 2021-10-08 17:06:00 +02:00
Mark Qvist de800f0ea7 Updated log output 2021-10-08 11:57:23 +02:00
Mark Qvist 5dad76879c Improved known destination saving on shared instances 2021-10-08 08:52:50 +02:00
Mark Qvist 75c3180933 Improved shared instance and local client handling 2021-10-03 15:23:12 +02:00
Mark Qvist 4c6ba97dca Updated readme 2021-09-26 12:34:19 +02:00
Mark Qvist cd6427cc9d Fixed rnstatus output 2021-09-25 23:56:56 +02:00
Mark Qvist 1749393732 Fixed rnstatus output 2021-09-25 23:44:59 +02:00
75 changed files with 8275 additions and 1996 deletions
+1
View File
@@ -3,6 +3,7 @@
testutils
TODO
Examples/RNS
RNS/Utilities/RNS
build
dist
docs/build
+1 -1
View File
@@ -15,7 +15,7 @@ import RNS
APP_NAME = "example_utilities"
# We initialise two lists of strings to use as app_data
fruits = ["Peach", "Quince", "Date palm", "Tangerine", "Pomelo", "Carambola", "Grape"]
fruits = ["Peach", "Quince", "Date", "Tangerine", "Pomelo", "Carambola", "Grape"]
noble_gases = ["Helium", "Neon", "Argon", "Krypton", "Xenon", "Radon", "Oganesson"]
# This initialisation is executed when the program is started
+49 -2
View File
@@ -22,6 +22,8 @@ APP_NAME = "example_utilities"
# This initialisation is executed when the users chooses
# to run as a server
def server(configpath):
global reticulum
# We must first initialise Reticulum
reticulum = RNS.Reticulum(configpath)
@@ -78,11 +80,32 @@ def announceLoop(destination):
def server_callback(message, packet):
global reticulum
# Tell the user that we received an echo request, and
# that we are going to send a reply to the requester.
# Sending the proof is handled automatically, since we
# set up the destination to prove all incoming packets.
RNS.log("Received packet from echo client, proof sent")
reception_stats = ""
if reticulum.is_connected_to_shared_instance:
reception_rssi = reticulum.get_packet_rssi(packet.packet_hash)
reception_snr = reticulum.get_packet_snr(packet.packet_hash)
if reception_rssi != None:
reception_stats += " [RSSI "+str(reception_rssi)+" dBm]"
if reception_snr != None:
reception_stats += " [SNR "+str(reception_snr)+" dBm]"
else:
if packet.rssi != None:
reception_stats += " [RSSI "+str(packet.rssi)+" dBm]"
if packet.snr != None:
reception_stats += " [SNR "+str(packet.snr)+" dB]"
RNS.log("Received packet from echo client, proof sent"+reception_stats)
##########################################################
@@ -92,6 +115,8 @@ def server_callback(message, packet):
# This initialisation is executed when the users chooses
# to run as a client
def client(destination_hexhash, configpath, timeout=None):
global reticulum
# We need a binary representation of the destination
# hash that was entered on the command line
try:
@@ -188,6 +213,8 @@ def client(destination_hexhash, configpath, timeout=None):
# This function is called when our reply destination
# receives a proof packet.
def packet_delivered(receipt):
global reticulum
if receipt.status == RNS.PacketReceipt.DELIVERED:
rtt = receipt.get_rtt()
if (rtt >= 1):
@@ -197,10 +224,30 @@ def packet_delivered(receipt):
rtt = round(rtt*1000, 3)
rttstring = str(rtt)+" milliseconds"
reception_stats = ""
if reticulum.is_connected_to_shared_instance:
reception_rssi = reticulum.get_packet_rssi(receipt.proof_packet.packet_hash)
reception_snr = reticulum.get_packet_snr(receipt.proof_packet.packet_hash)
if reception_rssi != None:
reception_stats += " [RSSI "+str(reception_rssi)+" dBm]"
if reception_snr != None:
reception_stats += " [SNR "+str(reception_snr)+" dB]"
else:
if receipt.proof_packet != None:
if receipt.proof_packet.rssi != None:
reception_stats += " [RSSI "+str(receipt.proof_packet.rssi)+" dBm]"
if receipt.proof_packet.snr != None:
reception_stats += " [SNR "+str(receipt.proof_packet.snr)+" dB]"
RNS.log(
"Valid reply received from "+
RNS.prettyhexrep(receipt.destination.hash)+
", round-trip time is "+rttstring
", round-trip time is "+rttstring+
reception_stats
)
# This function is called if a packet times out.
+6
View File
@@ -1,6 +1,12 @@
##########################################################
# This RNS example demonstrates a simple speedtest #
# program to measure link throughput. #
# #
# The current configuration is suited for testing fast #
# links. If you want to measure slow links like LoRa or #
# packet radio, you must significantly lower the #
# data_cap variable, which defines how much data is sent #
# for each test. #
##########################################################
import os
+1 -1
View File
@@ -1,6 +1,6 @@
MIT License, unless otherwise noted
Copyright (c) 2018 Mark Qvist / unsigned.io
Copyright (c) 2016-2022 Mark Qvist / unsigned.io
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
+25
View File
@@ -0,0 +1,25 @@
all: release
clean:
@echo Cleaning...
-rm -r ./build
-rm -r ./dist
remove_symlinks:
@echo Removing symlinks for build...
-rm Examples/RNS
-rm RNS/Utilities/RNS
create_symlinks:
@echo Creating symlinks...
-ln -s ../RNS ./Examples/
-ln -s ../../RNS ./RNS/Utilities/
build_wheel:
python3 setup.py sdist bdist_wheel
release: remove_symlinks build_wheel create_symlinks
upload:
@echo Uploading to PyPi...
twine upload dist/*
+125 -59
View File
@@ -1,7 +1,11 @@
Reticulum Network Stack β
==========
Reticulum is a cryptography-based networking stack for wide-area networks built on readily available hardware, and can operate even with very high latency and extremely low bandwidth. Reticulum allows you to build very wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.
<p align="center"><img width="200" src="https://unsigned.io/wp-content/uploads/2022/03/reticulum_logo_512.png"></p>
Reticulum is the cryptography-based networking stack for wide-area networks built on readily available hardware. It can operate even with very high latency and extremely low bandwidth. Reticulum allows you to build wide-area networks with off-the-shelf tools, and offers end-to-end encryption and connectivity, initiator anonymity, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable delivery acknowledgements and more.
The vision of Reticulum is to allow anyone to be their own network operator, and to make it cheap and easy to cover vast areas with a myriad of independent, interconnectable and autonomous networks. Reticulum **is not** *one network*, it **is a tool** for building *thousands of networks*. Networks without kill-switches, surveillance, censorship and control. Networks that can freely interoperate, associate and disassociate with each other, and require no central oversight. Networks for human beings. *Networks for the people*.
Reticulum is a complete networking stack, and does not need IP or higher layers, although it is easy to use IP (with TCP or UDP) as the underlying carrier for Reticulum. It is therefore trivial to tunnel Reticulum over the Internet or private IP networks.
@@ -9,6 +13,7 @@ Having no dependencies on traditional networking stacks free up overhead that ha
No kernel modules or drivers are required. Reticulum runs completely in userland, and can run on practically any system that runs Python 3.
## Read The Manual
The full documentation for Reticulum is available at [markqvist.github.io/Reticulum/manual/](https://markqvist.github.io/Reticulum/manual/).
You can also [download the Reticulum manual as a PDF](https://github.com/markqvist/Reticulum/raw/master/docs/Reticulum%20Manual.pdf)
@@ -16,29 +21,37 @@ You can also [download the Reticulum manual as a PDF](https://github.com/markqvi
For more info, see [unsigned.io/projects/reticulum](https://unsigned.io/projects/reticulum/)
## Notable Features
- Coordination-less globally unique adressing and identification
- Fully self-configuring multi-hop routing
- Asymmetric X25519 encryption and Ed25519 signatures as a basis for all communication
- Forward Secrecy with ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
- Reticulum uses the [Fernet](https://github.com/fernet/spec/blob/master/Spec.md) specification for on-the-wire / over-the-air encryption
- Keys are ephemeral and derived from an ECDH key exchange on Curve25519
- AES-128 in CBC mode with PKCS7 padding
- HMAC using SHA256 for authentication
- IVs are generated through os.urandom()
- Unforgeable packet delivery confirmations
- A variety of supported interface types
- An intuitive and easy-to-use API
- Reliable and efficient transfer of arbritrary amounts of data
- Reticulum can handle a few bytes of data or files of many gigabytes
- Sequencing, transfer coordination and checksumming is automatic
- The API is very easy to use, and provides transfer progress
- Lightweight, flexible and expandable Request/Response mechanism
- Efficient link establishment
- Total bandwidth cost of setting up a link is 3 packets totalling 237 bytes
- Low cost of keeping links open at only 0.62 bits per second
- Coordination-less globally unique adressing and identification
- Fully self-configuring multi-hop routing
- Complete initiator anonymity, communicate without revealing your identity
- Asymmetric X25519 encryption and Ed25519 signatures as a basis for all communication
- Forward Secrecy with ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
- Reticulum uses the [Fernet](https://github.com/fernet/spec/blob/master/Spec.md) specification for on-the-wire / over-the-air encryption
- Keys are ephemeral and derived from an ECDH key exchange on Curve25519
- AES-128 in CBC mode with PKCS7 padding
- HMAC using SHA256 for authentication
- IVs are generated through os.urandom()
- Unforgeable packet delivery confirmations
- A variety of supported interface types
- An intuitive and easy-to-use API
- Reliable and efficient transfer of arbritrary amounts of data
- Reticulum can handle a few bytes of data or files of many gigabytes
- Sequencing, transfer coordination and checksumming is automatic
- The API is very easy to use, and provides transfer progress
- Lightweight, flexible and expandable Request/Response mechanism
- Efficient link establishment
- Total bandwidth cost of setting up an encrypted link is 3 packets totalling 237 bytes
- Low cost of keeping links open at only 0.62 bits per second
## Examples of Reticulum Applications
If you want to quickly get an idea of what Reticulum can do, take a look at the following resources.
- [LXMF](https://github.com/markqvist/lxmf) is a distributed, delay and disruption tolerant message transfer protocol built on Reticulum
- For an off-grid, encrypted and resilient mesh communications platform, see [Nomad Network](https://github.com/markqvist/NomadNet)
- The Android, Linux and macOS app [Sideband](https://unsigned.io/sideband) has a graphical interface and focuses on ease of use.
## Where can Reticulum be used?
Over practically any medium that can support at least a half-duplex channel with 1.000 bits per second throughput, and an MTU of 500 bytes. Data radios, modems, LoRa radios, serial lines, AX.25 TNCs, amateur radio digital modes, ad-hoc WiFi, free-space optical links and similar systems are all examples of the types of interfaces Reticulum was designed for.
Over practically any medium that can support at least a half-duplex channel with 500 bits per second throughput, and an MTU of 500 bytes. Data radios, modems, LoRa radios, serial lines, AX.25 TNCs, amateur radio digital modes, ad-hoc WiFi, free-space optical links and similar systems are all examples of the types of interfaces Reticulum was designed for.
An open-source LoRa-based interface called [RNode](https://unsigned.io/projects/rnode/) has been designed specifically for use with Reticulum. It is possible to build yourself, or it can be purchased as a complete transceiver that just needs a USB connection to the host.
@@ -46,52 +59,105 @@ Reticulum can also be encapsulated over existing IP networks, so there's nothing
As an example, it's possible to set up a Raspberry Pi connected to both a LoRa radio, a packet radio TNC and a WiFi network. Once the interfaces are configured, Reticulum will take care of the rest, and any device on the WiFi network can communicate with nodes on the LoRa and packet radio sides of the network, and vice versa.
## Current Status
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered relatively stable at the moment, but could change if warranted.
## Supported interface types and devices
Reticulum implements a range of generalised interface types that covers most of the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. Currently, the following interfaces are supported:
- Any ethernet device
- LoRa using [RNode](https://unsigned.io/projects/rnode/)
- Packet Radio TNCs (with or without AX.25)
- Any device with a serial port
- TCP over IP networks
- UDP over IP networks
## Feature Roadmap
- More interface types for even broader compatibility
- ESP32 devices (ESP-Now, Bluetooth, etc.)
- AT-compatible modems
- CAN-bus
- ZeroMQ
- MQTT
- SPI
- i²c
- A delay and disruption tolerant message transfer protocol built on Reticulum, see [LXMF](https://github.com/markqvist/lxmf)
- A few useful-in-the-real-world apps built with Reticulum, see [Nomad Network](https://github.com/markqvist/NomadNet)
## Dependencies:
- Python 3
- cryptography.io
- pyserial
## How do I get started?
The best way to get started with the Reticulum Network Stack depends on what
you want to do. For full details and examples, have a look at the [Getting Started Fast](https://markqvist.github.io/Reticulum/manual/gettingstartedfast.html) section of the [Reticulum Manual](https://markqvist.github.io/Reticulum/manual/).
If you just need Reticulum as a dependency for another application, the easiest way is via pip:
To simply install Reticulum and related utilities on your system, the easiest way is via pip:
```bash
pip3 install rns
```
The default config file contains examples for using Reticulum with LoRa transceivers (specifically [RNode](https://unsigned.io/projects/rnode/)), packet radio TNCs/modems and UDP. By default a UDP interface is already enabled in the default config, which will enable Reticulum communication in your local ethernet broadcast domain.
You can then start any program that uses Reticulum, or start Reticulum as a system service with [the rnsd utility](https://markqvist.github.io/Reticulum/manual/using.html#the-rnsd-utility).
You can use the examples in the config file to expand communication over other mediums such as packet radio or LoRa, or over fast IP links using the UDP interface. I'll add in-depth tutorials and explanations on these topics later. For now, the included examples will hopefully be enough to get started.
When first started, Reticulum will create a default configuration file, providing basic connectivity to other Reticulum peers. The default config file contains examples for using Reticulum with LoRa transceivers (specifically [RNode](https://unsigned.io/projects/rnode/)), packet radio TNCs/modems, TCP and UDP.
You can use the examples in the config file to expand communication over many mediums such as packet radio or LoRa (with [RNode](https://unsigned.io/projects/rnode/)), serial ports, or over fast IP links and the Internet using the UDP and TCP interfaces. For more detailed examples, take a look at the [Supported Interfaces](https://markqvist.github.io/Reticulum/manual/interfaces.html) section of the [Reticulum Manual](https://markqvist.github.io/Reticulum/manual/).
## Current Status
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered relatively stable at the moment, but could change if warranted.
## Supported interface types and devices
Reticulum implements a range of generalised interface types that covers most of the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. I will gratefully accept pull requests for custom interfaces if they are generally useful.
Currently, the following interfaces are supported:
- Any ethernet device
- LoRa using [RNode](https://unsigned.io/projects/rnode/)
- Packet Radio TNCs (with or without AX.25)
- KISS-compatible hardware and software modems
- Any device with a serial port
- TCP over IP networks
- UDP over IP networks
## Development Roadmap
- Version 0.3.6
- Improving [the manual](https://markqvist.github.io/Reticulum/manual/) with sections specifically for beginners
- Support for radio and modem interfaces on Android
- GUI interface configuration tool
- Easy way to share interface configurations, see [#19](https://github.com/markqvist/Reticulum/discussions/19)
- Version 0.3.7
- More interface types for even broader compatibility
- Plain ESP32 devices (ESP-Now, WiFi, Bluetooth, etc.)
- More LoRa transceivers
- AT-compatible modems
- IR Transceivers
- AWDL / OWL
- HF Modems
- CAN-bus
- ZeroMQ
- MQTT
- SPI
- i²c
- Planned, but not yet scheduled
- Globally routable multicast
- A portable Reticulum implementation in C, see [#21](https://github.com/markqvist/Reticulum/discussions/21)
## Dependencies:
- Python 3.6
- cryptography.io
- netifaces
- pyserial
## Public Testnet
If you just want to get started experimenting without building any physical networks, you are welcome to join the Unsigned.io RNS Testnet. The testnet is just that, an informal network for testing and experimenting. It will be up most of the time, and anyone can join, but it also means that there's no guarantees for service availability.
The testnet runs the very latest version of Reticulum (often even a short while before it is publicly released). Sometimes experimental versions of Reticulum might be deployed to nodes on the testnet, which means strange behaviour might occur. If none of that scares you, you can join the testnet via eihter TCP or I2P. Just add one of the following interfaces to your Reticulum configuration file:
```
# For connecting over TCP/IP:
[[RNS Testnet Frankfurt]]
type = TCPClientInterface
interface_enabled = yes
outgoing = True
target_host = frankfurt.rns.unsigned.io
target_port = 4965
# For connecting over I2P:
[[RNS Testnet I2P Node A]]
type = I2PInterface
interface_enabled = yes
peers = ykzlw5ujbaqc2xkec4cpvgyxj257wcrmmgkuxqmqcur7cq3w3lha.b32.i2p
```
The testnet also contains a number of [Nomad Network](https://github.com/markqvist/nomadnet) nodes, and LXMF propagation nodes.
## Support Reticulum
You can help support the continued development of open, free and private communications systems by donating via one of the following channels:
- Ethereum: 0x81F7B979fEa6134bA9FD5c701b3501A2e61E897a
- Bitcoin: 3CPmacGm34qYvR6XWLVEJmi2aNe3PZqUuq
- Ko-Fi: https://ko-fi.com/markqvist
Are certain features in the development roadmap are important to you or your organisation? Make them a reality quickly by sponsoring their implementation.
## Caveat Emptor
Reticulum is experimental software, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it _has not_ been externally security audited, and there could very well be privacy-breaking bugs. If you want to help out, or help sponsor an audit, please do get in touch.
Reticulum is relatively young software, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it _has not_ been externally security audited, and there could very well be privacy-breaking bugs. If you want to help out, or help sponsor an audit, please do get in touch.
+35 -2
View File
@@ -1,5 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import base64
import math
import time
import RNS
from cryptography.fernet import Fernet
@@ -117,6 +140,9 @@ class Destination:
identity = RNS.Identity()
aspects = aspects+(identity.hexhash,)
if identity != None and self.type == Destination.PLAIN:
raise TypeError("Selected destination type PLAIN cannot hold an identity")
self.identity = identity
self.name = Destination.full_name(app_name, *aspects)
@@ -145,8 +171,11 @@ class Destination:
:param app_data: *bytes* containing the app_data.
:param path_response: Internal flag used by :ref:`RNS.Transport<api-transport>`. Ignore.
"""
if self.type != Destination.SINGLE:
raise TypeError("Only SINGLE destination types can be announced")
destination_hash = self.hash
random_hash = RNS.Identity.get_random_hash()
random_hash = RNS.Identity.get_random_hash()[0:5]+int(time.time()).to_bytes(5, "big")
if app_data == None and self.default_app_data != None:
if isinstance(self.default_app_data, bytes):
@@ -262,7 +291,11 @@ class Destination:
if plaintext != None:
if packet.packet_type == RNS.Packet.DATA:
if self.callbacks.packet != None:
self.callbacks.packet(plaintext, packet)
try:
self.callbacks.packet(plaintext, packet)
except Exception as e:
RNS.log("Error while executing receive callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def incoming_link_request(self, data, packet):
link = RNS.Link.validate_request(self, data, packet)
+95 -38
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import base64
import math
import os
@@ -14,6 +36,8 @@ from cryptography.hazmat.primitives.asymmetric.x25519 import X25519PrivateKey, X
from cryptography.hazmat.primitives.kdf.hkdf import HKDF
from cryptography.fernet import Fernet
cio_default_backend = default_backend()
class Identity:
"""
This class is used to manage identities in Reticulum. It provides methods
@@ -34,11 +58,12 @@ class Identity:
"""
# Non-configurable constants
FERNET_VERSION = 0x80
FERNET_OVERHEAD = 54 # In bytes
AES128_BLOCKSIZE = 16 # In bytes
HASHLENGTH = 256 # In bits
SIGLENGTH = KEYSIZE # In bits
FERNET_VERSION = 0x80
FERNET_OVERHEAD = 57 # In bytes
OPTIMISED_FERNET_OVERHEAD = 54 # In bytes
AES128_BLOCKSIZE = 16 # In bytes
HASHLENGTH = 256 # In bits
SIGLENGTH = KEYSIZE # In bits
TRUNCATED_HASHLENGTH = RNS.Reticulum.TRUNCATED_HASHLENGTH
"""
@@ -90,11 +115,27 @@ class Identity:
@staticmethod
def save_known_destinations():
RNS.log("Saving known destinations to storage...", RNS.LOG_VERBOSE)
file = open(RNS.Reticulum.storagepath+"/known_destinations","wb")
umsgpack.dump(Identity.known_destinations, file)
file.close()
RNS.log("Done saving known destinations to storage", RNS.LOG_VERBOSE)
try:
storage_known_destinations = {}
if os.path.isfile(RNS.Reticulum.storagepath+"/known_destinations"):
try:
file = open(RNS.Reticulum.storagepath+"/known_destinations","rb")
storage_known_destinations = umsgpack.load(file)
file.close()
except:
pass
for destination_hash in storage_known_destinations:
if not destination_hash in Identity.known_destinations:
Identity.known_destinations[destination_hash] = storage_known_destinations[destination_hash]
RNS.log("Saving known destinations to storage...", RNS.LOG_VERBOSE)
file = open(RNS.Reticulum.storagepath+"/known_destinations","wb")
umsgpack.dump(Identity.known_destinations, file)
file.close()
RNS.log("Done saving known destinations to storage", RNS.LOG_VERBOSE)
except Exception as e:
RNS.log("Error while saving known destinations to disk, the contained exception was: "+str(e), RNS.LOG_ERROR)
@staticmethod
def load_known_destinations():
@@ -107,7 +148,7 @@ class Identity:
except:
RNS.log("Error loading known destinations from disk, file will be recreated on exit", RNS.LOG_ERROR)
else:
RNS.log("Destinations file does not exist, so no known destinations loaded", RNS.LOG_VERBOSE)
RNS.log("Destinations file does not exist, no known destinations loaded", RNS.LOG_VERBOSE)
@staticmethod
def full_hash(data):
@@ -140,37 +181,47 @@ class Identity:
:param data: Data to be hashed as *bytes*.
:returns: Truncated SHA-256 hash of random data as *bytes*
"""
return Identity.truncated_hash(os.urandom(10))
return Identity.truncated_hash(os.urandom(Identity.TRUNCATED_HASHLENGTH//8))
@staticmethod
def validate_announce(packet):
if packet.packet_type == RNS.Packet.ANNOUNCE:
RNS.log("Validating announce from "+RNS.prettyhexrep(packet.destination_hash), RNS.LOG_DEBUG)
destination_hash = packet.destination_hash
public_key = packet.data[:Identity.KEYSIZE//8]
random_hash = packet.data[Identity.KEYSIZE//8:Identity.KEYSIZE//8+10]
signature = packet.data[Identity.KEYSIZE//8+10:Identity.KEYSIZE//8+10+Identity.KEYSIZE//8]
app_data = b""
if len(packet.data) > Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:
app_data = packet.data[Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:]
try:
if packet.packet_type == RNS.Packet.ANNOUNCE:
destination_hash = packet.destination_hash
public_key = packet.data[:Identity.KEYSIZE//8]
random_hash = packet.data[Identity.KEYSIZE//8:Identity.KEYSIZE//8+10]
signature = packet.data[Identity.KEYSIZE//8+10:Identity.KEYSIZE//8+10+Identity.KEYSIZE//8]
app_data = b""
if len(packet.data) > Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:
app_data = packet.data[Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:]
signed_data = destination_hash+public_key+random_hash+app_data
signed_data = destination_hash+public_key+random_hash+app_data
if not len(packet.data) > Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:
app_data = None
if not len(packet.data) > Identity.KEYSIZE//8+10+Identity.KEYSIZE//8:
app_data = None
announced_identity = Identity(create_keys=False)
announced_identity.load_public_key(public_key)
announced_identity = Identity(create_keys=False)
announced_identity.load_public_key(public_key)
if announced_identity.pub != None and announced_identity.validate(signature, signed_data):
RNS.Identity.remember(packet.get_hash(), destination_hash, public_key, app_data)
RNS.log("Stored valid announce from "+RNS.prettyhexrep(destination_hash), RNS.LOG_DEBUG)
del announced_identity
return True
else:
RNS.log("Received invalid announce", RNS.LOG_DEBUG)
del announced_identity
return False
if announced_identity.pub != None and announced_identity.validate(signature, signed_data):
RNS.Identity.remember(packet.get_hash(), destination_hash, public_key, app_data)
del announced_identity
if hasattr(packet, "transport_id") and packet.transport_id != None:
RNS.log("Valid announce for "+RNS.prettyhexrep(destination_hash)+" "+str(packet.hops)+" hops away, received via "+RNS.prettyhexrep(packet.transport_id)+" on "+str(packet.receiving_interface), RNS.LOG_EXTREME)
else:
RNS.log("Valid announce for "+RNS.prettyhexrep(destination_hash)+" "+str(packet.hops)+" hops away, received on "+str(packet.receiving_interface), RNS.LOG_EXTREME)
return True
else:
RNS.log("Received invalid announce for "+RNS.prettyhexrep(destination_hash), RNS.LOG_DEBUG)
del announced_identity
return False
except Exception as e:
RNS.log("Error occurred while validating announce. The contained exception was: "+str(e), RNS.LOG_ERROR)
return False
@staticmethod
def exit_handler():
@@ -376,11 +427,14 @@ class Identity:
)
shared_key = ephemeral_key.exchange(self.pub)
derived_key = derived_key = HKDF(
# TODO: Improve this re-allocation of HKDF
derived_key = HKDF(
algorithm=hashes.SHA256(),
length=32,
salt=self.get_salt(),
info=self.get_context(),
backend=cio_default_backend,
).derive(shared_key)
fernet = Fernet(base64.urlsafe_b64encode(derived_key))
@@ -408,11 +462,14 @@ class Identity:
peer_pub = X25519PublicKey.from_public_bytes(peer_pub_bytes)
shared_key = self.prv.exchange(peer_pub)
derived_key = derived_key = HKDF(
# TODO: Improve this re-allocation of HKDF
derived_key = HKDF(
algorithm=hashes.SHA256(),
length=32,
salt=self.get_salt(),
info=self.get_context(),
backend=cio_default_backend,
).derive(shared_key)
fernet = Fernet(base64.urlsafe_b64encode(derived_key))
@@ -443,7 +500,7 @@ class Identity:
return self.sig_prv.sign(message)
except Exception as e:
RNS.log("The identity "+str(self)+" could not sign the requested message. The contained exception was: "+str(e), RNS.LOG_ERROR)
raise e
raise e
else:
raise KeyError("Signing failed because identity does not hold a private key")
+87 -35
View File
@@ -1,8 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
from time import sleep
import sys
import serial
import threading
import time
import RNS
@@ -38,6 +58,7 @@ class AX25():
class AX25KISSInterface(Interface):
MAX_CHUNK = 32768
BITRATE_GUESS = 1200
owner = None
port = None
@@ -48,9 +69,18 @@ class AX25KISSInterface(Interface):
serial = None
def __init__(self, owner, name, callsign, ssid, port, speed, databits, parity, stopbits, preamble, txtail, persistence, slottime, flow_control):
import importlib
if importlib.util.find_spec('serial') != None:
import serial
else:
RNS.log("Using the AX.25 KISS interface requires a serial communication module to be installed.", RNS.LOG_CRITICAL)
RNS.log("You can install one with the command: python3 -m pip install pyserial", RNS.LOG_CRITICAL)
RNS.panic()
self.rxb = 0
self.txb = 0
self.pyserial = serial
self.serial = None
self.owner = owner
self.name = name
@@ -65,6 +95,7 @@ class AX25KISSInterface(Interface):
self.stopbits = stopbits
self.timeout = 100
self.online = False
self.bitrate = KISSInterface.BITRATE_GUESS
self.packet_queue = []
self.flow_control = flow_control
@@ -90,44 +121,48 @@ class AX25KISSInterface(Interface):
self.parity = serial.PARITY_ODD
try:
RNS.log("Opening serial port "+self.port+"...")
self.serial = serial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
self.open_port()
except Exception as e:
RNS.log("Could not open serial port for interface "+str(self), RNS.LOG_ERROR)
raise e
if self.serial.is_open:
# Allow time for interface to initialise before config
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring AX.25 KISS interface parameters...")
self.setPreamble(self.preamble)
self.setTxTail(self.txtail)
self.setPersistence(self.persistence)
self.setSlotTime(self.slottime)
self.setFlowControl(self.flow_control)
self.interface_ready = True
RNS.log("AX.25 KISS interface configured")
sleep(2)
self.configure_device()
else:
raise IOError("Could not open serial port")
def open_port(self):
RNS.log("Opening serial port "+self.port+"...", RNS.LOG_VERBOSE)
self.serial = self.pyserial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
def configure_device(self):
# Allow time for interface to initialise before config
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring AX.25 KISS interface parameters...")
self.setPreamble(self.preamble)
self.setTxTail(self.txtail)
self.setPersistence(self.persistence)
self.setSlotTime(self.slottime)
self.setFlowControl(self.flow_control)
self.interface_ready = True
RNS.log("AX.25 KISS interface configured")
def setPreamble(self, preamble):
preamble_ms = preamble
@@ -287,8 +322,6 @@ class AX25KISSInterface(Interface):
escape = False
data_buffer = data_buffer+bytes([byte])
elif (command == KISS.CMD_READY):
# TODO: add timeout and reset if ready
# command never arrives
self.process_queue()
else:
time_since_last = int(time.time()*1000) - last_read_ms
@@ -308,10 +341,29 @@ class AX25KISSInterface(Interface):
except Exception as e:
self.online = False
RNS.log("A serial port error occurred, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is being torn down. Restart Reticulum to attempt to open this interface again.", RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is now offline.", RNS.LOG_ERROR)
if RNS.Reticulum.panic_on_interface_error:
RNS.panic()
RNS.log("Reticulum will attempt to reconnect the interface periodically.", RNS.LOG_ERROR)
self.online = False
self.serial.close()
self.reconnect_port()
def reconnect_port(self):
while not self.online:
try:
time.sleep(5)
RNS.log("Attempting to reconnect serial port "+str(self.port)+" for "+str(self)+"...", RNS.LOG_VERBOSE)
self.open_port()
if self.serial.is_open:
self.configure_device()
except Exception as e:
RNS.log("Error while reconnecting port, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("Reconnected serial port for "+str(self))
def __str__(self):
return "AX25KISSInterface["+self.name+"]"
+352
View File
@@ -0,0 +1,352 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
import socketserver
import threading
import socket
import struct
import time
import sys
import RNS
class AutoInterface(Interface):
DEFAULT_DISCOVERY_PORT = 29716
DEFAULT_DATA_PORT = 42671
DEFAULT_GROUP_ID = "reticulum".encode("utf-8")
SCOPE_LINK = "2"
SCOPE_ADMIN = "4"
SCOPE_SITE = "5"
SCOPE_ORGANISATION = "8"
SCOPE_GLOBAL = "e"
PEERING_TIMEOUT = 7.5
DARWIN_IGNORE_IFS = ["awdl0", "llw0", "lo0", "en5"]
ANDROID_IGNORE_IFS = ["dummy0", "lo", "tun0"]
BITRATE_GUESS = 10*1000*1000
def __init__(self, owner, name, group_id=None, discovery_scope=None, discovery_port=None, data_port=None, allowed_interfaces=None, ignored_interfaces=None, configured_bitrate=None):
import importlib
if importlib.util.find_spec('netifaces') != None:
import netifaces
else:
RNS.log("Using AutoInterface requires the netifaces module.", RNS.LOG_CRITICAL)
RNS.log("You can install it with the command: python3 -m pip install netifaces", RNS.LOG_CRITICAL)
RNS.panic()
self.netifaces = netifaces
self.rxb = 0
self.txb = 0
self.IN = True
self.OUT = False
self.name = name
self.online = False
self.peers = {}
self.link_local_addresses = []
self.adopted_interfaces = {}
self.multicast_echoes = {}
self.timed_out_interfaces = {}
self.outbound_udp_socket = None
self.announce_interval = AutoInterface.PEERING_TIMEOUT/6.0
self.peer_job_interval = AutoInterface.PEERING_TIMEOUT*1.1
self.peering_timeout = AutoInterface.PEERING_TIMEOUT
self.multicast_echo_timeout = AutoInterface.PEERING_TIMEOUT/2
if allowed_interfaces == None:
self.allowed_interfaces = []
else:
self.allowed_interfaces = allowed_interfaces
if ignored_interfaces == None:
self.ignored_interfaces = []
else:
self.ignored_interfaces = ignored_interfaces
if group_id == None:
self.group_id = AutoInterface.DEFAULT_GROUP_ID
else:
self.group_id = group_id.encode("utf-8")
if discovery_port == None:
self.discovery_port = AutoInterface.DEFAULT_DISCOVERY_PORT
else:
self.discovery_port = discovery_port
if data_port == None:
self.data_port = AutoInterface.DEFAULT_DATA_PORT
else:
self.data_port = data_port
if discovery_scope == None:
self.discovery_scope = AutoInterface.SCOPE_LINK
elif str(discovery_scope).lower() == "link":
self.discovery_scope = AutoInterface.SCOPE_LINK
elif str(discovery_scope).lower() == "admin":
self.discovery_scope = AutoInterface.SCOPE_ADMIN
elif str(discovery_scope).lower() == "site":
self.discovery_scope = AutoInterface.SCOPE_SITE
elif str(discovery_scope).lower() == "organisation":
self.discovery_scope = AutoInterface.SCOPE_ORGANISATION
elif str(discovery_scope).lower() == "global":
self.discovery_scope = AutoInterface.SCOPE_GLOBAL
self.group_hash = RNS.Identity.full_hash(self.group_id)
g = self.group_hash
#gt = "{:02x}".format(g[1]+(g[0]<<8))
gt = "0"
gt += ":"+"{:02x}".format(g[3]+(g[2]<<8))
gt += ":"+"{:02x}".format(g[5]+(g[4]<<8))
gt += ":"+"{:02x}".format(g[7]+(g[6]<<8))
gt += ":"+"{:02x}".format(g[9]+(g[8]<<8))
gt += ":"+"{:02x}".format(g[11]+(g[10]<<8))
gt += ":"+"{:02x}".format(g[13]+(g[12]<<8))
self.mcast_discovery_address = "ff1"+self.discovery_scope+":"+gt
suitable_interfaces = 0
for ifname in self.netifaces.interfaces():
if RNS.vendor.platformutils.is_darwin() and ifname in AutoInterface.DARWIN_IGNORE_IFS and not ifname in self.allowed_interfaces:
RNS.log(str(self)+" skipping Darwin AWDL or tethering interface "+str(ifname), RNS.LOG_EXTREME)
elif RNS.vendor.platformutils.is_darwin() and ifname == "lo0":
RNS.log(str(self)+" skipping Darwin loopback interface "+str(ifname), RNS.LOG_EXTREME)
elif RNS.vendor.platformutils.is_android() and ifname in AutoInterface.ANDROID_IGNORE_IFS and not ifname in self.allowed_interfaces:
RNS.log(str(self)+" skipping Android system interface "+str(ifname), RNS.LOG_EXTREME)
elif ifname in self.ignored_interfaces:
RNS.log(str(self)+" ignoring disallowed interface "+str(ifname), RNS.LOG_EXTREME)
else:
if len(self.allowed_interfaces) > 0 and not ifname in self.allowed_interfaces:
RNS.log(str(self)+" ignoring interface "+str(ifname)+" since it was not allowed", RNS.LOG_EXTREME)
else:
addresses = self.netifaces.ifaddresses(ifname)
if self.netifaces.AF_INET6 in addresses:
link_local_addr = None
for address in addresses[self.netifaces.AF_INET6]:
if "addr" in address:
if address["addr"].startswith("fe80:"):
link_local_addr = address["addr"]
self.link_local_addresses.append(link_local_addr.split("%")[0])
self.adopted_interfaces[ifname] = link_local_addr.split("%")[0]
self.multicast_echoes[ifname] = time.time()
RNS.log(str(self)+" Selecting link-local address "+str(link_local_addr)+" for interface "+str(ifname), RNS.LOG_EXTREME)
if link_local_addr == None:
RNS.log(str(self)+" No link-local IPv6 address configured for "+str(ifname)+", skipping interface", RNS.LOG_EXTREME)
else:
mcast_addr = self.mcast_discovery_address
RNS.log(str(self)+" Creating multicast discovery listener on "+str(ifname)+" with address "+str(mcast_addr), RNS.LOG_EXTREME)
# Struct with interface index
if_struct = struct.pack("I", socket.if_nametoindex(ifname))
# Set up multicast socket
discovery_socket = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM)
discovery_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
discovery_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEPORT, 1)
discovery_socket.setsockopt(socket.IPPROTO_IPV6, socket.IPV6_MULTICAST_IF, if_struct)
# Join multicast group
mcast_group = socket.inet_pton(socket.AF_INET6, mcast_addr) + if_struct
discovery_socket.setsockopt(socket.IPPROTO_IPV6, socket.IPV6_JOIN_GROUP, mcast_group)
# Bind socket
addr_info = socket.getaddrinfo(mcast_addr+"%"+ifname, self.discovery_port, socket.AF_INET6, socket.SOCK_DGRAM)
discovery_socket.bind(addr_info[0][4])
# Set up thread for discovery packets
def discovery_loop():
self.discovery_handler(discovery_socket, ifname)
thread = threading.Thread(target=discovery_loop)
thread.setDaemon(True)
thread.start()
suitable_interfaces += 1
if suitable_interfaces == 0:
RNS.log(str(self)+" could not autoconfigure. This interface currently provides no connectivity.", RNS.LOG_WARNING)
else:
self.receives = True
peering_wait = self.announce_interval*1.2
RNS.log(str(self)+" discovering peers for "+str(round(peering_wait, 2))+" seconds...", RNS.LOG_VERBOSE)
def handlerFactory(callback):
def createHandler(*args, **keys):
return AutoInterfaceHandler(callback, *args, **keys)
return createHandler
self.owner = owner
socketserver.UDPServer.address_family = socket.AF_INET6
for ifname in self.adopted_interfaces:
local_addr = self.adopted_interfaces[ifname]+"%"+ifname
addr_info = socket.getaddrinfo(local_addr, self.data_port, socket.AF_INET6, socket.SOCK_DGRAM)
address = addr_info[0][4]
self.server = socketserver.UDPServer(address, handlerFactory(self.processIncoming))
thread = threading.Thread(target=self.server.serve_forever)
thread.setDaemon(True)
thread.start()
job_thread = threading.Thread(target=self.peer_jobs)
job_thread.setDaemon(True)
job_thread.start()
time.sleep(peering_wait)
if configured_bitrate != None:
self.bitrate = configured_bitrate
else:
self.bitrate = AutoInterface.BITRATE_GUESS
self.online = True
def discovery_handler(self, socket, ifname):
def announce_loop():
self.announce_handler(ifname)
thread = threading.Thread(target=announce_loop)
thread.setDaemon(True)
thread.start()
while True:
data, ipv6_src = socket.recvfrom(1024)
expected_hash = RNS.Identity.full_hash(self.group_id+ipv6_src[0].encode("utf-8"))
if data == expected_hash:
self.add_peer(ipv6_src[0], ifname)
else:
RNS.log(str(self)+" received peering packet on "+str(ifname)+" from "+str(ipv6_src[0])+", but authentication hash was incorrect.", RNS.LOG_DEBUG)
def peer_jobs(self):
while True:
time.sleep(self.peer_job_interval)
now = time.time()
timed_out_peers = []
# Check for timed out peers
for peer_addr in self.peers:
peer = self.peers[peer_addr]
last_heard = peer[1]
if now > last_heard+self.peering_timeout:
timed_out_peers.append(peer_addr)
# Remove any timed out peers
for peer_addr in timed_out_peers:
removed_peer = self.peers.pop(peer_addr)
RNS.log(str(self)+" removed peer "+str(peer_addr)+" on "+str(removed_peer[0]), RNS.LOG_DEBUG)
for ifname in self.adopted_interfaces:
last_multicast_echo = 0
if ifname in self.multicast_echoes:
last_multicast_echo = self.multicast_echoes[ifname]
if now - last_multicast_echo > self.multicast_echo_timeout:
if ifname in self.timed_out_interfaces and self.timed_out_interfaces[ifname] == False:
RNS.log("Multicast echo timeout for "+str(ifname)+". Carrier lost.", RNS.LOG_WARNING)
self.timed_out_interfaces[ifname] = True
else:
if ifname in self.timed_out_interfaces and self.timed_out_interfaces[ifname] == True:
RNS.log(str(self)+" Carrier recovered on "+str(ifname), RNS.LOG_WARNING)
self.timed_out_interfaces[ifname] = False
def announce_handler(self, ifname):
while True:
self.peer_announce(ifname)
time.sleep(self.announce_interval)
def peer_announce(self, ifname):
try:
link_local_address = self.adopted_interfaces[ifname]
discovery_token = RNS.Identity.full_hash(self.group_id+link_local_address.encode("utf-8"))
announce_socket = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM)
addr_info = socket.getaddrinfo(self.mcast_discovery_address, self.discovery_port, socket.AF_INET6, socket.SOCK_DGRAM)
ifis = struct.pack("I", socket.if_nametoindex(ifname))
announce_socket.setsockopt(socket.IPPROTO_IPV6, socket.IPV6_MULTICAST_IF, ifis)
announce_socket.sendto(discovery_token, addr_info[0][4])
except Exception as e:
if (ifname in self.timed_out_interfaces and self.timed_out_interfaces[ifname] == False) or not ifname in self.timed_out_interfaces:
RNS.log(str(self)+" Detected possible carrier loss on "+str(ifname)+": "+str(e), RNS.LOG_WARNING)
else:
pass
def add_peer(self, addr, ifname):
if addr in self.link_local_addresses:
ifname = None
for interface_name in self.adopted_interfaces:
if self.adopted_interfaces[interface_name] == addr:
ifname = interface_name
if ifname != None:
self.multicast_echoes[ifname] = time.time()
else:
RNS.log(str(self)+" received multicast echo on unexpected interface "+str(ifname), RNS.LOG_WARNING)
else:
if not addr in self.peers:
self.peers[addr] = [ifname, time.time()]
RNS.log(str(self)+" added peer "+str(addr)+" on "+str(ifname), RNS.LOG_DEBUG)
else:
self.refresh_peer(addr)
def refresh_peer(self, addr):
self.peers[addr][1] = time.time()
def processIncoming(self, data):
self.rxb += len(data)
self.owner.inbound(data, self)
def processOutgoing(self,data):
for peer in self.peers:
try:
if self.outbound_udp_socket == None:
self.outbound_udp_socket = socket.socket(socket.AF_INET6, socket.SOCK_DGRAM)
peer_addr = str(peer)+"%"+str(self.peers[peer][0])
addr_info = socket.getaddrinfo(peer_addr, self.data_port, socket.AF_INET6, socket.SOCK_DGRAM)
self.outbound_udp_socket.sendto(data, addr_info[0][4])
except Exception as e:
RNS.log("Could not transmit on "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
self.txb += len(data)
def __str__(self):
return "AutoInterface["+self.name+"]"
class AutoInterfaceHandler(socketserver.BaseRequestHandler):
def __init__(self, callback, *args, **keys):
self.callback = callback
socketserver.BaseRequestHandler.__init__(self, *args, **keys)
def handle(self):
data = self.request[0]
self.callback(data)
+622
View File
@@ -0,0 +1,622 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
import socketserver
import threading
import platform
import socket
import time
import sys
import os
import RNS
import asyncio
class HDLC():
FLAG = 0x7E
ESC = 0x7D
ESC_MASK = 0x20
@staticmethod
def escape(data):
data = data.replace(bytes([HDLC.ESC]), bytes([HDLC.ESC, HDLC.ESC^HDLC.ESC_MASK]))
data = data.replace(bytes([HDLC.FLAG]), bytes([HDLC.ESC, HDLC.FLAG^HDLC.ESC_MASK]))
return data
class KISS():
FEND = 0xC0
FESC = 0xDB
TFEND = 0xDC
TFESC = 0xDD
CMD_DATA = 0x00
CMD_UNKNOWN = 0xFE
@staticmethod
def escape(data):
data = data.replace(bytes([0xdb]), bytes([0xdb, 0xdd]))
data = data.replace(bytes([0xc0]), bytes([0xdb, 0xdc]))
return data
# TODO: Neater shutdown of the event loop and
# better error handling is needed. Sometimes
# errors occur in I2P that leave tunnel setup
# hanging indefinitely, and right now we have
# no way of catching it. Sometimes the server
# and client tasks are also not cancelled on
# shutdown, which leads to errors dumped to
# the console. This should also be remedied.
class I2PController:
def __init__(self, rns_storagepath):
import RNS.vendor.i2plib as i2plib
import RNS.vendor.i2plib.utils
self.client_tunnels = {}
self.server_tunnels = {}
self.loop = None
self.i2plib = i2plib
self.utils = i2plib.utils
self.sam_address = i2plib.get_sam_address()
self.storagepath = rns_storagepath+"/i2p"
if not os.path.isdir(self.storagepath):
os.makedirs(self.storagepath)
def start(self):
asyncio.set_event_loop(asyncio.new_event_loop())
self.loop = asyncio.get_event_loop()
try:
self.loop.run_forever()
except Exception as e:
RNS.log("Exception on event loop for "+str(self)+": "+str(e), RNS.LOG_ERROR)
finally:
self.loop.close()
def stop(self):
for task in asyncio.Task.all_tasks(loop=self.loop):
task.cancel()
self.loop.stop()
def get_free_port(self):
return self.i2plib.utils.get_free_port()
def client_tunnel(self, owner, i2p_destination):
self.client_tunnels[i2p_destination] = False
while True:
if not self.client_tunnels[i2p_destination]:
try:
async def tunnel_up():
RNS.log("Bringing up I2P tunnel to "+str(owner)+", this may take a while...", RNS.LOG_INFO)
tunnel = self.i2plib.ClientTunnel(i2p_destination, owner.local_addr, sam_address=self.sam_address, loop=self.loop)
await tunnel.run()
owner.awaiting_i2p_tunnel = False
RNS.log(str(owner)+ " tunnel setup complete", RNS.LOG_VERBOSE)
try:
self.loop.ext_owner = self
future = asyncio.run_coroutine_threadsafe(tunnel_up(), self.loop).result()
self.client_tunnels[i2p_destination] = True
except Exception as e:
RNS.log("Error while setting up I2P tunnel: "+str(e))
raise e
except Exception as e:
raise IOError("Could not connect to I2P SAM API while configuring to "+str(owner)+". Check that I2P is running and SAM is enabled.")
time.sleep(5)
def server_tunnel(self, owner):
i2p_dest_hash = RNS.Identity.full_hash(RNS.Identity.full_hash(owner.name.encode("utf-8")))
i2p_keyfile = self.storagepath+"/"+RNS.hexrep(i2p_dest_hash, delimit=False)+".i2p"
i2p_dest = None
if not os.path.isfile(i2p_keyfile):
coro = self.i2plib.new_destination(sam_address=self.sam_address, loop=self.loop)
i2p_dest = asyncio.run_coroutine_threadsafe(coro, self.loop).result()
key_file = open(i2p_keyfile, "w")
key_file.write(i2p_dest.private_key.base64)
key_file.close()
else:
key_file = open(i2p_keyfile, "r")
prvd = key_file.read()
key_file.close()
i2p_dest = self.i2plib.Destination(data=prvd, has_private_key=True)
i2p_b32 = i2p_dest.base32
owner.b32 = i2p_b32
self.server_tunnels[i2p_b32] = False
while self.server_tunnels[i2p_b32] == False:
try:
async def tunnel_up():
RNS.log(str(owner)+" Bringing up I2P endpoint, this may take a while...", RNS.LOG_INFO)
tunnel = self.i2plib.ServerTunnel((owner.bind_ip, owner.bind_port), loop=self.loop, destination=i2p_dest, sam_address=self.sam_address)
await tunnel.run()
RNS.log(str(owner)+ " endpoint setup complete. Now reachable at: "+str(i2p_dest.base32)+".b32.i2p", RNS.LOG_VERBOSE)
asyncio.run_coroutine_threadsafe(tunnel_up(), self.loop).result()
self.server_tunnels[i2p_b32] = True
except Exception as e:
raise IOError("Could not connect to I2P SAM API while configuring "+str(self)+". Check that I2P is running and SAM is enabled.")
time.sleep(5)
def get_loop(self):
return asyncio.get_event_loop()
class ThreadingI2PServer(socketserver.ThreadingMixIn, socketserver.TCPServer):
pass
class I2PInterfacePeer(Interface):
RECONNECT_WAIT = 15
RECONNECT_MAX_TRIES = None
# TCP socket options
I2P_USER_TIMEOUT = 40
I2P_PROBE_AFTER = 10
I2P_PROBE_INTERVAL = 5
I2P_PROBES = 6
def __init__(self, parent_interface, owner, name, target_i2p_dest=None, connected_socket=None, max_reconnect_tries=None):
self.rxb = 0
self.txb = 0
self.IN = True
self.OUT = False
self.socket = None
self.parent_interface = parent_interface
self.parent_count = True
self.name = name
self.initiator = False
self.reconnecting = False
self.never_connected = True
self.owner = owner
self.writing = False
self.online = False
self.detached = False
self.kiss_framing = False
self.i2p_tunneled = True
self.i2p_dest = None
self.i2p_tunnel_ready = False
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
self.bitrate = I2PInterface.BITRATE_GUESS
if max_reconnect_tries == None:
self.max_reconnect_tries = I2PInterfacePeer.RECONNECT_MAX_TRIES
else:
self.max_reconnect_tries = max_reconnect_tries
if connected_socket != None:
self.receives = True
self.target_ip = None
self.target_port = None
self.socket = connected_socket
if platform.system() == "Linux":
self.set_timeouts_linux()
elif platform.system() == "Darwin":
self.set_timeouts_osx()
elif target_i2p_dest != None:
self.receives = True
self.initiator = True
self.bind_ip = "127.0.0.1"
self.bind_port = self.parent_interface.i2p.get_free_port()
self.local_addr = (self.bind_ip, self.bind_port)
self.target_ip = self.bind_ip
self.target_port = self.bind_port
self.awaiting_i2p_tunnel = True
def tunnel_job():
self.parent_interface.i2p.client_tunnel(self, target_i2p_dest)
thread = threading.Thread(target=tunnel_job)
thread.setDaemon(True)
thread.start()
def wait_job():
while self.awaiting_i2p_tunnel:
time.sleep(0.25)
if not self.kiss_framing:
self.wants_tunnel = True
if not self.connect(initial=True):
thread = threading.Thread(target=self.reconnect)
thread.setDaemon(True)
thread.start()
else:
thread = threading.Thread(target=self.read_loop)
thread.setDaemon(True)
thread.start()
thread = threading.Thread(target=wait_job)
thread.setDaemon(True)
thread.start()
def set_timeouts_linux(self):
if not self.i2p_tunneled:
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, int(I2PInterfacePeer.TCP_USER_TIMEOUT * 1000))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, int(I2PInterfacePeer.TCP_PROBE_AFTER))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, int(I2PInterfacePeer.TCP_PROBE_INTERVAL))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPCNT, int(I2PInterfacePeer.TCP_PROBES))
else:
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, int(I2PInterfacePeer.I2P_USER_TIMEOUT * 1000))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, int(I2PInterfacePeer.I2P_PROBE_AFTER))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, int(I2PInterfacePeer.I2P_PROBE_INTERVAL))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPCNT, int(I2PInterfacePeer.I2P_PROBES))
def set_timeouts_osx(self):
if hasattr(socket, "TCP_KEEPALIVE"):
TCP_KEEPIDLE = socket.TCP_KEEPALIVE
else:
TCP_KEEPIDLE = 0x10
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
if not self.i2p_tunneled:
self.socket.setsockopt(socket.IPPROTO_TCP, TCP_KEEPIDLE, int(I2PInterfacePeer.TCP_PROBE_AFTER))
else:
self.socket.setsockopt(socket.IPPROTO_TCP, TCP_KEEPIDLE, int(I2PInterfacePeer.I2P_PROBE_AFTER))
def detach(self):
if self.socket != None:
if hasattr(self.socket, "close"):
if callable(self.socket.close):
RNS.log("Detaching "+str(self), RNS.LOG_DEBUG)
self.detached = True
try:
self.socket.shutdown(socket.SHUT_RDWR)
except Exception as e:
RNS.log("Error while shutting down socket for "+str(self)+": "+str(e))
try:
self.socket.close()
except Exception as e:
RNS.log("Error while closing socket for "+str(self)+": "+str(e))
self.socket = None
def connect(self, initial=False):
try:
self.socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.socket.connect((self.target_ip, self.target_port))
self.online = True
except Exception as e:
if initial:
if not self.awaiting_i2p_tunnel:
RNS.log("Initial connection for "+str(self)+" could not be established: "+str(e), RNS.LOG_ERROR)
RNS.log("Leaving unconnected and retrying connection in "+str(I2PInterfacePeer.RECONNECT_WAIT)+" seconds.", RNS.LOG_ERROR)
return False
else:
raise e
if platform.system() == "Linux":
self.set_timeouts_linux()
elif platform.system() == "Darwin":
self.set_timeouts_osx()
self.online = True
self.writing = False
self.never_connected = False
if not self.kiss_framing and self.wants_tunnel:
RNS.Transport.synthesize_tunnel(self)
return True
def reconnect(self):
if self.initiator:
if not self.reconnecting:
self.reconnecting = True
attempts = 0
while not self.online:
time.sleep(I2PInterfacePeer.RECONNECT_WAIT)
attempts += 1
if self.max_reconnect_tries != None and attempts > self.max_reconnect_tries:
RNS.log("Max reconnection attempts reached for "+str(self), RNS.LOG_ERROR)
self.teardown()
break
try:
self.connect()
except Exception as e:
if not self.awaiting_i2p_tunnel:
RNS.log("Connection attempt for "+str(self)+" failed: "+str(e), RNS.LOG_DEBUG)
else:
RNS.log(str(self)+" still waiting for I2P tunnel to appear", RNS.LOG_VERBOSE)
if not self.never_connected:
RNS.log(str(self)+" Re-established connection via I2P tunnel", RNS.LOG_INFO)
self.reconnecting = False
thread = threading.Thread(target=self.read_loop)
thread.setDaemon(True)
thread.start()
if not self.kiss_framing:
RNS.Transport.synthesize_tunnel(self)
else:
RNS.log("Attempt to reconnect on a non-initiator I2P interface. This should not happen.", RNS.LOG_ERROR)
raise IOError("Attempt to reconnect on a non-initiator I2P interface")
def processIncoming(self, data):
self.rxb += len(data)
if hasattr(self, "parent_interface") and self.parent_interface != None and self.parent_count:
self.parent_interface.rxb += len(data)
self.owner.inbound(data, self)
def processOutgoing(self, data):
if self.online:
while self.writing:
time.sleep(0.01)
try:
self.writing = True
if self.kiss_framing:
data = bytes([KISS.FEND])+bytes([KISS.CMD_DATA])+KISS.escape(data)+bytes([KISS.FEND])
else:
data = bytes([HDLC.FLAG])+HDLC.escape(data)+bytes([HDLC.FLAG])
self.socket.sendall(data)
self.writing = False
self.txb += len(data)
if hasattr(self, "parent_interface") and self.parent_interface != None and self.parent_count:
self.parent_interface.txb += len(data)
except Exception as e:
RNS.log("Exception occurred while transmitting via "+str(self)+", tearing down interface", RNS.LOG_ERROR)
RNS.log("The contained exception was: "+str(e), RNS.LOG_ERROR)
self.teardown()
def read_loop(self):
try:
in_frame = False
escape = False
data_buffer = b""
command = KISS.CMD_UNKNOWN
while True:
data_in = self.socket.recv(4096)
if len(data_in) > 0:
pointer = 0
while pointer < len(data_in):
byte = data_in[pointer]
pointer += 1
if self.kiss_framing:
# Read loop for KISS framing
if (in_frame and byte == KISS.FEND and command == KISS.CMD_DATA):
in_frame = False
self.processIncoming(data_buffer)
elif (byte == KISS.FEND):
in_frame = True
command = KISS.CMD_UNKNOWN
data_buffer = b""
elif (in_frame and len(data_buffer) < RNS.Reticulum.MTU):
if (len(data_buffer) == 0 and command == KISS.CMD_UNKNOWN):
# We only support one HDLC port for now, so
# strip off the port nibble
byte = byte & 0x0F
command = byte
elif (command == KISS.CMD_DATA):
if (byte == KISS.FESC):
escape = True
else:
if (escape):
if (byte == KISS.TFEND):
byte = KISS.FEND
if (byte == KISS.TFESC):
byte = KISS.FESC
escape = False
data_buffer = data_buffer+bytes([byte])
else:
# Read loop for HDLC framing
if (in_frame and byte == HDLC.FLAG):
in_frame = False
self.processIncoming(data_buffer)
elif (byte == HDLC.FLAG):
in_frame = True
data_buffer = b""
elif (in_frame and len(data_buffer) < RNS.Reticulum.MTU):
if (byte == HDLC.ESC):
escape = True
else:
if (escape):
if (byte == HDLC.FLAG ^ HDLC.ESC_MASK):
byte = HDLC.FLAG
if (byte == HDLC.ESC ^ HDLC.ESC_MASK):
byte = HDLC.ESC
escape = False
data_buffer = data_buffer+bytes([byte])
else:
self.online = False
if self.initiator and not self.detached:
RNS.log("Socket for "+str(self)+" was closed, attempting to reconnect...", RNS.LOG_WARNING)
self.reconnect()
else:
RNS.log("Socket for remote client "+str(self)+" was closed.", RNS.LOG_VERBOSE)
self.teardown()
break
except Exception as e:
self.online = False
RNS.log("An interface error occurred for "+str(self)+", the contained exception was: "+str(e), RNS.LOG_WARNING)
if self.initiator:
RNS.log("Attempting to reconnect...", RNS.LOG_WARNING)
self.reconnect()
else:
self.teardown()
def teardown(self):
if self.initiator and not self.detached:
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is being torn down. Restart Reticulum to attempt to open this interface again.", RNS.LOG_ERROR)
if RNS.Reticulum.panic_on_interface_error:
RNS.panic()
else:
RNS.log("The interface "+str(self)+" is being torn down.", RNS.LOG_VERBOSE)
self.online = False
self.OUT = False
self.IN = False
if hasattr(self, "parent_interface") and self.parent_interface != None:
self.parent_interface.clients -= 1
if self in RNS.Transport.interfaces:
if not self.initiator:
RNS.Transport.interfaces.remove(self)
def __str__(self):
return "I2PInterfacePeer["+str(self.name)+"]"
class I2PInterface(Interface):
BITRATE_GUESS = 256*1000
def __init__(self, owner, name, rns_storagepath, peers, connectable = True):
self.rxb = 0
self.txb = 0
self.online = False
self.clients = 0
self.owner = owner
self.connectable = connectable
self.i2p_tunneled = True
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
self.b32 = None
self.i2p = I2PController(rns_storagepath)
self.IN = True
self.OUT = False
self.name = name
self.receives = True
self.bind_ip = "127.0.0.1"
self.bind_port = self.i2p.get_free_port()
self.address = (self.bind_ip, self.bind_port)
self.bitrate = I2PInterface.BITRATE_GUESS
i2p_thread = threading.Thread(target=self.i2p.start)
i2p_thread.setDaemon(True)
i2p_thread.start()
def handlerFactory(callback):
def createHandler(*args, **keys):
return I2PInterfaceHandler(callback, *args, **keys)
return createHandler
ThreadingI2PServer.allow_reuse_address = True
self.server = ThreadingI2PServer(self.address, handlerFactory(self.incoming_connection))
thread = threading.Thread(target=self.server.serve_forever)
thread.setDaemon(True)
thread.start()
if self.connectable:
def tunnel_job():
self.i2p.server_tunnel(self)
thread = threading.Thread(target=tunnel_job)
thread.setDaemon(True)
thread.start()
if peers != None:
for peer_addr in peers:
interface_name = peer_addr
peer_interface = I2PInterfacePeer(self, self.owner, interface_name, peer_addr)
peer_interface.OUT = True
peer_interface.IN = True
peer_interface.parent_interface = self
peer_interface.parent_count = False
RNS.Transport.interfaces.append(peer_interface)
self.online = True
def incoming_connection(self, handler):
RNS.log("Accepting incoming I2P connection", RNS.LOG_VERBOSE)
interface_name = "Connected peer on "+self.name
spawned_interface = I2PInterfacePeer(self, self.owner, interface_name, connected_socket=handler.request)
spawned_interface.OUT = True
spawned_interface.IN = True
spawned_interface.parent_interface = self
spawned_interface.online = True
spawned_interface.bitrate = self.bitrate
spawned_interface.ifac_size = self.ifac_size
spawned_interface.ifac_netname = self.ifac_netname
spawned_interface.ifac_netkey = self.ifac_netkey
RNS.log("Spawned new I2PInterface Peer: "+str(spawned_interface), RNS.LOG_VERBOSE)
RNS.Transport.interfaces.append(spawned_interface)
self.clients += 1
spawned_interface.read_loop()
def processOutgoing(self, data):
pass
def detach(self):
self.i2p.stop()
def __str__(self):
return "I2PInterface["+self.name+"]"
class I2PInterfaceHandler(socketserver.BaseRequestHandler):
def __init__(self, callback, *args, **keys):
self.callback = callback
socketserver.BaseRequestHandler.__init__(self, *args, **keys)
def handle(self):
self.callback(handler=self)
+65
View File
@@ -1,4 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import time
import threading
class Interface:
IN = False
@@ -7,6 +31,10 @@ class Interface:
RPT = False
name = None
MODE_FULL = 0x01
MODE_POINT_TO_POINT = 0x02
MODE_ACCESS_POINT = 0x03
def __init__(self):
self.rxb = 0
self.txb = 0
@@ -15,5 +43,42 @@ class Interface:
def get_hash(self):
return RNS.Identity.full_hash(str(self).encode("utf-8"))
def process_announce_queue(self):
if not hasattr(self, "announce_cap"):
self.announce_cap = RNS.Reticulum.ANNOUNCE_CAP
if hasattr(self, "announce_queue"):
try:
now = time.time()
stale = []
for a in self.announce_queue:
if now > a["time"]+RNS.Reticulum.QUEUED_ANNOUNCE_LIFE:
stale.append(a)
for s in stale:
self.announce_queue.remove(s)
if len(self.announce_queue) > 0:
min_hops = min(entry["hops"] for entry in self.announce_queue)
entries = list(filter(lambda e: e["hops"] == min_hops, self.announce_queue))
entries.sort(key=lambda e: e["time"])
selected = entries[0]
now = time.time()
tx_time = (len(selected["raw"])*8) / self.bitrate
wait_time = (tx_time / self.announce_cap)
self.announce_allowed_at = now + wait_time
self.processOutgoing(selected["raw"])
self.announce_queue.remove(selected)
if len(self.announce_queue) > 0:
timer = threading.Timer(wait_time, self.process_announce_queue)
timer.start()
except Exception as e:
self.announce_queue = []
RNS.log("Error while processing announce queue on "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("The announce queue for this interface has been cleared.", RNS.LOG_ERROR)
def detach(self):
pass
+91 -32
View File
@@ -1,7 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
from time import sleep
import sys
import serial
import threading
import time
import RNS
@@ -30,6 +51,7 @@ class KISS():
class KISSInterface(Interface):
MAX_CHUNK = 32768
BITRATE_GUESS = 1200
owner = None
port = None
@@ -40,12 +62,21 @@ class KISSInterface(Interface):
serial = None
def __init__(self, owner, name, port, speed, databits, parity, stopbits, preamble, txtail, persistence, slottime, flow_control, beacon_interval, beacon_data):
import importlib
if importlib.util.find_spec('serial') != None:
import serial
else:
RNS.log("Using the KISS interface requires a serial communication module to be installed.", RNS.LOG_CRITICAL)
RNS.log("You can install one with the command: python3 -m pip install pyserial", RNS.LOG_CRITICAL)
RNS.panic()
self.rxb = 0
self.txb = 0
if beacon_data == None:
beacon_data = ""
self.pyserial = serial
self.serial = None
self.owner = owner
self.name = name
@@ -59,6 +90,7 @@ class KISSInterface(Interface):
self.beacon_i = beacon_interval
self.beacon_d = beacon_data.encode("utf-8")
self.first_tx = None
self.bitrate = KISSInterface.BITRATE_GUESS
self.packet_queue = []
self.flow_control = flow_control
@@ -78,44 +110,52 @@ class KISSInterface(Interface):
self.parity = serial.PARITY_ODD
try:
RNS.log("Opening serial port "+self.port+"...")
self.serial = serial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
self.open_port()
except Exception as e:
RNS.log("Could not open serial port "+self.port, RNS.LOG_ERROR)
raise e
if self.serial.is_open:
# Allow time for interface to initialise before config
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring KISS interface parameters...")
self.setPreamble(self.preamble)
self.setTxTail(self.txtail)
self.setPersistence(self.persistence)
self.setSlotTime(self.slottime)
self.setFlowControl(self.flow_control)
self.interface_ready = True
RNS.log("KISS interface configured")
self.configure_device()
else:
raise IOError("Could not open serial port")
def open_port(self):
RNS.log("Opening serial port "+self.port+"...", RNS.LOG_VERBOSE)
self.serial = self.pyserial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
def configure_device(self):
# Allow time for interface to initialise before config
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring KISS interface parameters...")
self.setPreamble(self.preamble)
self.setTxTail(self.txtail)
self.setPersistence(self.persistence)
self.setSlotTime(self.slottime)
self.setFlowControl(self.flow_control)
self.interface_ready = True
RNS.log("KISS interface configured")
def setPreamble(self, preamble):
preamble_ms = preamble
preamble = int(preamble_ms / 10)
@@ -283,10 +323,29 @@ class KISSInterface(Interface):
except Exception as e:
self.online = False
RNS.log("A serial port error occurred, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is being torn down. Restart Reticulum to attempt to open this interface again.", RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is now offline.", RNS.LOG_ERROR)
if RNS.Reticulum.panic_on_interface_error:
RNS.panic()
RNS.log("Reticulum will attempt to reconnect the interface periodically.", RNS.LOG_ERROR)
self.online = False
self.serial.close()
self.reconnect_port()
def reconnect_port(self):
while not self.online:
try:
time.sleep(5)
RNS.log("Attempting to reconnect serial port "+str(self.port)+" for "+str(self)+"...", RNS.LOG_VERBOSE)
self.open_port()
if self.serial.is_open:
self.configure_device()
except Exception as e:
RNS.log("Error while reconnecting port, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("Reconnected serial port for "+str(self))
def __str__(self):
return "KISSInterface["+self.name+"]"
+87 -15
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
import socketserver
import threading
@@ -22,6 +44,7 @@ class ThreadingTCPServer(socketserver.ThreadingMixIn, socketserver.TCPServer):
pass
class LocalClientInterface(Interface):
RECONNECT_WAIT = 3
def __init__(self, owner, name, target_port = None, connected_socket=None):
self.rxb = 0
@@ -32,7 +55,11 @@ class LocalClientInterface(Interface):
self.OUT = False
self.socket = None
self.parent_interface = None
self.reconnecting = False
self.never_connected = True
self.detached = False
self.name = name
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
if connected_socket != None:
self.receives = True
@@ -46,13 +73,10 @@ class LocalClientInterface(Interface):
self.receives = True
self.target_ip = "127.0.0.1"
self.target_port = target_port
self.socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.socket.connect((self.target_ip, self.target_port))
self.is_connected_to_shared_instance = True
self.connect()
self.owner = owner
self.bitrate = 1000*1000*1000
self.online = True
self.writing = False
@@ -61,6 +85,47 @@ class LocalClientInterface(Interface):
thread.setDaemon(True)
thread.start()
def connect(self):
self.socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
self.socket.connect((self.target_ip, self.target_port))
self.online = True
self.is_connected_to_shared_instance = True
self.never_connected = False
return True
def reconnect(self):
if self.is_connected_to_shared_instance:
if not self.reconnecting:
self.reconnecting = True
attempts = 0
while not self.online:
time.sleep(LocalClientInterface.RECONNECT_WAIT)
attempts += 1
try:
self.connect()
except Exception as e:
RNS.log("Connection attempt for "+str(self)+" failed: "+str(e), RNS.LOG_DEBUG)
if not self.never_connected:
RNS.log("Reconnected TCP socket for "+str(self)+".", RNS.LOG_INFO)
self.reconnecting = False
thread = threading.Thread(target=self.read_loop)
thread.setDaemon(True)
thread.start()
RNS.Transport.shared_connection_reappeared()
else:
RNS.log("Attempt to reconnect on a non-initiator shared local interface. This should not happen.", RNS.LOG_ERROR)
raise IOError("Attempt to reconnect on a non-initiator local interface")
def processIncoming(self, data):
self.rxb += len(data)
if hasattr(self, "parent_interface") and self.parent_interface != None:
@@ -68,6 +133,7 @@ class LocalClientInterface(Interface):
self.owner.inbound(data, self)
def processOutgoing(self, data):
if self.online:
while self.writing:
@@ -119,8 +185,14 @@ class LocalClientInterface(Interface):
escape = False
data_buffer = data_buffer+bytes([byte])
else:
RNS.log("Socket for "+str(self)+" was closed, tearing down interface", RNS.LOG_VERBOSE)
self.teardown(nowarning=True)
self.online = False
if self.is_connected_to_shared_instance and not self.detached:
RNS.log("Socket for "+str(self)+" was closed, attempting to reconnect...", RNS.LOG_WARNING)
RNS.Transport.shared_connection_disappeared()
self.reconnect()
else:
self.teardown(nowarning=True)
break
@@ -168,13 +240,10 @@ class LocalClientInterface(Interface):
RNS.panic()
if self.is_connected_to_shared_instance:
# TODO: Maybe add automatic recovery here.
# Needs thinking through, since user needs
# to now that all connectivity has been cut
# while service is recovering. Better for
# now to take down entire stack.
RNS.log("Lost connection to local shared RNS instance. Exiting now.", RNS.LOG_CRITICAL)
RNS.panic()
if nowarning == False:
RNS.log("Permanently lost connection to local shared RNS instance. Exiting now.", RNS.LOG_CRITICAL)
RNS.exit()
def __str__(self):
@@ -192,6 +261,7 @@ class LocalServerInterface(Interface):
self.IN = True
self.OUT = False
self.name = "Reticulum"
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
if (bindport != None):
self.receives = True
@@ -215,6 +285,7 @@ class LocalServerInterface(Interface):
thread.setDaemon(True)
thread.start()
self.bitrate = 1000*1000*1000
self.online = True
@@ -227,7 +298,8 @@ class LocalServerInterface(Interface):
spawned_interface.target_ip = handler.client_address[0]
spawned_interface.target_port = str(handler.client_address[1])
spawned_interface.parent_interface = self
RNS.log("Accepting new connection to shared instance: "+str(spawned_interface), RNS.LOG_VERBOSE)
spawned_interface.bitrate = self.bitrate
RNS.log("Accepting new connection to shared instance: "+str(spawned_interface), RNS.LOG_EXTREME)
RNS.Transport.interfaces.append(spawned_interface)
RNS.Transport.local_client_interfaces.append(spawned_interface)
self.clients += 1
+191 -37
View File
@@ -1,8 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
from time import sleep
import sys
import serial
import threading
import time
import math
@@ -31,8 +51,11 @@ class KISS():
CMD_STAT_SNR = 0x24
CMD_BLINK = 0x30
CMD_RANDOM = 0x40
CMD_PLATFORM = 0x48
CMD_MCU = 0x49
CMD_FW_VERSION = 0x50
CMD_ROM_READ = 0x51
CMD_RESET = 0x55
DETECT_REQ = 0x73
DETECT_RESP = 0x46
@@ -46,6 +69,9 @@ class KISS():
ERROR_TXFAILED = 0x02
ERROR_EEPROM_LOCKED = 0x03
PLATFORM_AVR = 0x90
PLATFORM_ESP32 = 0x80
@staticmethod
def escape(data):
data = data.replace(bytes([0xdb]), bytes([0xdb, 0xdd]))
@@ -71,10 +97,22 @@ class RNodeInterface(Interface):
CALLSIGN_MAX_LEN = 32
REQUIRED_FW_VER_MAJ = 1
REQUIRED_FW_VER_MIN = 26
def __init__(self, owner, name, port, frequency = None, bandwidth = None, txpower = None, sf = None, cr = None, flow_control = False, id_interval = None, id_callsign = None):
import importlib
if importlib.util.find_spec('serial') != None:
import serial
else:
RNS.log("Using the RNode interface requires a serial communication module to be installed.", RNS.LOG_CRITICAL)
RNS.log("You can install one with the command: python3 -m pip install pyserial", RNS.LOG_CRITICAL)
RNS.panic()
self.rxb = 0
self.txb = 0
self.pyserial = serial
self.serial = None
self.owner = owner
self.name = name
@@ -93,6 +131,12 @@ class RNodeInterface(Interface):
self.cr = cr
self.state = KISS.RADIO_STATE_OFF
self.bitrate = 0
self.platform = None
self.mcu = None
self.detected = False
self.firmware_ok = False
self.maj_version = 0
self.min_version = 0
self.last_id = 0
self.first_tx = None
@@ -150,46 +194,64 @@ class RNodeInterface(Interface):
raise ValueError("The configuration for "+str(self)+" contains errors, interface is offline")
try:
RNS.log("Opening serial port "+self.port+"...")
self.serial = serial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
self.open_port()
except Exception as e:
RNS.log("Could not open serial port for interface "+str(self), RNS.LOG_ERROR)
raise e
if self.serial.is_open:
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring RNode interface...", RNS.LOG_VERBOSE)
self.initRadio()
if (self.validateRadioState()):
self.interface_ready = True
RNS.log(str(self)+" is configured and powered up")
sleep(1.0)
else:
RNS.log("After configuring "+str(self)+", the reported radio parameters did not match your configuration.", RNS.LOG_ERROR)
RNS.log("Make sure that your hardware actually supports the parameters specified in the configuration", RNS.LOG_ERROR)
RNS.log("Aborting RNode startup", RNS.LOG_ERROR)
self.serial.close()
raise IOError("RNode interface did not pass validation")
self.configure_device()
else:
raise IOError("Could not open serial port")
def open_port(self):
RNS.log("Opening serial port "+self.port+"...")
self.serial = self.pyserial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
def configure_device(self):
sleep(2.0)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.detect()
sleep(0.1)
if not self.detected:
raise IOError("Could not detect device")
else:
if self.platform == KISS.PLATFORM_ESP32:
RNS.log("Resetting ESP32-based device before configuration...", RNS.LOG_VERBOSE)
self.hard_reset()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
RNS.log("Configuring RNode interface...", RNS.LOG_VERBOSE)
self.initRadio()
if (self.validateRadioState()):
self.interface_ready = True
RNS.log(str(self)+" is configured and powered up")
sleep(1.0)
else:
RNS.log("After configuring "+str(self)+", the reported radio parameters did not match your configuration.", RNS.LOG_ERROR)
RNS.log("Make sure that your hardware actually supports the parameters specified in the configuration", RNS.LOG_ERROR)
RNS.log("Aborting RNode startup", RNS.LOG_ERROR)
self.serial.close()
raise IOError("RNode interface did not pass configuration validation")
def initRadio(self):
self.setFrequency()
@@ -199,6 +261,19 @@ class RNodeInterface(Interface):
self.setCodingRate()
self.setRadioState(KISS.RADIO_STATE_ON)
def detect(self):
kiss_command = bytes([KISS.FEND, KISS.CMD_DETECT, KISS.DETECT_REQ, KISS.FEND, KISS.CMD_FW_VERSION, 0x00, KISS.FEND, KISS.CMD_PLATFORM, 0x00, KISS.FEND, KISS.CMD_MCU, 0x00, KISS.FEND])
written = self.serial.write(kiss_command)
if written != len(kiss_command):
raise IOError("An IO error occurred while detecting hardware for "+self(str))
def hard_reset(self):
kiss_command = bytes([KISS.FEND, KISS.CMD_RESET, 0xf8, KISS.FEND])
written = self.serial.write(kiss_command)
if written != len(kiss_command):
raise IOError("An IO error occurred while restarting device")
sleep(2.25);
def setFrequency(self):
c1 = self.frequency >> 24
c2 = self.frequency >> 16 & 0xFF
@@ -245,13 +320,28 @@ class RNodeInterface(Interface):
raise IOError("An IO error occurred while configuring coding rate for "+self(str))
def setRadioState(self, state):
self.state = state
kiss_command = bytes([KISS.FEND])+bytes([KISS.CMD_RADIO_STATE])+bytes([state])+bytes([KISS.FEND])
written = self.serial.write(kiss_command)
if written != len(kiss_command):
raise IOError("An IO error occurred while configuring radio state for "+self(str))
def validate_firmware(self):
if (self.maj_version >= RNodeInterface.REQUIRED_FW_VER_MAJ):
if (self.min_version >= RNodeInterface.REQUIRED_FW_VER_MIN):
self.firmware_ok = True
if self.firmware_ok:
return
RNS.log("The firmware version of the connected RNode is "+str(self.maj_version)+"."+str(self.min_version), RNS.LOG_ERROR)
RNS.log("This version of Reticulum requires at least version "+str(RNodeInterface.REQUIRED_FW_VER_MAJ)+"."+str(RNodeInterface.REQUIRED_FW_VER_MIN), RNS.LOG_ERROR)
RNS.log("Please update your RNode firmware with rnodeconf (https://github.com/markqvist/rnodeconfigutil/)")
RNS.panic()
def validateRadioState(self):
RNS.log("Validating radio configuration for "+str(self)+"...", RNS.LOG_VERBOSE)
RNS.log("Wating for radio configuration validation for "+str(self)+"...", RNS.LOG_VERBOSE)
sleep(0.25);
if (self.frequency != self.r_frequency):
RNS.log("Frequency mismatch", RNS.LOG_ERROR)
@@ -265,6 +355,9 @@ class RNodeInterface(Interface):
if (self.sf != self.r_sf):
RNS.log("Spreading factor mismatch", RNS.LOG_ERROR)
self.validcfg = False
if (self.state != self.r_state):
RNS.log("Radio state mismatch", RNS.LOG_ERROR)
self.validcfg = False
if (self.validcfg):
return True
@@ -283,6 +376,8 @@ class RNodeInterface(Interface):
def processIncoming(self, data):
self.rxb += len(data)
self.owner.inbound(data, self)
self.r_stat_rssi = None
self.r_stat_snr = None
def processOutgoing(self,data):
@@ -403,8 +498,30 @@ class RNodeInterface(Interface):
self.updateBitrate()
elif (command == KISS.CMD_RADIO_STATE):
self.r_state = byte
if self.r_state:
pass
#RNS.log(str(self)+" Radio reporting state is online", RNS.LOG_DEBUG)
else:
RNS.log(str(self)+" Radio reporting state is offline", RNS.LOG_DEBUG)
elif (command == KISS.CMD_RADIO_LOCK):
self.r_lock = byte
elif (command == KISS.CMD_FW_VERSION):
if (byte == KISS.FESC):
escape = True
else:
if (escape):
if (byte == KISS.TFEND):
byte = KISS.FEND
if (byte == KISS.TFESC):
byte = KISS.FESC
escape = False
command_buffer = command_buffer+bytes([byte])
if (len(command_buffer) == 2):
self.maj_version = int(command_buffer[0])
self.min_version = int(command_buffer[1])
self.validate_firmware()
elif (command == KISS.CMD_STAT_RX):
if (byte == KISS.FESC):
escape = True
@@ -439,15 +556,33 @@ class RNodeInterface(Interface):
self.r_stat_snr = int.from_bytes(bytes([byte]), byteorder="big", signed=True) * 0.25
elif (command == KISS.CMD_RANDOM):
self.r_random = byte
elif (command == KISS.CMD_PLATFORM):
self.platform = byte
elif (command == KISS.CMD_MCU):
self.mcu = byte
elif (command == KISS.CMD_ERROR):
if (byte == KISS.ERROR_INITRADIO):
RNS.log(str(self)+" hardware initialisation error (code "+RNS.hexrep(byte)+")", RNS.LOG_ERROR)
raise IOError("Radio initialisation failure")
elif (byte == KISS.ERROR_INITRADIO):
RNS.log(str(self)+" hardware TX error (code "+RNS.hexrep(byte)+")", RNS.LOG_ERROR)
raise IOError("Hardware transmit failure")
else:
RNS.log(str(self)+" hardware error (code "+RNS.hexrep(byte)+")", RNS.LOG_ERROR)
raise IOError("Unknown hardware failure")
elif (command == KISS.CMD_RESET):
if (byte == 0xF8):
if self.platform == KISS.PLATFORM_ESP32:
if self.online:
RNS.log("Detected reset while device was online, reinitialising device...", RNS.LOG_ERROR)
raise IOError("ESP32 reset")
elif (command == KISS.CMD_READY):
self.process_queue()
elif (command == KISS.CMD_DETECT):
if byte == KISS.DETECT_RESP:
self.detected = True
else:
self.detected = False
else:
time_since_last = int(time.time()*1000) - last_read_ms
@@ -469,11 +604,30 @@ class RNodeInterface(Interface):
except Exception as e:
self.online = False
RNS.log("A serial port error occurred, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is being torn down. Restart Reticulum to attempt to open this interface again.", RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is now offline.", RNS.LOG_ERROR)
if RNS.Reticulum.panic_on_interface_error:
RNS.panic()
def __str__(self):
return "RNodeInterface["+self.name+"]"
RNS.log("Reticulum will attempt to reconnect the interface periodically.", RNS.LOG_ERROR)
self.online = False
self.serial.close()
self.reconnect_port()
def reconnect_port(self):
while not self.online:
try:
time.sleep(3.5)
RNS.log("Attempting to reconnect serial port "+str(self.port)+" for "+str(self)+"...", RNS.LOG_VERBOSE)
self.open_port()
if self.serial.is_open:
self.configure_device()
except Exception as e:
RNS.log("Error while reconnecting port, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("Reconnected serial port for "+str(self))
def __str__(self):
return "RNodeInterface["+str(self.name)+"]"
+82 -23
View File
@@ -1,7 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
from time import sleep
import sys
import serial
import threading
import time
import RNS
@@ -31,9 +52,18 @@ class SerialInterface(Interface):
serial = None
def __init__(self, owner, name, port, speed, databits, parity, stopbits):
import importlib
if importlib.util.find_spec('serial') != None:
import serial
else:
RNS.log("Using the Serial interface requires a serial communication module to be installed.", RNS.LOG_CRITICAL)
RNS.log("You can install one with the command: python3 -m pip install pyserial", RNS.LOG_CRITICAL)
RNS.panic()
self.rxb = 0
self.txb = 0
self.pyserial = serial
self.serial = None
self.owner = owner
self.name = name
@@ -44,6 +74,7 @@ class SerialInterface(Interface):
self.stopbits = stopbits
self.timeout = 100
self.online = False
self.bitrate = self.speed
if parity.lower() == "e" or parity.lower() == "even":
self.parity = serial.PARITY_EVEN
@@ -52,35 +83,43 @@ class SerialInterface(Interface):
self.parity = serial.PARITY_ODD
try:
RNS.log("Opening serial port "+self.port+"...")
self.serial = serial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
self.open_port()
except Exception as e:
RNS.log("Could not open serial port for interface "+str(self), RNS.LOG_ERROR)
raise e
if self.serial.is_open:
sleep(0.5)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
self.configure_device()
else:
raise IOError("Could not open serial port")
def open_port(self):
RNS.log("Opening serial port "+self.port+"...", RNS.LOG_VERBOSE)
self.serial = self.pyserial.Serial(
port = self.port,
baudrate = self.speed,
bytesize = self.databits,
parity = self.parity,
stopbits = self.stopbits,
xonxoff = False,
rtscts = False,
timeout = 0,
inter_byte_timeout = None,
write_timeout = None,
dsrdtr = False,
)
def configure_device(self):
sleep(0.5)
thread = threading.Thread(target=self.readLoop)
thread.setDaemon(True)
thread.start()
self.online = True
RNS.log("Serial port "+self.port+" is now open")
def processIncoming(self, data):
self.rxb += len(data)
self.owner.inbound(data, self)
@@ -132,13 +171,33 @@ class SerialInterface(Interface):
in_frame = False
escape = False
sleep(0.08)
except Exception as e:
self.online = False
RNS.log("A serial port error occurred, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is being torn down. Restart Reticulum to attempt to open this interface again.", RNS.LOG_ERROR)
RNS.log("The interface "+str(self)+" experienced an unrecoverable error and is now offline.", RNS.LOG_ERROR)
if RNS.Reticulum.panic_on_interface_error:
RNS.panic()
RNS.log("Reticulum will attempt to reconnect the interface periodically.", RNS.LOG_ERROR)
self.online = False
self.serial.close()
self.reconnect_port()
def reconnect_port(self):
while not self.online:
try:
time.sleep(5)
RNS.log("Attempting to reconnect serial port "+str(self.port)+" for "+str(self)+"...", RNS.LOG_VERBOSE)
self.open_port()
if self.serial.is_open:
self.configure_device()
except Exception as e:
RNS.log("Error while reconnecting port, the contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log("Reconnected serial port for "+str(self))
def __str__(self):
return "SerialInterface["+self.name+"]"
+159 -38
View File
@@ -1,7 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
import socketserver
import threading
import netifaces
import platform
import socket
import time
@@ -20,10 +41,26 @@ class HDLC():
data = data.replace(bytes([HDLC.FLAG]), bytes([HDLC.ESC, HDLC.FLAG^HDLC.ESC_MASK]))
return data
class KISS():
FEND = 0xC0
FESC = 0xDB
TFEND = 0xDC
TFESC = 0xDD
CMD_DATA = 0x00
CMD_UNKNOWN = 0xFE
@staticmethod
def escape(data):
data = data.replace(bytes([0xdb]), bytes([0xdb, 0xdd]))
data = data.replace(bytes([0xc0]), bytes([0xdb, 0xdc]))
return data
class ThreadingTCPServer(socketserver.ThreadingMixIn, socketserver.TCPServer):
pass
class TCPClientInterface(Interface):
BITRATE_GUESS = 10*1000*1000
RECONNECT_WAIT = 5
RECONNECT_MAX_TRIES = None
@@ -33,7 +70,12 @@ class TCPClientInterface(Interface):
TCP_PROBE_INTERVAL = 3
TCP_PROBES = 5
def __init__(self, owner, name, target_ip=None, target_port=None, connected_socket=None, max_reconnect_tries=None):
I2P_USER_TIMEOUT = 40
I2P_PROBE_AFTER = 10
I2P_PROBE_INTERVAL = 5
I2P_PROBES = 6
def __init__(self, owner, name, target_ip=None, target_port=None, connected_socket=None, max_reconnect_tries=None, kiss_framing=False, i2p_tunneled = False):
self.rxb = 0
self.txb = 0
@@ -49,7 +91,11 @@ class TCPClientInterface(Interface):
self.writing = False
self.online = False
self.detached = False
self.kiss_framing = kiss_framing
self.i2p_tunneled = i2p_tunneled
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
self.bitrate = TCPClientInterface.BITRATE_GUESS
if max_reconnect_tries == None:
self.max_reconnect_tries = TCPClientInterface.RECONNECT_MAX_TRIES
else:
@@ -80,16 +126,23 @@ class TCPClientInterface(Interface):
thread = threading.Thread(target=self.read_loop)
thread.setDaemon(True)
thread.start()
self.wants_tunnel = True
if not self.kiss_framing:
self.wants_tunnel = True
def set_timeouts_linux(self):
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, int(TCPClientInterface.TCP_USER_TIMEOUT * 1000))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, int(TCPClientInterface.TCP_PROBE_AFTER))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, int(TCPClientInterface.TCP_PROBE_INTERVAL))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPCNT, int(TCPClientInterface.TCP_PROBES))
if not self.i2p_tunneled:
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, int(TCPClientInterface.TCP_USER_TIMEOUT * 1000))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, int(TCPClientInterface.TCP_PROBE_AFTER))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, int(TCPClientInterface.TCP_PROBE_INTERVAL))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPCNT, int(TCPClientInterface.TCP_PROBES))
else:
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_USER_TIMEOUT, int(TCPClientInterface.I2P_USER_TIMEOUT * 1000))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, int(TCPClientInterface.I2P_PROBE_AFTER))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, int(TCPClientInterface.I2P_PROBE_INTERVAL))
self.socket.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPCNT, int(TCPClientInterface.I2P_PROBES))
def set_timeouts_osx(self):
if hasattr(socket, "TCP_KEEPALIVE"):
@@ -97,9 +150,13 @@ class TCPClientInterface(Interface):
else:
TCP_KEEPIDLE = 0x10
sock.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
sock.setsockopt(socket.IPPROTO_TCP, TCP_KEEPIDLE, int(TCPClientInterface.TCP_PROBE_AFTER))
self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)
if not self.i2p_tunneled:
self.socket.setsockopt(socket.IPPROTO_TCP, TCP_KEEPIDLE, int(TCPClientInterface.TCP_PROBE_AFTER))
else:
self.socket.setsockopt(socket.IPPROTO_TCP, TCP_KEEPIDLE, int(TCPClientInterface.I2P_PROBE_AFTER))
def detach(self):
if self.socket != None:
if hasattr(self.socket, "close"):
@@ -173,7 +230,8 @@ class TCPClientInterface(Interface):
thread = threading.Thread(target=self.read_loop)
thread.setDaemon(True)
thread.start()
RNS.Transport.synthesize_tunnel(self)
if not self.kiss_framing:
RNS.Transport.synthesize_tunnel(self)
else:
RNS.log("Attempt to reconnect on a non-initiator TCP interface. This should not happen.", RNS.LOG_ERROR)
@@ -193,7 +251,12 @@ class TCPClientInterface(Interface):
try:
self.writing = True
data = bytes([HDLC.FLAG])+HDLC.escape(data)+bytes([HDLC.FLAG])
if self.kiss_framing:
data = bytes([KISS.FEND])+bytes([KISS.CMD_DATA])+KISS.escape(data)+bytes([KISS.FEND])
else:
data = bytes([HDLC.FLAG])+HDLC.escape(data)+bytes([HDLC.FLAG])
self.socket.sendall(data)
self.writing = False
self.txb += len(data)
@@ -211,6 +274,7 @@ class TCPClientInterface(Interface):
in_frame = False
escape = False
data_buffer = b""
command = KISS.CMD_UNKNOWN
while True:
data_in = self.socket.recv(4096)
@@ -219,23 +283,53 @@ class TCPClientInterface(Interface):
while pointer < len(data_in):
byte = data_in[pointer]
pointer += 1
if (in_frame and byte == HDLC.FLAG):
in_frame = False
self.processIncoming(data_buffer)
elif (byte == HDLC.FLAG):
in_frame = True
data_buffer = b""
elif (in_frame and len(data_buffer) < RNS.Reticulum.MTU):
if (byte == HDLC.ESC):
escape = True
else:
if (escape):
if (byte == HDLC.FLAG ^ HDLC.ESC_MASK):
byte = HDLC.FLAG
if (byte == HDLC.ESC ^ HDLC.ESC_MASK):
byte = HDLC.ESC
escape = False
data_buffer = data_buffer+bytes([byte])
if self.kiss_framing:
# Read loop for KISS framing
if (in_frame and byte == KISS.FEND and command == KISS.CMD_DATA):
in_frame = False
self.processIncoming(data_buffer)
elif (byte == KISS.FEND):
in_frame = True
command = KISS.CMD_UNKNOWN
data_buffer = b""
elif (in_frame and len(data_buffer) < RNS.Reticulum.MTU):
if (len(data_buffer) == 0 and command == KISS.CMD_UNKNOWN):
# We only support one HDLC port for now, so
# strip off the port nibble
byte = byte & 0x0F
command = byte
elif (command == KISS.CMD_DATA):
if (byte == KISS.FESC):
escape = True
else:
if (escape):
if (byte == KISS.TFEND):
byte = KISS.FEND
if (byte == KISS.TFESC):
byte = KISS.FESC
escape = False
data_buffer = data_buffer+bytes([byte])
else:
# Read loop for HDLC framing
if (in_frame and byte == HDLC.FLAG):
in_frame = False
self.processIncoming(data_buffer)
elif (byte == HDLC.FLAG):
in_frame = True
data_buffer = b""
elif (in_frame and len(data_buffer) < RNS.Reticulum.MTU):
if (byte == HDLC.ESC):
escape = True
else:
if (escape):
if (byte == HDLC.FLAG ^ HDLC.ESC_MASK):
byte = HDLC.FLAG
if (byte == HDLC.ESC ^ HDLC.ESC_MASK):
byte = HDLC.ESC
escape = False
data_buffer = data_buffer+bytes([byte])
else:
self.online = False
if self.initiator and not self.detached:
@@ -275,7 +369,8 @@ class TCPClientInterface(Interface):
self.parent_interface.clients -= 1
if self in RNS.Transport.interfaces:
RNS.Transport.interfaces.remove(self)
if not self.initiator:
RNS.Transport.interfaces.remove(self)
def __str__(self):
@@ -283,14 +378,31 @@ class TCPClientInterface(Interface):
class TCPServerInterface(Interface):
BITRATE_GUESS = 10*1000*1000
@staticmethod
def get_address_for_if(name):
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['addr']
import importlib
if importlib.util.find_spec('netifaces') != None:
import netifaces
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['addr']
else:
RNS.log("Getting interface addresses from device names requires the netifaces module.", RNS.LOG_CRITICAL)
RNS.log("You can install it with the command: python3 -m pip install netifaces", RNS.LOG_CRITICAL)
RNS.panic()
@staticmethod
def get_broadcast_for_if(name):
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['broadcast']
import importlib
if importlib.util.find_spec('netifaces') != None:
import netifaces
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['broadcast']
else:
RNS.log("Getting interface addresses from device names requires the netifaces module.", RNS.LOG_CRITICAL)
RNS.log("You can install it with the command: python3 -m pip install netifaces", RNS.LOG_CRITICAL)
RNS.panic()
def __init__(self, owner, name, device=None, bindip=None, bindport=None):
def __init__(self, owner, name, device=None, bindip=None, bindport=None, i2p_tunneled=False):
self.rxb = 0
self.txb = 0
self.online = False
@@ -300,6 +412,9 @@ class TCPServerInterface(Interface):
self.OUT = False
self.name = name
self.i2p_tunneled = i2p_tunneled
self.mode = RNS.Interfaces.Interface.Interface.MODE_FULL
if device != None:
bindip = TCPServerInterface.get_address_for_if(device)
@@ -319,6 +434,8 @@ class TCPServerInterface(Interface):
ThreadingTCPServer.allow_reuse_address = True
self.server = ThreadingTCPServer(address, handlerFactory(self.incoming_connection))
self.bitrate = TCPServerInterface.BITRATE_GUESS
thread = threading.Thread(target=self.server.serve_forever)
thread.setDaemon(True)
thread.start()
@@ -329,12 +446,16 @@ class TCPServerInterface(Interface):
def incoming_connection(self, handler):
RNS.log("Accepting incoming TCP connection", RNS.LOG_VERBOSE)
interface_name = "Client on "+self.name
spawned_interface = TCPClientInterface(self.owner, interface_name, target_ip=None, target_port=None, connected_socket=handler.request)
spawned_interface = TCPClientInterface(self.owner, interface_name, target_ip=None, target_port=None, connected_socket=handler.request, i2p_tunneled=self.i2p_tunneled)
spawned_interface.OUT = self.OUT
spawned_interface.IN = self.IN
spawned_interface.target_ip = handler.client_address[0]
spawned_interface.target_port = str(handler.client_address[1])
spawned_interface.parent_interface = self
spawned_interface.bitrate = self.bitrate
spawned_interface.ifac_size = self.ifac_size
spawned_interface.ifac_netname = self.ifac_netname
spawned_interface.ifac_netkey = self.ifac_netkey
spawned_interface.online = True
RNS.log("Spawned new TCPClient Interface: "+str(spawned_interface), RNS.LOG_VERBOSE)
RNS.Transport.interfaces.append(spawned_interface)
@@ -353,4 +474,4 @@ class TCPInterfaceHandler(socketserver.BaseRequestHandler):
socketserver.BaseRequestHandler.__init__(self, *args, **keys)
def handle(self):
self.callback(handler=self)
self.callback(handler=self)
+42 -3
View File
@@ -1,7 +1,28 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from .Interface import Interface
import socketserver
import threading
import netifaces
import socket
import time
import sys
@@ -9,13 +30,29 @@ import RNS
class UDPInterface(Interface):
BITRATE_GUESS = 10*1000*1000
@staticmethod
def get_address_for_if(name):
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['addr']
import importlib
if importlib.util.find_spec('netifaces') != None:
import netifaces
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['addr']
else:
RNS.log("Getting interface addresses from device names requires the netifaces module.", RNS.LOG_CRITICAL)
RNS.log("You can install it with the command: python3 -m pip install netifaces", RNS.LOG_CRITICAL)
RNS.panic()
@staticmethod
def get_broadcast_for_if(name):
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['broadcast']
import importlib
if importlib.util.find_spec('netifaces') != None:
import netifaces
return netifaces.ifaddresses(name)[netifaces.AF_INET][0]['broadcast']
else:
RNS.log("Getting interface addresses from device names requires the netifaces module.", RNS.LOG_CRITICAL)
RNS.log("You can install it with the command: python3 -m pip install netifaces", RNS.LOG_CRITICAL)
RNS.panic()
def __init__(self, owner, name, device=None, bindip=None, bindport=None, forwardip=None, forwardport=None):
self.rxb = 0
@@ -24,6 +61,7 @@ class UDPInterface(Interface):
self.OUT = False
self.name = name
self.online = False
self.bitrate = UDPInterface.BITRATE_GUESS
if device != None:
if bindip == None:
@@ -44,6 +82,7 @@ class UDPInterface(Interface):
self.owner = owner
address = (self.bind_ip, self.bind_port)
socketserver.UDPServer.address_family = socket.AF_INET
self.server = socketserver.UDPServer(address, handlerFactory(self.processIncoming))
thread = threading.Thread(target=self.server.serve_forever)
+22
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import os
import glob
+81 -24
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from cryptography.hazmat.backends import default_backend
from cryptography.hazmat.primitives import hashes
from cryptography.hazmat.primitives import serialization
@@ -15,6 +37,8 @@ import RNS
import traceback
cio_default_backend = default_backend()
class LinkCallbacks:
def __init__(self):
self.link_established = None
@@ -43,13 +67,9 @@ class Link:
ECPUBSIZE = 32+32
KEYSIZE = 32
MDU = math.floor((RNS.Reticulum.MTU-RNS.Reticulum.HEADER_MINSIZE-RNS.Identity.FERNET_OVERHEAD)/RNS.Identity.AES128_BLOCKSIZE)*RNS.Identity.AES128_BLOCKSIZE - 1
MDU = math.floor((RNS.Reticulum.MTU-RNS.Reticulum.IFAC_MIN_SIZE-RNS.Reticulum.HEADER_MINSIZE-RNS.Identity.OPTIMISED_FERNET_OVERHEAD)/RNS.Identity.AES128_BLOCKSIZE)*RNS.Identity.AES128_BLOCKSIZE - 1
# This value is set at a reasonable level for a 1 Kb/s channel.
#
# TODO: Find a way to automatically raise or lower this according to
# channel bandwidth and utilisation.
ESTABLISHMENT_TIMEOUT_PER_HOP = 5
ESTABLISHMENT_TIMEOUT_PER_HOP = RNS.Reticulum.DEFAULT_PER_HOP_TIMEOUT
"""
Default timeout for link establishment in seconds per hop to destination.
"""
@@ -99,7 +119,7 @@ class Link:
except Exception as e:
RNS.log("Validating link request failed", RNS.LOG_VERBOSE)
traceback.print_exc()
RNS.log("exc: "+str(e))
return None
else:
@@ -127,6 +147,7 @@ class Link:
self.keepalive = Link.KEEPALIVE
self.watchdog_lock = False
self.status = Link.PENDING
self.activated_at = None
self.type = RNS.Destination.LINK
self.owner = owner
self.destination = destination
@@ -182,7 +203,7 @@ class Link:
self.start_watchdog()
self.packet.send()
self.had_outbound()
RNS.log("Link request "+RNS.prettyhexrep(self.link_id)+" sent to "+str(self.destination), RNS.LOG_VERBOSE)
RNS.log("Link request "+RNS.prettyhexrep(self.link_id)+" sent to "+str(self.destination), RNS.LOG_DEBUG)
def load_peer(self, peer_pub_bytes, peer_sig_pub_bytes):
@@ -202,11 +223,14 @@ class Link:
def handshake(self):
self.status = Link.HANDSHAKE
self.shared_key = self.prv.exchange(self.peer_pub)
# TODO: Improve this re-allocation of HKDF
self.derived_key = HKDF(
algorithm=hashes.SHA256(),
length=32,
salt=self.get_salt(),
info=self.get_context(),
backend=cio_default_backend,
).derive(self.shared_key)
def prove(self):
@@ -218,6 +242,7 @@ class Link:
proof.send()
self.had_outbound()
def prove_packet(self, packet):
signature = self.sign(packet.packet_hash)
# TODO: Hardcoded as explicit proof for now
@@ -249,6 +274,7 @@ class Link:
self.had_outbound()
self.status = Link.ACTIVE
self.activated_at = time.time()
if self.callbacks.link_established != None:
thread = threading.Thread(target=self.callbacks.link_established, args=(self,))
thread.setDaemon(True)
@@ -292,7 +318,7 @@ class Link:
packed_request = umsgpack.packb(unpacked_request)
if timeout == None:
timeout = self.rtt * self.traffic_timeout_factor + RNS.Resource.RESPONSE_MAX_GRACE_TIME
timeout = self.rtt * self.traffic_timeout_factor + RNS.Resource.RESPONSE_MAX_GRACE_TIME/4.0
if len(packed_request) <= Link.MDU:
request_packet = RNS.Packet(self, packed_request, RNS.Packet.DATA, context = RNS.Packet.REQUEST)
@@ -338,6 +364,8 @@ class Link:
rtt = umsgpack.unpackb(plaintext)
self.rtt = max(measured_rtt, rtt)
self.status = Link.ACTIVE
self.activated_at = time.time()
if self.owner.callbacks.link_established != None:
self.owner.callbacks.link_established(self)
@@ -425,7 +453,11 @@ class Link:
self.destination.links.remove(self)
if self.callbacks.link_closed != None:
self.callbacks.link_closed(self)
try:
self.callbacks.link_closed(self)
except Exception as e:
RNS.log("Error while executing link closed callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def start_watchdog(self):
thread = threading.Thread(target=self.__watchdog_job)
@@ -455,7 +487,7 @@ class Link:
sleep_time = next_check - time.time()
if time.time() >= self.request_time + self.establishment_timeout:
if self.initiator:
RNS.log("Timeout waiting link request proof", RNS.LOG_DEBUG)
RNS.log("Timeout waiting for link request proof", RNS.LOG_DEBUG)
else:
RNS.log("Timeout waiting for RTT packet from link initiator", RNS.LOG_DEBUG)
@@ -527,7 +559,7 @@ class Link:
else:
response_resource = RNS.Resource(packed_response, self, request_id = request_id, is_response = True)
else:
identity_string = RNS.prettyhexrep(self.get_remote_identity()) if self.get_remote_identity() != None else "<Unknown>"
identity_string = str(self.get_remote_identity()) if self.get_remote_identity() != None else "<Unknown>"
RNS.log("Request "+RNS.prettyhexrep(request_id)+" from "+identity_string+" not allowed for: "+str(path), RNS.LOG_DEBUG)
def handle_response(self, request_id, response_data, response_size, response_transfer_size):
@@ -546,7 +578,8 @@ class Link:
break
if remove != None:
self.pending_requests.remove(remove)
if remove in self.pending_requests:
self.pending_requests.remove(remove)
def request_resource_concluded(self, resource):
if resource.status == RNS.Resource.COMPLETE:
@@ -598,7 +631,10 @@ class Link:
elif self.destination.proof_strategy == RNS.Destination.PROVE_APP:
if self.destination.callbacks.proof_requested:
self.destination.callbacks.proof_requested(packet)
try:
self.destination.callbacks.proof_requested(packet)
except Exception as e:
RNS.log("Error while executing proof request callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
elif packet.context == RNS.Packet.LINKIDENTIFY:
plaintext = self.decrypt(packet.data)
@@ -613,7 +649,10 @@ class Link:
if identity.validate(signature, signed_data):
self.__remote_identity = identity
if self.callbacks.remote_identified != None:
self.callbacks.remote_identified(self.__remote_identity)
try:
self.callbacks.remote_identified(self.__remote_identity)
except Exception as e:
RNS.log("Error while executing remote identified callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
elif packet.context == RNS.Packet.REQUEST:
try:
@@ -659,8 +698,11 @@ class Link:
pass
elif self.resource_strategy == Link.ACCEPT_APP:
if self.callbacks.resource != None:
if self.callbacks.resource(resource):
RNS.Resource.accept(packet, self.callbacks.resource_concluded)
try:
if self.callbacks.resource(resource):
RNS.Resource.accept(packet, self.callbacks.resource_concluded)
except Exception as e:
RNS.log("Error while executing resource accept callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
elif self.resource_strategy == Link.ACCEPT_ALL:
RNS.Resource.accept(packet, self.callbacks.resource_concluded)
@@ -751,7 +793,7 @@ class Link:
return plaintext
except Exception as e:
RNS.log("Decryption failed on link "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
RNS.log(traceback.format_exc(), RNS.LOG_ERROR)
# RNS.log(traceback.format_exc(), RNS.LOG_ERROR)
# TODO: Think long about implications here
# self.teardown()
@@ -933,7 +975,10 @@ class RequestReceipt():
self.link.pending_requests.remove(self)
if self.callbacks.failed != None:
self.callbacks.failed(self)
try:
self.callbacks.failed(self)
except Exception as e:
RNS.log("Error while executing request failed callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def __response_timeout_job(self):
@@ -951,7 +996,10 @@ class RequestReceipt():
self.link.pending_requests.remove(self)
if self.callbacks.failed != None:
self.callbacks.failed(self)
try:
self.callbacks.failed(self)
except Exception as e:
RNS.log("Error while executing request timed out callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def response_resource_progress(self, resource):
@@ -967,7 +1015,10 @@ class RequestReceipt():
self.progress = resource.get_progress()
if self.callbacks.progress != None:
self.callbacks.progress(self)
try:
self.callbacks.progress(self)
except Exception as e:
RNS.log("Error while executing response progress callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
else:
resource.cancel()
@@ -987,10 +1038,16 @@ class RequestReceipt():
self.packet_receipt.callbacks.delivery(self.packet_receipt)
if self.callbacks.progress != None:
self.callbacks.progress(self)
try:
self.callbacks.progress(self)
except Exception as e:
RNS.log("Error while executing response progress callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
if self.callbacks.response != None:
self.callbacks.response(self)
try:
self.callbacks.response(self)
except Exception as e:
RNS.log("Error while executing response received callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def get_request_id(self):
"""
@@ -1034,4 +1091,4 @@ class RequestReceiptCallbacks:
def __init__(self):
self.response = None
self.failed = None
self.progress = None
self.progress = None
+71 -28
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import threading
import struct
import math
@@ -72,12 +94,10 @@ class Packet:
"""
PLAIN_MDU = MDU
"""
The maximum size of the payload data in a single unencrypted packet
The maximum size of the payload data in a single unencrypted packet
"""
# This value is set at a reasonable
# level for a 1 Kb/s channel.
TIMEOUT_PER_HOP = 5
TIMEOUT_PER_HOP = RNS.Reticulum.DEFAULT_PER_HOP_TIMEOUT
def __init__(self, destination, data, packet_type = DATA, context = NONE, transport_type = RNS.Transport.BROADCAST, header_type = HEADER_1, transport_id = None, attached_interface = None, create_receipt = True):
if destination != None:
@@ -113,6 +133,8 @@ class Packet:
self.attached_interface = attached_interface
self.receiving_interface = None
self.rssi = None
self.snr = None
def get_packed_flags(self):
if self.context == Packet.LRPROOF:
@@ -185,27 +207,33 @@ class Packet:
def unpack(self):
self.flags = self.raw[0]
self.hops = self.raw[1]
try:
self.flags = self.raw[0]
self.hops = self.raw[1]
self.header_type = (self.flags & 0b11000000) >> 6
self.transport_type = (self.flags & 0b00110000) >> 4
self.destination_type = (self.flags & 0b00001100) >> 2
self.packet_type = (self.flags & 0b00000011)
self.header_type = (self.flags & 0b11000000) >> 6
self.transport_type = (self.flags & 0b00110000) >> 4
self.destination_type = (self.flags & 0b00001100) >> 2
self.packet_type = (self.flags & 0b00000011)
if self.header_type == Packet.HEADER_2:
self.transport_id = self.raw[2:12]
self.destination_hash = self.raw[12:22]
self.context = ord(self.raw[22:23])
self.data = self.raw[23:]
else:
self.transport_id = None
self.destination_hash = self.raw[2:12]
self.context = ord(self.raw[12:13])
self.data = self.raw[13:]
if self.header_type == Packet.HEADER_2:
self.transport_id = self.raw[2:12]
self.destination_hash = self.raw[12:22]
self.context = ord(self.raw[22:23])
self.data = self.raw[23:]
else:
self.transport_id = None
self.destination_hash = self.raw[2:12]
self.context = ord(self.raw[12:13])
self.data = self.raw[13:]
self.packed = False
self.update_hash()
self.packed = False
self.update_hash()
return True
except Exception as e:
RNS.log("Received malformed packet, dropping it. The contained exception was: "+str(e), RNS.LOG_EXTREME)
return False
def send(self):
"""
@@ -328,6 +356,7 @@ class PacketReceipt:
self.destination = packet.destination
self.callbacks = PacketReceiptCallbacks()
self.concluded_at = None
self.proof_packet = None
if packet.destination.type == RNS.Destination.LINK:
self.timeout = packet.destination.rtt * packet.destination.traffic_timeout_factor
@@ -344,12 +373,12 @@ class PacketReceipt:
# Validate a proof packet
def validate_proof_packet(self, proof_packet):
if hasattr(proof_packet, "link") and proof_packet.link:
return self.validate_link_proof(proof_packet.data, proof_packet.link)
return self.validate_link_proof(proof_packet.data, proof_packet.link, proof_packet)
else:
return self.validate_proof(proof_packet.data)
return self.validate_proof(proof_packet.data, proof_packet)
# Validate a raw proof for a link
def validate_link_proof(self, proof, link):
def validate_link_proof(self, proof, link, proof_packet=None):
# TODO: Hardcoded as explicit proofs for now
if True or len(proof) == PacketReceipt.EXPL_LENGTH:
# This is an explicit proof
@@ -361,6 +390,8 @@ class PacketReceipt:
self.status = PacketReceipt.DELIVERED
self.proved = True
self.concluded_at = time.time()
self.proof_packet = proof_packet
if self.callbacks.delivery != None:
self.callbacks.delivery(self)
return True
@@ -388,7 +419,7 @@ class PacketReceipt:
return False
# Validate a raw proof
def validate_proof(self, proof):
def validate_proof(self, proof, proof_packet=None):
if len(proof) == PacketReceipt.EXPL_LENGTH:
# This is an explicit proof
proof_hash = proof[:RNS.Identity.HASHLENGTH//8]
@@ -399,8 +430,14 @@ class PacketReceipt:
self.status = PacketReceipt.DELIVERED
self.proved = True
self.concluded_at = time.time()
self.proof_packet = proof_packet
if self.callbacks.delivery != None:
self.callbacks.delivery(self)
try:
self.callbacks.delivery(self)
except Exception as e:
RNS.log("Error while executing proof validated callback. The contained exception was: "+str(e), RNS.LOG_ERROR)
return True
else:
return False
@@ -417,8 +454,14 @@ class PacketReceipt:
self.status = PacketReceipt.DELIVERED
self.proved = True
self.concluded_at = time.time()
self.proof_packet = proof_packet
if self.callbacks.delivery != None:
self.callbacks.delivery(self)
try:
self.callbacks.delivery(self)
except Exception as e:
RNS.log("Error while executing proof validated callback. The contained exception was: "+str(e), RNS.LOG_ERROR)
return True
else:
return False
+52 -29
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import os
import bz2
@@ -123,7 +145,10 @@ class Resource:
RNS.log("Accepting resource advertisement for "+RNS.prettyhexrep(resource.hash), RNS.LOG_DEBUG)
if resource.link.callbacks.resource_started != None:
resource.link.callbacks.resource_started(resource)
try:
resource.link.callbacks.resource_started(resource)
except Exception as e:
RNS.log("Error while executing resource started callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
resource.hashmap_update(0, resource.hashmap_raw)
@@ -140,6 +165,8 @@ class Resource:
def __init__(self, data, link, advertise=True, auto_compress=True, callback=None, progress_callback=None, timeout = None, segment_index = 1, original_hash = None, request_id = None, is_response = False):
data_size = None
resource_data = None
self.assembly_lock = False
if hasattr(data, "read"):
data_size = os.stat(data.name).st_size
self.total_size = data_size
@@ -321,10 +348,6 @@ class Resource:
self.request_next()
def get_map_hash(self, data):
# TODO: This will break if running unencrypted,
# uncompressed transfers on streams with long blocks
# of identical bytes. Doing so would be very silly
# anyways but maybe it should be handled gracefully.
return RNS.Identity.full_hash(data+self.random_hash)[:Resource.MAPHASH_LEN]
def advertise(self):
@@ -405,15 +428,6 @@ class Resource:
sleep_time = self.last_activity + (rtt*(self.part_timeout_factor+window_remaining)) + Resource.RETRY_GRACE_TIME - time.time()
# TODO: Remove debug info
# RNS.log("rtt "+str(rtt))
# RNS.log("ptof "+str(self.part_timeout_factor))
# RNS.log("wait "+str((rtt*self.part_timeout_factor) + Resource.RETRY_GRACE_TIME))
# RNS.log("sleep "+str(sleep_time))
# RNS.log("wndw "+str(self.window))
# RNS.log("wndwr "+str(window_remaining))
# RNS.log("")
if sleep_time < 0:
if self.retries_left > 0:
RNS.log("Timed out waiting for parts, requesting retry", RNS.LOG_DEBUG)
@@ -506,7 +520,10 @@ class Resource:
if self.segment_index == self.total_segments:
if self.callback != None:
self.data = open(self.storagepath, "rb")
self.callback(self)
try:
self.callback(self)
except Exception as e:
RNS.log("Error while executing resource assembled callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
try:
self.data.close()
@@ -540,7 +557,10 @@ class Resource:
# If all segments were processed, we'll
# signal that the resource sending concluded
if self.callback != None:
self.callback(self)
try:
self.callback(self)
except Exception as e:
RNS.log("Error while executing resource concluded callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
else:
# Otherwise we'll recursively create the
# next segment of the resource
@@ -596,20 +616,17 @@ class Resource:
cp += 1
if self.__progress_callback != None:
self.__progress_callback(self)
# TODO: Remove debug info
# RNS.log("outstanding_parts "+str(self.outstanding_parts))
# RNS.log("total_parts "+str(self.total_parts))
# RNS.log("received_count "+str(self.received_count))
try:
self.__progress_callback(self)
except Exception as e:
RNS.log("Error while executing progress callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
i += 1
self.receiving_part = False
# TODO: Remove
#if self.outstanding_parts == 0 and self.received_count == self.total_parts:
if self.received_count == self.total_parts:
if self.received_count == self.total_parts and not self.assembly_lock:
self.assembly_lock = True
self.assemble()
elif self.outstanding_parts == 0:
# TODO: Figure out if there is a mathematically
@@ -754,7 +771,10 @@ class Resource:
self.status = Resource.AWAITING_PROOF
if self.__progress_callback != None:
self.__progress_callback(self)
try:
self.__progress_callback(self)
except Exception as e:
RNS.log("Error while executing progress callback from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def cancel(self):
"""
@@ -774,8 +794,11 @@ class Resource:
self.link.cancel_incoming_resource(self)
if self.callback != None:
self.link.resource_concluded(self)
self.callback(self)
try:
self.link.resource_concluded(self)
self.callback(self)
except Exception as e:
RNS.log("Error while executing callbacks on resource cancel from "+str(self)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
def set_callback(self, callback):
self.callback = callback
@@ -804,7 +827,7 @@ class Resource:
return progress
def __str__(self):
return RNS.prettyhexrep(self.hash)+str(self.link)
return "<"+RNS.hexrep(self.hash)+"/"+RNS.hexrep(self.link.link_id)+">"
class ResourceAdvertisement:
+492 -307
View File
File diff suppressed because it is too large Load Diff
+478 -137
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import os
import RNS
import time
@@ -30,11 +52,12 @@ class Transport:
"""
Maximum amount of hops that Reticulum will transport a packet.
"""
PATHFINDER_C = 2.0 # Decay constant
PATHFINDER_R = 1 # Retransmit retries
PATHFINDER_T = 10 # Retry grace period
PATHFINDER_RW = 10 # Random window for announce rebroadcast
PATHFINDER_E = 60*60*24*7 # Path expiration in seconds
PATHFINDER_G = 5 # Retry grace period
PATHFINDER_RW = 0.5 # Random window for announce rebroadcast
PATHFINDER_E = 60*60*24*7 # Path expiration of one week
AP_PATH_TIME = 60*60*24 # Path expiration of one day for Access Point paths
# TODO: Calculate an optimal number for this in
# various situations
@@ -58,6 +81,7 @@ class Transport:
# TODO: "destination_table" should really be renamed to "path_table"
# Notes on memory usage: 1 megabyte of memory can store approximately
# 55.100 path table entries or approximately 22.300 link table entries.
announce_table = {} # A table for storing announces currently waiting to be retransmitted
destination_table = {} # A lookup table containing the next hop to a given destination
reverse_table = {} # A lookup table for storing packet hashes used to return proofs and replies
@@ -76,6 +100,12 @@ class Transport:
# Reticulum instance
local_client_interfaces = []
local_client_rssi_cache = []
local_client_snr_cache = []
LOCAL_CLIENT_CACHE_MAXSIZE = 512
pending_local_path_requests = {}
jobs_locked = False
jobs_running = False
job_interval = 0.250
@@ -91,6 +121,7 @@ class Transport:
@staticmethod
def start(reticulum_instance):
Transport.jobs_running = True
Transport.owner = reticulum_instance
if Transport.identity == None:
@@ -125,6 +156,7 @@ class Transport:
Transport.control_destinations.append(Transport.tunnel_synthesize_handler)
Transport.control_hashes.append(Transport.tunnel_synthesize_destination.hash)
Transport.jobs_running = False
thread = threading.Thread(target=Transport.jobloop)
thread.setDaemon(True)
thread.start()
@@ -265,12 +297,12 @@ class Transport:
for destination_hash in Transport.announce_table:
announce_entry = Transport.announce_table[destination_hash]
if announce_entry[2] > Transport.PATHFINDER_R:
RNS.log("Dropping announce for "+RNS.prettyhexrep(destination_hash)+", retries exceeded", RNS.LOG_DEBUG)
RNS.log("Completed announce processing for "+RNS.prettyhexrep(destination_hash)+", retry limit reached", RNS.LOG_EXTREME)
Transport.announce_table.pop(destination_hash)
break
else:
if time.time() > announce_entry[1]:
announce_entry[1] = time.time() + math.pow(Transport.PATHFINDER_C, announce_entry[4]) + Transport.PATHFINDER_T + Transport.PATHFINDER_RW
announce_entry[1] = time.time() + Transport.PATHFINDER_G + Transport.PATHFINDER_RW
announce_entry[2] += 1
packet = announce_entry[5]
block_rebroadcasts = announce_entry[7]
@@ -300,6 +332,7 @@ class Transport:
RNS.log("Rebroadcasting announce as path response for "+RNS.prettyhexrep(announce_destination.hash)+" with hop count "+str(new_packet.hops), RNS.LOG_DEBUG)
else:
RNS.log("Rebroadcasting announce for "+RNS.prettyhexrep(announce_destination.hash)+" with hop count "+str(new_packet.hops), RNS.LOG_DEBUG)
outgoing.append(new_packet)
# This handles an edge case where a peer sends a past
@@ -357,7 +390,7 @@ class Transport:
expires = tunnel_entry[3]
if time.time() > expires:
stale_tunnels.append(tunnel_id)
RNS.log("Tunnel "+RNS.prettyhexrep(tunnel_id)+" timed out and was removed", RNS.LOG_DEBUG)
RNS.log("Tunnel "+RNS.prettyhexrep(tunnel_id)+" timed out and was removed", RNS.LOG_EXTREME)
else:
stale_tunnel_paths = []
tunnel_paths = tunnel_entry[2]
@@ -366,7 +399,7 @@ class Transport:
if time.time() > tunnel_path_entry[0] + Transport.DESTINATION_TIMEOUT:
stale_tunnel_paths.append(tunnel_path)
RNS.log("Tunnel path to "+RNS.prettyhexrep(tunnel_path)+" timed out and was removed", RNS.LOG_DEBUG)
RNS.log("Tunnel path to "+RNS.prettyhexrep(tunnel_path)+" timed out and was removed", RNS.LOG_EXTREME)
for tunnel_path in stale_tunnel_paths:
tunnel_paths.pop(tunnel_path)
@@ -375,9 +408,9 @@ class Transport:
if ti > 0:
if ti == 1:
RNS.log("Removed "+str(ti)+" tunnel path", RNS.LOG_DEBUG)
RNS.log("Removed "+str(ti)+" tunnel path", RNS.LOG_EXTREME)
else:
RNS.log("Removed "+str(ti)+" tunnel paths", RNS.LOG_DEBUG)
RNS.log("Removed "+str(ti)+" tunnel paths", RNS.LOG_EXTREME)
@@ -388,9 +421,9 @@ class Transport:
if i > 0:
if i == 1:
RNS.log("Dropped "+str(i)+" reverse table entry", RNS.LOG_DEBUG)
RNS.log("Released "+str(i)+" reverse table entry", RNS.LOG_EXTREME)
else:
RNS.log("Dropped "+str(i)+" reverse table entries", RNS.LOG_DEBUG)
RNS.log("Released "+str(i)+" reverse table entries", RNS.LOG_EXTREME)
@@ -401,9 +434,9 @@ class Transport:
if i > 0:
if i == 1:
RNS.log("Dropped "+str(i)+" link", RNS.LOG_DEBUG)
RNS.log("Released "+str(i)+" link", RNS.LOG_EXTREME)
else:
RNS.log("Dropped "+str(i)+" links", RNS.LOG_DEBUG)
RNS.log("Released "+str(i)+" links", RNS.LOG_EXTREME)
i = 0
for destination_hash in stale_paths:
@@ -412,9 +445,9 @@ class Transport:
if i > 0:
if i == 1:
RNS.log("Removed "+str(i)+" path", RNS.LOG_DEBUG)
RNS.log("Removed "+str(i)+" path", RNS.LOG_EXTREME)
else:
RNS.log("Removed "+str(i)+" paths", RNS.LOG_DEBUG)
RNS.log("Removed "+str(i)+" paths", RNS.LOG_EXTREME)
i = 0
for tunnel_id in stale_tunnels:
@@ -423,9 +456,9 @@ class Transport:
if i > 0:
if i == 1:
RNS.log("Removed "+str(i)+" tunnel", RNS.LOG_DEBUG)
RNS.log("Removed "+str(i)+" tunnel", RNS.LOG_EXTREME)
else:
RNS.log("Removed "+str(i)+" tunnels", RNS.LOG_DEBUG)
RNS.log("Removed "+str(i)+" tunnels", RNS.LOG_EXTREME)
Transport.tables_last_culled = time.time()
@@ -439,18 +472,40 @@ class Transport:
for packet in outgoing:
packet.send()
@staticmethod
def transmit(interface, raw):
try:
if hasattr(interface, "ifac_identity") and interface.ifac_identity != None:
# Calculate packet access code
ifac = interface.ifac_identity.sign(raw)[-interface.ifac_size:]
# Set IFAC flag
new_header = bytes([raw[0] | 0x80, raw[1]])
# Assemble new payload with IFAC and send it
new_raw = new_header+ifac+raw[2:]
interface.processOutgoing(new_raw)
else:
interface.processOutgoing(raw)
except Exception as e:
RNS.log("Error while transmitting on "+str(interface)+". The contained exception was: "+str(e), RNS.LOG_ERROR)
@staticmethod
def outbound(packet):
while (Transport.jobs_running):
# TODO: Profile actual impact here on faster links
sleep(0.01)
Transport.jobs_locked = True
# TODO: This updateHash call might be redundant
packet.update_hash()
sent = False
outbound_time = time.time()
# Check if we have a known path for the destination in the path table
if packet.packet_type != RNS.Packet.ANNOUNCE and packet.destination_hash in Transport.destination_table:
if packet.packet_type != RNS.Packet.ANNOUNCE and packet.destination.type != RNS.Destination.PLAIN and packet.destination.type != RNS.Destination.GROUP and packet.destination_hash in Transport.destination_table:
outbound_interface = Transport.destination_table[packet.destination_hash][5]
# If there's more than one hop to the destination, and we know
@@ -466,7 +521,7 @@ class Transport:
new_raw += packet.raw[1:2]
new_raw += Transport.destination_table[packet.destination_hash][1]
new_raw += packet.raw[2:]
outbound_interface.processOutgoing(new_raw)
Transport.transmit(outbound_interface, new_raw)
Transport.destination_table[packet.destination_hash][0] = time.time()
sent = True
@@ -485,7 +540,7 @@ class Transport:
new_raw += packet.raw[1:2]
new_raw += Transport.destination_table[packet.destination_hash][1]
new_raw += packet.raw[2:]
outbound_interface.processOutgoing(new_raw)
Transport.transmit(outbound_interface, new_raw)
Transport.destination_table[packet.destination_hash][0] = time.time()
sent = True
@@ -493,7 +548,7 @@ class Transport:
# directly reachable, and also on which interface, so we
# simply transmit the packet directly on that one.
else:
outbound_interface.processOutgoing(packet.raw)
Transport.transmit(outbound_interface, packet.raw)
sent = True
# If we don't have a known path for the destination, we'll
@@ -505,20 +560,88 @@ class Transport:
for interface in Transport.interfaces:
if interface.OUT:
should_transmit = True
if packet.destination.type == RNS.Destination.LINK:
if packet.destination.status == RNS.Link.CLOSED:
should_transmit = False
if interface != packet.destination.attached_interface:
should_transmit = False
if packet.attached_interface != None and interface != packet.attached_interface:
should_transmit = False
if packet.packet_type == RNS.Packet.ANNOUNCE:
if packet.attached_interface == None:
if interface.mode == RNS.Interfaces.Interface.Interface.MODE_ACCESS_POINT:
RNS.log("Blocking announce broadcast on "+str(interface)+" due to AP mode", RNS.LOG_EXTREME)
should_transmit = False
else:
# Currently, annouces originating locally are always
# allowed, and do not conform to bandwidth caps.
# TODO: Rethink whether this is actually optimal.
if packet.hops > 0:
if not hasattr(interface, "announce_cap"):
interface.announce_cap = RNS.Reticulum.ANNOUNCE_CAP
if not hasattr(interface, "announce_allowed_at"):
interface.announce_allowed_at = 0
if not hasattr(interface, "announce_queue"):
interface.announce_queue = []
queued_announces = True if len(interface.announce_queue) > 0 else False
if not queued_announces and outbound_time > interface.announce_allowed_at:
tx_time = (len(packet.raw)*8) / interface.bitrate
wait_time = (tx_time / interface.announce_cap)
interface.announce_allowed_at = outbound_time + wait_time
# TODO: Clean
# wait_time_str = str(round(wait_time*1000,3))+"ms"
# RNS.log("Next announce on "+str(interface)+" allowed in "+wait_time_str, RNS.LOG_EXTREME)
else:
should_transmit = False
if not len(interface.announce_queue) >= RNS.Reticulum.MAX_QUEUED_ANNOUNCES:
entry = {"time": outbound_time, "hops": packet.hops, "raw": packet.raw}
queued_announces = True if len(interface.announce_queue) > 0 else False
interface.announce_queue.append(entry)
if not queued_announces:
wait_time = max(interface.announce_allowed_at - time.time(), 0)
timer = threading.Timer(wait_time, interface.process_announce_queue)
timer.start()
wait_time_str = str(round(wait_time*1000,3))+"ms"
ql_str = str(len(interface.announce_queue))
RNS.log("Added announce to queue (height "+ql_str+") on "+str(interface)+" for processing in "+wait_time_str, RNS.LOG_EXTREME)
else:
wait_time = max(interface.announce_allowed_at - time.time(), 0)
wait_time_str = str(round(wait_time*1000,3))+"ms"
ql_str = str(len(interface.announce_queue))
RNS.log("Added announce to queue (height "+ql_str+") on "+str(interface)+" for processing in "+wait_time_str, RNS.LOG_EXTREME)
else:
pass
else:
pass
if should_transmit:
if not stored_hash:
Transport.packet_hashlist.append(packet.packet_hash)
stored_hash = True
interface.processOutgoing(packet.raw)
def send_packet():
Transport.transmit(interface, packet.raw)
thread = threading.Thread(target=send_packet)
thread.daemon = True
thread.start()
sent = True
if sent:
@@ -559,34 +682,121 @@ class Transport:
return True
if packet.context == RNS.Packet.CACHE_REQUEST:
return True
if packet.destination_type == RNS.Destination.PLAIN:
return True
if packet.packet_type != RNS.Packet.ANNOUNCE:
if packet.hops > 1:
RNS.log("Dropped PLAIN packet "+RNS.prettyhexrep(packet.hash)+" with "+str(packet.hops)+" hops", RNS.LOG_DEBUG)
return False
else:
return True
else:
RNS.log("Dropped invalid PLAIN announce packet", RNS.LOG_DEBUG)
return False
if packet.destination_type == RNS.Destination.GROUP:
if packet.packet_type != RNS.Packet.ANNOUNCE:
if packet.hops > 1:
RNS.log("Dropped GROUP packet "+RNS.prettyhexrep(packet.hash)+" with "+str(packet.hops)+" hops", RNS.LOG_DEBUG)
return False
else:
return True
else:
RNS.log("Dropped invalid GROUP announce packet", RNS.LOG_DEBUG)
return False
if not packet.packet_hash in Transport.packet_hashlist:
return True
else:
if packet.packet_type == RNS.Packet.ANNOUNCE:
return True
if packet.destination_type == RNS.Destination.SINGLE:
return True
else:
RNS.log("Dropped invalid announce packet", RNS.LOG_DEBUG)
return False
RNS.log("Filtered packet with hash "+RNS.prettyhexrep(packet.packet_hash), RNS.LOG_DEBUG)
RNS.log("Filtered packet with hash "+RNS.prettyhexrep(packet.packet_hash), RNS.LOG_EXTREME)
return False
@staticmethod
def inbound(raw, interface=None):
# If interface access codes are enabled,
# we must authenticate each packet.
if interface != None and hasattr(interface, "ifac_identity") and interface.ifac_identity != None:
# Check that IFAC flag is set
if raw[0] & 0x80 == 0x80:
if len(raw) > 2+interface.ifac_size:
# Extract IFAC
ifac = raw[2:2+interface.ifac_size]
# Unset IFAC flag
new_header = bytes([raw[0] & 0x7f, raw[1]])
# Re-assemble packet
new_raw = new_header+raw[2+interface.ifac_size:]
# Calculate expected IFAC
expected_ifac = interface.ifac_identity.sign(new_raw)[-interface.ifac_size:]
# Check it
if ifac == expected_ifac:
raw = new_raw
else:
return
else:
return
else:
# If the IFAC flag is not set, but should be,
# drop the packet.
return
else:
# If the interface does not have IFAC enabled,
# check the received packet IFAC flag.
if raw[0] & 0x80 == 0x80:
# If the flag is set, drop the packet
return
while (Transport.jobs_running):
sleep(0.01)
if Transport.identity == None:
return
Transport.jobs_locked = True
packet = RNS.Packet(None, raw)
packet.unpack()
if not packet.unpack():
return
packet.receiving_interface = interface
packet.hops += 1
if len(Transport.local_client_interfaces) > 0:
if interface != None:
if hasattr(interface, "r_stat_rssi"):
if interface.r_stat_rssi != None:
packet.rssi = interface.r_stat_rssi
if len(Transport.local_client_interfaces) > 0:
Transport.local_client_rssi_cache.append([packet.packet_hash, packet.rssi])
while len(Transport.local_client_rssi_cache) > Transport.LOCAL_CLIENT_CACHE_MAXSIZE:
Transport.local_client_rssi_cache.pop()
if hasattr(interface, "r_stat_snr"):
if interface.r_stat_rssi != None:
packet.snr = interface.r_stat_snr
if len(Transport.local_client_interfaces) > 0:
Transport.local_client_snr_cache.append([packet.packet_hash, packet.snr])
while len(Transport.local_client_snr_cache) > Transport.LOCAL_CLIENT_CACHE_MAXSIZE:
Transport.local_client_snr_cache.pop()
if len(Transport.local_client_interfaces) > 0:
if Transport.is_local_client_interface(interface):
packet.hops -= 1
elif Transport.interface_to_shared_instance(interface):
packet.hops -= 1
@@ -612,12 +822,12 @@ class Transport:
if from_local_client:
for interface in Transport.interfaces:
if interface != packet.receiving_interface:
interface.processOutgoing(packet.raw)
Transport.transmit(interface, packet.raw)
# If the packet was not from a local client, send
# it directly to all local clients
else:
for interface in Transport.local_client_interfaces:
interface.processOutgoing(packet.raw)
Transport.transmit(interface, packet.raw)
# General transport handling. Takes care of directing
@@ -670,7 +880,7 @@ class Transport:
new_raw += packet.raw[2:]
outbound_interface = Transport.destination_table[packet.destination_hash][5]
outbound_interface.processOutgoing(new_raw)
Transport.transmit(outbound_interface, new_raw)
Transport.destination_table[packet.destination_hash][0] = time.time()
if packet.packet_type == RNS.Packet.LINKREQUEST:
@@ -698,7 +908,7 @@ class Transport:
# TODO: There should probably be some kind of REJECT
# mechanism here, to signal to the source that their
# expected path failed.
RNS.log("Got packet in transport, but no known path to final destination. Dropping packet.", RNS.LOG_DEBUG)
RNS.log("Got packet in transport, but no known path to final destination "+RNS.prettyhexrep(packet.destination_hash)+". Dropping packet.", RNS.LOG_EXTREME)
# Link transport handling. Directs packets according
# to entries in the link tables
@@ -731,7 +941,7 @@ class Transport:
new_raw = packet.raw[0:1]
new_raw += struct.pack("!B", packet.hops)
new_raw += packet.raw[2:]
outbound_interface.processOutgoing(new_raw)
Transport.transmit(outbound_interface, new_raw)
Transport.link_table[packet.destination_hash][0] = time.time()
else:
pass
@@ -742,7 +952,7 @@ class Transport:
# of queued announce rebroadcasts once handed to the next node.
if packet.packet_type == RNS.Packet.ANNOUNCE:
local_destination = next((d for d in Transport.destinations if d.hash == packet.destination_hash), None)
if local_destination == None and RNS.Identity.validate_announce(packet):
if local_destination == None and RNS.Identity.validate_announce(packet):
if packet.transport_id != None:
received_from = packet.transport_id
@@ -762,7 +972,7 @@ class Transport:
if packet.hops-1 == announce_entry[4]+1 and announce_entry[2] > 0:
now = time.time()
if now < announce_entry[1]:
RNS.log("Rebroadcasted announce for "+RNS.prettyhexrep(packet.destination_hash)+" has been passed on to next node, no further tries needed", RNS.LOG_DEBUG)
RNS.log("Rebroadcasted announce for "+RNS.prettyhexrep(packet.destination_hash)+" has been passed on to another node, no further tries needed", RNS.LOG_DEBUG)
Transport.announce_table.pop(packet.destination_hash)
else:
@@ -775,7 +985,8 @@ class Transport:
# First, check that the announce is not for a destination
# local to this system, and that hops are less than the max
if (not any(packet.destination_hash == d.hash for d in Transport.destinations) and packet.hops < Transport.PATHFINDER_M+1):
random_blob = packet.data[RNS.Identity.KEYSIZE//8+10:RNS.Identity.KEYSIZE//8+20]
random_blob = packet.data[RNS.Identity.KEYSIZE//8:RNS.Identity.KEYSIZE//8+RNS.Reticulum.TRUNCATED_HASHLENGTH//8]
announce_emitted = int.from_bytes(random_blob[5:10], "big")
random_blobs = []
if packet.destination_hash in Transport.destination_table:
random_blobs = Transport.destination_table[packet.destination_hash][4]
@@ -796,9 +1007,19 @@ class Transport:
else:
# If an announce arrives with a larger hop
# count than we already have in the table,
# ignore it, unless the path is expired
if (time.time() > Transport.destination_table[packet.destination_hash][3]):
# We also check that the announce hash is
# ignore it, unless the path is expired, or
# the emission timestamp is more recent.
now = time.time()
path_expires = Transport.destination_table[packet.destination_hash][3]
path_announce_emitted = 0
for path_random_blob in random_blobs:
path_announce_emitted = max(path_announce_emitted, int.from_bytes(path_random_blob[5:10], "big"))
if path_announce_emitted >= announce_emitted:
break
if (now >= path_expires):
# We also check that the announce is
# different from ones we've already heard,
# to avoid loops in the network
if not random_blob in random_blobs:
@@ -809,7 +1030,13 @@ class Transport:
else:
should_add = False
else:
should_add = False
if (announce_emitted > path_announce_emitted):
if not random_blob in random_blobs:
RNS.log("Replacing destination table entry for "+str(RNS.prettyhexrep(packet.destination_hash))+" with new announce, since it was more recently emitted", RNS.LOG_DEBUG)
should_add = True
else:
should_add = False
else:
# If this destination is unknown in our table
# we should add it
@@ -818,20 +1045,26 @@ class Transport:
if should_add:
now = time.time()
retries = 0
expires = now + Transport.PATHFINDER_E
announce_hops = packet.hops
local_rebroadcasts = 0
block_rebroadcasts = False
attached_interface = None
retransmit_timeout = now + math.pow(Transport.PATHFINDER_C, packet.hops) + (RNS.rand() * Transport.PATHFINDER_RW)
retransmit_timeout = now + (RNS.rand() * Transport.PATHFINDER_RW)
if packet.receiving_interface.mode == RNS.Interfaces.Interface.Interface.MODE_ACCESS_POINT:
expires = now + Transport.AP_PATH_TIME
else:
expires = now + Transport.PATHFINDER_E
random_blobs.append(random_blob)
if (RNS.Reticulum.transport_enabled() or Transport.from_local_client(packet)) and packet.context != RNS.Packet.PATH_RESPONSE:
# If the announce is from a local client,
# we announce it immediately, but only one
# time.
# Insert announce into announce table for retransmission
if Transport.from_local_client(packet):
# If the announce is from a local client,
# it is announced immediately, but only one time.
retransmit_timeout = now
retries = Transport.PATHFINDER_R
@@ -847,6 +1080,29 @@ class Transport:
attached_interface
]
# TODO: Check from_local_client once and store result
elif Transport.from_local_client(packet) and packet.context == RNS.Packet.PATH_RESPONSE:
# If this is a path response from a local client,
# check if any external interfaces have pending
# path requests.
if packet.destination_hash in Transport.pending_local_path_requests:
desiring_interface = Transport.pending_local_path_requests.pop(packet.destination_hash)
retransmit_timeout = now
retries = Transport.PATHFINDER_R
Transport.announce_table[packet.destination_hash] = [
now,
retransmit_timeout,
retries,
received_from,
announce_hops,
packet,
local_rebroadcasts,
block_rebroadcasts,
attached_interface
]
# If we have any local clients connected, we re-
# transmit the announce to them immediately
if (len(Transport.local_client_interfaces)):
@@ -858,8 +1114,8 @@ class Transport:
announce_data = packet.data
if Transport.from_local_client(packet) and packet.context == RNS.Packet.PATH_RESPONSE:
for interface in Transport.interfaces:
if packet.receiving_interface != interface:
for local_interface in Transport.local_client_interfaces:
if packet.receiving_interface != local_interface:
new_announce = RNS.Packet(
announce_destination,
announce_data,
@@ -868,7 +1124,7 @@ class Transport:
header_type = RNS.Packet.HEADER_2,
transport_type = Transport.TRANSPORT,
transport_id = Transport.identity.hash,
attached_interface = interface
attached_interface = local_interface
)
new_announce.hops = packet.hops
@@ -876,23 +1132,24 @@ class Transport:
else:
for local_interface in Transport.local_client_interfaces:
new_announce = RNS.Packet(
announce_destination,
announce_data,
RNS.Packet.ANNOUNCE,
context = announce_context,
header_type = RNS.Packet.HEADER_2,
transport_type = Transport.TRANSPORT,
transport_id = Transport.identity.hash,
attached_interface = local_interface
)
if packet.receiving_interface != local_interface:
new_announce = RNS.Packet(
announce_destination,
announce_data,
RNS.Packet.ANNOUNCE,
context = announce_context,
header_type = RNS.Packet.HEADER_2,
transport_type = Transport.TRANSPORT,
transport_id = Transport.identity.hash,
attached_interface = local_interface
)
new_announce.hops = packet.hops
new_announce.send()
new_announce.hops = packet.hops
new_announce.send()
destination_table_entry = [now, received_from, announce_hops, expires, random_blobs, packet.receiving_interface, packet]
Transport.destination_table[packet.destination_hash] = destination_table_entry
RNS.log("Path to "+RNS.prettyhexrep(packet.destination_hash)+" is now "+str(announce_hops)+" hops away via "+RNS.prettyhexrep(received_from)+" on "+str(packet.receiving_interface), RNS.LOG_VERBOSE)
RNS.log("Destination "+RNS.prettyhexrep(packet.destination_hash)+" is now "+str(announce_hops)+" hops away via "+RNS.prettyhexrep(received_from)+" on "+str(packet.receiving_interface), RNS.LOG_DEBUG)
# If the receiving interface is a tunnel, we add the
# announce to the tunnels table
@@ -902,33 +1159,34 @@ class Transport:
paths[packet.destination_hash] = destination_table_entry
expires = time.time() + Transport.DESTINATION_TIMEOUT
tunnel_entry[3] = expires
RNS.log("Path to "+RNS.prettyhexrep(packet.destination_hash)+" associated with tunnel "+RNS.prettyhexrep(packet.receiving_interface.tunnel_id), RNS.LOG_VERBOSE)
RNS.log("Path to "+RNS.prettyhexrep(packet.destination_hash)+" associated with tunnel "+RNS.prettyhexrep(packet.receiving_interface.tunnel_id), RNS.LOG_DEBUG)
# Call externally registered callbacks from apps
# wanting to know when an announce arrives
for handler in Transport.announce_handlers:
try:
# Check that the announced destination matches
# the handlers aspect filter
execute_callback = False
if handler.aspect_filter == None:
# If the handlers aspect filter is set to
# None, we execute the callback in all cases
execute_callback = True
else:
announce_identity = RNS.Identity.recall(packet.destination_hash)
handler_expected_hash = RNS.Destination.hash_from_name_and_identity(handler.aspect_filter, announce_identity)
if packet.destination_hash == handler_expected_hash:
if packet.context != RNS.Packet.PATH_RESPONSE:
for handler in Transport.announce_handlers:
try:
# Check that the announced destination matches
# the handlers aspect filter
execute_callback = False
if handler.aspect_filter == None:
# If the handlers aspect filter is set to
# None, we execute the callback in all cases
execute_callback = True
if execute_callback:
handler.received_announce(
destination_hash=packet.destination_hash,
announced_identity=announce_identity,
app_data=RNS.Identity.recall_app_data(packet.destination_hash)
)
except Exception as e:
RNS.log("Error while processing external announce callback.", RNS.LOG_ERROR)
RNS.log("The contained exception was: "+str(e), RNS.LOG_ERROR)
else:
announce_identity = RNS.Identity.recall(packet.destination_hash)
handler_expected_hash = RNS.Destination.hash_from_name_and_identity(handler.aspect_filter, announce_identity)
if packet.destination_hash == handler_expected_hash:
execute_callback = True
if execute_callback:
handler.received_announce(
destination_hash=packet.destination_hash,
announced_identity=announce_identity,
app_data=RNS.Identity.recall_app_data(packet.destination_hash)
)
except Exception as e:
RNS.log("Error while processing external announce callback.", RNS.LOG_ERROR)
RNS.log("The contained exception was: "+str(e), RNS.LOG_ERROR)
# Handling for linkrequests to local destinations
elif packet.packet_type == RNS.Packet.LINKREQUEST:
@@ -955,8 +1213,11 @@ class Transport:
elif destination.proof_strategy == RNS.Destination.PROVE_APP:
if destination.callbacks.proof_requested:
if destination.callbacks.proof_requested(packet):
packet.prove()
try:
if destination.callbacks.proof_requested(packet):
packet.prove()
except Exception as e:
RNS.log("Error while executing proof request callback. The contained exception was: "+str(e), RNS.LOG_ERROR)
# Handling for proofs and link-request proofs
elif packet.packet_type == RNS.Packet.PROOF:
@@ -973,7 +1234,7 @@ class Transport:
new_raw += struct.pack("!B", packet.hops)
new_raw += packet.raw[2:]
Transport.link_table[packet.destination_hash][7] = True
link_entry[4].processOutgoing(new_raw)
Transport.transmit(link_entry[4], new_raw)
else:
RNS.log("Link request proof received on wrong interface, not transporting it.", RNS.LOG_DEBUG)
else:
@@ -1006,7 +1267,7 @@ class Transport:
new_raw = packet.raw[0:1]
new_raw += struct.pack("!B", packet.hops)
new_raw += packet.raw[2:]
reverse_entry[0].processOutgoing(new_raw)
Transport.transmit(reverse_entry[0], new_raw)
else:
RNS.log("Proof received on wrong interface, not transporting it.", RNS.LOG_DEBUG)
@@ -1092,6 +1353,7 @@ class Transport:
interface.tunnel_id = tunnel_id
paths = tunnel_entry[2]
deprecated_paths = []
for destination_hash, path_entry in paths.items():
received_from = path_entry[1]
announce_hops = path_entry[2]
@@ -1111,15 +1373,20 @@ class Transport:
else:
RNS.log("Did not restore path to "+RNS.prettyhexrep(packet.destination_hash)+" because a newer path with fewer hops exist", RNS.LOG_DEBUG)
else:
should_add = True
if time.time() < expires:
should_add = True
else:
RNS.log("Did not restore path to "+RNS.prettyhexrep(packet.destination_hash)+" because it has expired", RNS.LOG_DEBUG)
if should_add:
Transport.destination_table[destination_hash] = new_entry
RNS.log("Restored path to "+RNS.prettyhexrep(packet.destination_hash)+" is now "+str(announce_hops)+" hops away via "+RNS.prettyhexrep(received_from)+" on "+str(receiving_interface), RNS.LOG_DEBUG)
else:
deprecated_paths.append(destination_hash)
for deprecated_path in deprecated_paths:
RNS.log("Removing path to "+RNS.prettyhexrep(deprecated_path)+" from tunnel "+RNS.prettyhexrep(tunnel_id), RNS.LOG_DEBUG)
paths.pop(deprecated_path)
@staticmethod
def register_destination(destination):
@@ -1315,88 +1582,140 @@ class Transport:
return None
@staticmethod
def request_path(destination_hash):
def expire_path(destination_hash):
if destination_hash in Transport.destination_table:
Transport.destination_table[destination_hash][0] = 0
Transport.tables_last_culled = 0
return True
else:
return False
@staticmethod
def request_path(destination_hash, on_interface=None):
"""
Requests a path to the destination from the network. If
another reachable peer on the network knows a path, it
will announce it.
:param destination_hash: A destination hash as *bytes*.
:param on_interface: If specified, the path request will only be sent on this interface. In normal use, Reticulum handles this automatically, and this parameter should not be used.
"""
path_request_data = destination_hash + RNS.Identity.get_random_hash()
path_request_dst = RNS.Destination(None, RNS.Destination.OUT, RNS.Destination.PLAIN, Transport.APP_NAME, "path", "request")
packet = RNS.Packet(path_request_dst, path_request_data, packet_type = RNS.Packet.DATA, transport_type = RNS.Transport.BROADCAST, header_type = RNS.Packet.HEADER_1)
packet.send()
if RNS.Reticulum.transport_enabled():
path_request_data = destination_hash+Transport.identity.hash+RNS.Identity.get_random_hash()
else:
path_request_data = destination_hash+RNS.Identity.get_random_hash()
@staticmethod
def request_path_on_interface(destination_hash, interface):
path_request_data = destination_hash + RNS.Identity.get_random_hash()
path_request_dst = RNS.Destination(None, RNS.Destination.OUT, RNS.Destination.PLAIN, Transport.APP_NAME, "path", "request")
packet = RNS.Packet(path_request_dst, path_request_data, packet_type = RNS.Packet.DATA, transport_type = RNS.Transport.BROADCAST, header_type = RNS.Packet.HEADER_1, attached_interface = interface)
packet = RNS.Packet(path_request_dst, path_request_data, packet_type = RNS.Packet.DATA, transport_type = RNS.Transport.BROADCAST, header_type = RNS.Packet.HEADER_1, attached_interface = on_interface)
packet.send()
@staticmethod
def path_request_handler(data, packet):
if len(data) >= RNS.Identity.TRUNCATED_HASHLENGTH//8:
Transport.path_request(
data[:RNS.Identity.TRUNCATED_HASHLENGTH//8],
Transport.from_local_client(packet),
packet.receiving_interface
)
try:
# If there is at least bytes enough for a destination
# hash in the packet, we assume those bytes are the
# destination being requested.
if len(data) >= RNS.Identity.TRUNCATED_HASHLENGTH//8:
# If there is also enough bytes for a trasport
# instance ID and at least one random byte, we
# assume the next bytes to be the trasport ID
# of the requesting transport instance.
if len(data) > (RNS.Identity.TRUNCATED_HASHLENGTH//8)*2:
requesting_transport_instance = data[RNS.Identity.TRUNCATED_HASHLENGTH//8:(RNS.Identity.TRUNCATED_HASHLENGTH//8)*2]
else:
requesting_transport_instance = None
Transport.path_request(
data[:RNS.Identity.TRUNCATED_HASHLENGTH//8],
Transport.from_local_client(packet),
packet.receiving_interface,
requesting_transport_instance,
)
except Exception as e:
RNS.log("Error while handling path request. The contained exception was: "+str(e), RNS.LOG_ERROR)
@staticmethod
def path_request(destination_hash, is_from_local_client, attached_interface):
RNS.log("Path request for "+RNS.prettyhexrep(destination_hash), RNS.LOG_DEBUG)
def path_request(destination_hash, is_from_local_client, attached_interface, requestor_transport_id=None):
if attached_interface != None:
interface_str = " on "+str(attached_interface)
else:
interface_str = ""
# TODO: Clean
# RNS.log("Path request for "+RNS.prettyhexrep(destination_hash)+interface_str, RNS.LOG_DEBUG)
destination_exists_on_local_client = False
if len(Transport.local_client_interfaces) > 0:
if destination_hash in Transport.destination_table:
destination_interface = Transport.destination_table[destination_hash][5]
if Transport.is_local_client_interface(destination_interface):
destination_exists_on_local_client = True
Transport.pending_local_path_requests[destination_hash] = attached_interface
local_destination = next((d for d in Transport.destinations if d.hash == destination_hash), None)
if local_destination != None:
RNS.log("Destination is local to this system, announcing", RNS.LOG_DEBUG)
local_destination.announce(path_response=True)
RNS.log("Answering path request for "+RNS.prettyhexrep(destination_hash)+interface_str+", destination is local to this system", RNS.LOG_DEBUG)
elif (RNS.Reticulum.transport_enabled() or is_from_local_client) and destination_hash in Transport.destination_table:
RNS.log("Path found, inserting announce for transmission", RNS.LOG_DEBUG)
elif (RNS.Reticulum.transport_enabled() or is_from_local_client) and (destination_hash in Transport.destination_table):
packet = Transport.destination_table[destination_hash][6]
next_hop = Transport.destination_table[destination_hash][1]
received_from = Transport.destination_table[destination_hash][5]
now = time.time()
retries = Transport.PATHFINDER_R
local_rebroadcasts = 0
block_rebroadcasts = True
announce_hops = packet.hops
if is_from_local_client:
retransmit_timeout = now
if requestor_transport_id != None and next_hop == requestor_transport_id:
# TODO: Find a bandwidth efficient way to invalidate our
# known path on this signal. The obvious way of signing
# path requests with transport instance keys is quite
# inefficient. There is probably a better way. Doing
# path invalidation here would decrease the network
# convergence time.
RNS.log("Not answering path request for "+RNS.prettyhexrep(destination_hash)+interface_str+", since next hop is the requestor", RNS.LOG_DEBUG)
else:
# TODO: Look at this timing
retransmit_timeout = now + Transport.PATH_REQUEST_GRACE # + (RNS.rand() * Transport.PATHFINDER_RW)
RNS.log("Answering path request for "+RNS.prettyhexrep(destination_hash)+interface_str+", path is known", RNS.LOG_DEBUG)
# This handles an edge case where a peer sends a past
# request for a destination just after an announce for
# said destination has arrived, but before it has been
# rebroadcast locally. In such a case the actual announce
# is temporarily held, and then reinserted when the path
# request has been served to the peer.
if packet.destination_hash in Transport.announce_table:
held_entry = Transport.announce_table[packet.destination_hash]
Transport.held_announces[packet.destination_hash] = held_entry
Transport.announce_table[packet.destination_hash] = [now, retransmit_timeout, retries, received_from, announce_hops, packet, local_rebroadcasts, block_rebroadcasts, attached_interface]
now = time.time()
retries = Transport.PATHFINDER_R
local_rebroadcasts = 0
block_rebroadcasts = True
announce_hops = packet.hops
if is_from_local_client:
retransmit_timeout = now
else:
# TODO: Look at this timing
retransmit_timeout = now + Transport.PATH_REQUEST_GRACE # + (RNS.rand() * Transport.PATHFINDER_RW)
# This handles an edge case where a peer sends a past
# request for a destination just after an announce for
# said destination has arrived, but before it has been
# rebroadcast locally. In such a case the actual announce
# is temporarily held, and then reinserted when the path
# request has been served to the peer.
if packet.destination_hash in Transport.announce_table:
held_entry = Transport.announce_table[packet.destination_hash]
Transport.held_announces[packet.destination_hash] = held_entry
Transport.announce_table[packet.destination_hash] = [now, retransmit_timeout, retries, received_from, announce_hops, packet, local_rebroadcasts, block_rebroadcasts, attached_interface]
elif is_from_local_client:
# Forward path request on all interfaces
# except the local client
RNS.log("Forwarding path request from local client for "+RNS.prettyhexrep(destination_hash)+interface_str+" to all other interfaces", RNS.LOG_DEBUG)
for interface in Transport.interfaces:
if not interface == attached_interface:
Transport.request_path_on_interface(destination_hash, interface)
Transport.request_path(destination_hash, interface)
elif not is_from_local_client and len(Transport.local_client_interfaces) > 0:
# Forward the path request on all local
# client interfaces
RNS.log("Forwarding path request for "+RNS.prettyhexrep(destination_hash)+interface_str+" to local clients", RNS.LOG_DEBUG)
for interface in Transport.local_client_interfaces:
Transport.request_path_on_interface(destination_hash, interface)
Transport.request_path(destination_hash, interface)
else:
RNS.log("No known path to requested destination, ignoring request", RNS.LOG_DEBUG)
RNS.log("Ignoring path request for "+RNS.prettyhexrep(destination_hash)+interface_str+", no path known", RNS.LOG_DEBUG)
@staticmethod
def from_local_client(packet):
@@ -1430,6 +1749,28 @@ class Transport:
for interface in Transport.local_client_interfaces:
interface.detach()
@staticmethod
def shared_connection_disappeared():
for link in Transport.active_links:
link.teardown()
for link in Transport.pending_links:
link.teardown()
Transport.announce_table = {}
Transport.destination_table = {}
Transport.reverse_table = {}
Transport.link_table = {}
Transport.held_announces = {}
Transport.announce_handlers = []
Transport.tunnels = {}
@staticmethod
def shared_connection_reappeared():
if Transport.owner.is_connected_to_shared_instance:
for registered_destination in Transport.destinations:
if registered_destination.type == RNS.Destination.SINGLE:
registered_destination.announce(path_response=True)
@staticmethod
def exit_handler():
+22
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import os
import glob
+124 -33
View File
@@ -1,5 +1,27 @@
#!/usr/bin/env python3
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import sys
import time
@@ -8,44 +30,81 @@ import argparse
from RNS._version import __version__
def program_setup(configdir, destination_hexhash, verbosity):
try:
dest_len = (RNS.Reticulum.TRUNCATED_HASHLENGTH//8)*2
if len(destination_hexhash) != dest_len:
raise ValueError("Destination length is invalid, must be {hex} hexadecimal characters ({byte} bytes).".format(hex=dest_len, byte=dest_len//2))
def program_setup(configdir, table, drop, destination_hexhash, verbosity, timeout):
if table:
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity)
table = sorted(reticulum.get_path_table(), key=lambda e: (e["interface"], e["hops"]) )
for path in table:
exp_str = RNS.timestamp_str(path["expires"])
if path["hops"] == 1:
m_str = " "
else:
m_str = "s"
print(RNS.prettyhexrep(path["hash"])+" is "+str(path["hops"])+" hop"+m_str+" away via "+RNS.prettyhexrep(path["via"])+" on "+path["interface"]+" expires "+RNS.timestamp_str(path["expires"]))
elif drop:
try:
destination_hash = bytes.fromhex(destination_hexhash)
dest_len = (RNS.Reticulum.TRUNCATED_HASHLENGTH//8)*2
if len(destination_hexhash) != dest_len:
raise ValueError("Destination length is invalid, must be {hex} hexadecimal characters ({byte} bytes).".format(hex=dest_len, byte=dest_len//2))
try:
destination_hash = bytes.fromhex(destination_hexhash)
except Exception as e:
raise ValueError("Invalid destination entered. Check your input.")
except Exception as e:
raise ValueError("Invalid destination entered. Check your input.")
except Exception as e:
print(str(e))
exit()
print(str(e))
exit()
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity)
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity)
if not RNS.Transport.has_path(destination_hash):
RNS.Transport.request_path(destination_hash)
print("Path to "+RNS.prettyhexrep(destination_hash)+" requested ", end=" ")
sys.stdout.flush()
if reticulum.drop_path(destination_hash):
print("Dropped path to "+RNS.prettyhexrep(destination_hash))
else:
print("Unable to drop path to "+RNS.prettyhexrep(destination_hash)+". Does it exist?")
i = 0
syms = "⢄⢂⢁⡁⡈⡐⡠"
while not RNS.Transport.has_path(destination_hash):
time.sleep(0.1)
print(("\b\b"+syms[i]+" "), end="")
sys.stdout.flush()
i = (i+1)%len(syms)
hops = RNS.Transport.hops_to(destination_hash)
next_hop = RNS.prettyhexrep(reticulum.get_next_hop(destination_hash))
next_hop_interface = reticulum.get_next_hop_if_name(destination_hash)
if hops != 1:
ms = "s"
else:
ms = ""
try:
dest_len = (RNS.Reticulum.TRUNCATED_HASHLENGTH//8)*2
if len(destination_hexhash) != dest_len:
raise ValueError("Destination length is invalid, must be {hex} hexadecimal characters ({byte} bytes).".format(hex=dest_len, byte=dest_len//2))
try:
destination_hash = bytes.fromhex(destination_hexhash)
except Exception as e:
raise ValueError("Invalid destination entered. Check your input.")
except Exception as e:
print(str(e))
exit()
print("\rPath found, destination "+RNS.prettyhexrep(destination_hash)+" is "+str(hops)+" hop"+ms+" away via "+next_hop+" on "+next_hop_interface)
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity)
if not RNS.Transport.has_path(destination_hash):
RNS.Transport.request_path(destination_hash)
print("Path to "+RNS.prettyhexrep(destination_hash)+" requested ", end=" ")
sys.stdout.flush()
i = 0
syms = "⢄⢂⢁⡁⡈⡐⡠"
limit = time.time()+timeout
while not RNS.Transport.has_path(destination_hash) and time.time()<limit:
time.sleep(0.1)
print(("\b\b"+syms[i]+" "), end="")
sys.stdout.flush()
i = (i+1)%len(syms)
if RNS.Transport.has_path(destination_hash):
hops = RNS.Transport.hops_to(destination_hash)
next_hop = RNS.prettyhexrep(reticulum.get_next_hop(destination_hash))
next_hop_interface = reticulum.get_next_hop_if_name(destination_hash)
if hops != 1:
ms = "s"
else:
ms = ""
print("\rPath found, destination "+RNS.prettyhexrep(destination_hash)+" is "+str(hops)+" hop"+ms+" away via "+next_hop+" on "+next_hop_interface)
else:
print("\r \rPath not found")
def main():
@@ -65,6 +124,31 @@ def main():
version="rnpath {version}".format(version=__version__)
)
parser.add_argument(
"-t",
"--table",
action="store_true",
help="show all known paths",
default=False
)
parser.add_argument(
"-d",
"--drop",
action="store_true",
help="remove the path to a destination",
default=False
)
parser.add_argument(
"-w",
action="store",
metavar="seconds",
type=float,
help="timeout before giving up",
default=15
)
parser.add_argument(
"destination",
nargs="?",
@@ -82,12 +166,19 @@ def main():
else:
configarg = None
if not args.destination:
if not args.table and not args.destination:
print("")
parser.print_help()
print("")
else:
program_setup(configdir = configarg, destination_hexhash = args.destination, verbosity = args.verbose)
program_setup(
configdir = configarg,
table = args.table,
drop = args.drop,
destination_hexhash = args.destination,
verbosity = args.verbose,
timeout = args.w,
)
except KeyboardInterrupt:
print("")
+43 -1
View File
@@ -1,5 +1,27 @@
#!/usr/bin/env python3
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import os
import sys
@@ -101,11 +123,31 @@ def program_setup(configdir, destination_hexhash, size=DEFAULT_PROBE_SIZE, full_
rtt = round(rtt*1000, 3)
rttstring = str(rtt)+" milliseconds"
reception_stats = ""
if reticulum.is_connected_to_shared_instance:
reception_rssi = reticulum.get_packet_rssi(receipt.proof_packet.packet_hash)
reception_snr = reticulum.get_packet_snr(receipt.proof_packet.packet_hash)
if reception_rssi != None:
reception_stats += " [RSSI "+str(reception_rssi)+" dBm]"
if reception_snr != None:
reception_stats += " [SNR "+str(reception_snr)+" dB]"
else:
if receipt.proof_packet != None:
if receipt.proof_packet.rssi != None:
reception_stats += " [RSSI "+str(receipt.proof_packet.rssi)+" dBm]"
if receipt.proof_packet.snr != None:
reception_stats += " [SNR "+str(receipt.proof_packet.snr)+" dB]"
print(
"Valid reply received from "+
RNS.prettyhexrep(receipt.destination.hash)+
"\nRound-trip time is "+rttstring+
" over "+str(hops)+" hop"+ms
" over "+str(hops)+" hop"+ms+
reception_stats
)
Regular → Executable
+381 -5
View File
@@ -1,16 +1,46 @@
#!/usr/bin/env python3
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import argparse
import time
from RNS._version import __version__
def program_setup(configdir, verbosity = 0, quietness = 0):
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity-quietness)
def program_setup(configdir, verbosity = 0, quietness = 0, service = False):
targetloglevel = 3+verbosity-quietness
if service:
RNS.logdest = RNS.LOG_FILE
RNS.logfile = RNS.Reticulum.configdir+"/logfile"
targetloglevel = None
reticulum = RNS.Reticulum(configdir=configdir, loglevel=targetloglevel)
RNS.log("Started rnsd version {version}".format(version=__version__), RNS.LOG_NOTICE)
while True:
input()
time.sleep(1)
def main():
try:
@@ -18,20 +48,366 @@ def main():
parser.add_argument("--config", action="store", default=None, help="path to alternative Reticulum config directory", type=str)
parser.add_argument('-v', '--verbose', action='count', default=0)
parser.add_argument('-q', '--quiet', action='count', default=0)
parser.add_argument('-s', '--service', action='store_true', default=False, help="rnsd is running as a service and should log to file")
parser.add_argument("--exampleconfig", action='store_true', default=False, help="print verbose configuration example to stdout and exit")
parser.add_argument("--version", action="version", version="rnsd {version}".format(version=__version__))
args = parser.parse_args()
if args.exampleconfig:
print(__example_rns_config__)
exit()
if args.config:
configarg = args.config
else:
configarg = None
program_setup(configdir = configarg, verbosity=args.verbose, quietness=args.quiet)
program_setup(configdir = configarg, verbosity=args.verbose, quietness=args.quiet, service=args.service)
except KeyboardInterrupt:
print("")
exit()
__example_rns_config__ = '''# This is an example Reticulum config file.
# You should probably edit it to include any additional,
# interfaces and settings you might need.
[reticulum]
# If you enable Transport, your system will route traffic
# for other peers, pass announces and serve path requests.
# This should be done for systems that are suited to act
# as transport nodes, ie. if they are stationary and
# always-on. This directive is optional and can be removed
# for brevity.
enable_transport = False
# By default, the first program to launch the Reticulum
# Network Stack will create a shared instance, that other
# programs can communicate with. Only the shared instance
# opens all the configured interfaces directly, and other
# local programs communicate with the shared instance over
# a local socket. This is completely transparent to the
# user, and should generally be turned on. This directive
# is optional and can be removed for brevity.
share_instance = Yes
# If you want to run multiple *different* shared instances
# on the same system, you will need to specify different
# shared instance ports for each. The defaults are given
# below, and again, these options can be left out if you
# don't need them.
shared_instance_port = 37428
instance_control_port = 37429
# You can configure Reticulum to panic and forcibly close
# if an unrecoverable interface error occurs, such as the
# hardware device for an interface disappearing. This is
# an optional directive, and can be left out for brevity.
# This behaviour is disabled by default.
panic_on_interface_error = No
[logging]
# Valid log levels are 0 through 7:
# 0: Log only critical information
# 1: Log errors and lower log levels
# 2: Log warnings and lower log levels
# 3: Log notices and lower log levels
# 4: Log info and lower (this is the default)
# 5: Verbose logging
# 6: Debug logging
# 7: Extreme logging
loglevel = 4
# The interfaces section defines the physical and virtual
# interfaces Reticulum will use to communicate on. This
# section will contain examples for a variety of interface
# types. You can modify these or use them as a basis for
# your own config, or simply remove the unused ones.
[interfaces]
# This interface enables communication with other
# link-local Reticulum nodes over UDP. It does not
# need any functional IP infrastructure like routers
# or DHCP servers, but will require that at least link-
# local IPv6 is enabled in your operating system, which
# should be enabled by default in almost any OS. See
# the Reticulum Manual for more configuration options.
[[Default Interface]]
type = AutoInterface
interface_enabled = True
# The following example enables communication with other
# local Reticulum peers using UDP broadcasts.
[[UDP Interface]]
type = UDPInterface
interface_enabled = False
listen_ip = 0.0.0.0
listen_port = 4242
forward_ip = 255.255.255.255
forward_port = 4242
# The above configuration will allow communication
# within the local broadcast domains of all local
# IP interfaces.
# Instead of specifying listen_ip, listen_port,
# forward_ip and forward_port, you can also bind
# to a specific network device like below.
# device = eth0
# port = 4242
# Assuming the eth0 device has the address
# 10.55.0.72/24, the above configuration would
# be equivalent to the following manual setup.
# Note that we are both listening and forwarding to
# the broadcast address of the network segments.
# listen_ip = 10.55.0.255
# listen_port = 4242
# forward_ip = 10.55.0.255
# forward_port = 4242
# You can of course also communicate only with
# a single IP address
# listen_ip = 10.55.0.15
# listen_port = 4242
# forward_ip = 10.55.0.16
# forward_port = 4242
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[TCP Server Interface]]
type = TCPServerInterface
interface_enabled = False
# This configuration will listen on all IP
# interfaces on port 4242
listen_ip = 0.0.0.0
listen_port = 4242
# Alternatively you can bind to a specific IP
# listen_ip = 10.0.0.88
# listen_port = 4242
# Or a specific network device
# device = eth0
# port = 4242
# To connect to a TCP server interface, you would
# naturally use the TCP client interface. Here's
# an example. The target_host can either be an IP
# address or a hostname
[[TCP Client Interface]]
type = TCPClientInterface
interface_enabled = False
target_host = 127.0.0.1
target_port = 4242
# This example shows how to make your Reticulum
# instance available over I2P, and connect to
# another I2P peer. Please be aware that you
# must have an I2P router running on your system
# with the SAMv3 API enabled for this to work.
[[I2P]]
type = I2PInterface
interface_enabled = yes
connectable = yes
peers = 5urvjicpzi7q3ybztsef4i5ow2aq4soktfj7zedz53s47r54jnqq.b32.i2p
# Here's an example of how to add a LoRa interface
# using the RNode LoRa transceiver.
[[RNode LoRa Interface]]
type = RNodeInterface
# Enable interface if you want use it!
interface_enabled = False
# Serial port for the device
port = /dev/ttyUSB0
# Set frequency to 867.2 MHz
frequency = 867200000
# Set LoRa bandwidth to 125 KHz
bandwidth = 125000
# Set TX power to 7 dBm (5 mW)
txpower = 7
# Select spreading factor 8. Valid
# range is 7 through 12, with 7
# being the fastest and 12 having
# the longest range.
spreadingfactor = 8
# Select coding rate 5. Valid range
# is 5 throough 8, with 5 being the
# fastest, and 8 the longest range.
codingrate = 5
# You can configure the RNode to send
# out identification on the channel with
# a set interval by configuring the
# following two parameters. The trans-
# ceiver will only ID if the set
# interval has elapsed since it's last
# actual transmission. The interval is
# configured in seconds.
# This option is commented out and not
# used by default.
# id_callsign = MYCALL-0
# id_interval = 600
# For certain homebrew RNode interfaces
# with low amounts of RAM, using packet
# flow control can be useful. By default
# it is disabled.
flow_control = False
# An example KISS modem interface. Useful for running
# Reticulum over packet radio hardware.
[[Packet Radio KISS Interface]]
type = KISSInterface
# Enable interface if you want use it!
interface_enabled = False
# Serial port for the device
port = /dev/ttyUSB1
# Set the serial baud-rate and other
# configuration parameters.
speed = 115200
databits = 8
parity = none
stopbits = 1
# Set the modem preamble. A 150ms
# preamble should be a reasonable
# default, but may need to be
# increased for radios with slow-
# opening squelch and long TX/RX
# turnaround
preamble = 150
# Set the modem TX tail. In most
# cases this should be kept as low
# as possible to not waste airtime.
txtail = 10
# Configure CDMA parameters. These
# settings are reasonable defaults.
persistence = 200
slottime = 20
# You can configure the interface to send
# out identification on the channel with
# a set interval by configuring the
# following two parameters. The KISS
# interface will only ID if the set
# interval has elapsed since it's last
# actual transmission. The interval is
# configured in seconds.
# This option is commented out and not
# used by default.
# id_callsign = MYCALL-0
# id_interval = 600
# Whether to use KISS flow-control.
# This is useful for modems that have
# a small internal packet buffer, but
# support packet flow control instead.
flow_control = false
# If you're using Reticulum on amateur radio spectrum,
# you might want to use the AX.25 KISS interface. This
# way, Reticulum will automatically encapsulate it's
# traffic in AX.25 and also identify your stations
# transmissions with your callsign and SSID.
#
# Only do this if you really need to! Reticulum doesn't
# need the AX.25 layer for anything, and it incurs extra
# overhead on every packet to encapsulate in AX.25.
#
# A more efficient way is to use the plain KISS interface
# with the beaconing functionality described above.
[[Packet Radio AX.25 KISS Interface]]
type = AX25KISSInterface
# Set the station callsign and SSID
callsign = NO1CLL
ssid = 0
# Enable interface if you want use it!
interface_enabled = False
# Serial port for the device
port = /dev/ttyUSB2
# Set the serial baud-rate and other
# configuration parameters.
speed = 115200
databits = 8
parity = none
stopbits = 1
# Whether to use KISS flow-control.
# This is useful for modems with a
# small internal packet buffer.
flow_control = false
# Set the modem preamble. A 150ms
# preamble should be a reasonable
# default, but may need to be
# increased for radios with slow-
# opening squelch and long TX/RX
# turnaround
preamble = 150
# Set the modem TX tail. In most
# cases this should be kept as low
# as possible to not waste airtime.
txtail = 10
# Configure CDMA parameters. These
# settings are reasonable defaults.
persistence = 200
slottime = 20
'''
if __name__ == "__main__":
main()
main()
+102 -10
View File
@@ -1,5 +1,27 @@
#!/usr/bin/env python3
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import RNS
import argparse
@@ -26,34 +48,88 @@ def size_str(num, suffix='B'):
def program_setup(configdir, dispall=False, verbosity = 0):
reticulum = RNS.Reticulum(configdir = configdir, loglevel = 3+verbosity)
ifstats = reticulum.get_interface_stats()
if ifstats != None:
for ifstat in ifstats:
stats = None
try:
stats = reticulum.get_interface_stats()
except Exception as e:
pass
if stats != None:
for ifstat in stats["interfaces"]:
name = ifstat["name"]
print("")
if dispall or not (name.startswith("LocalInterface[") or name.startswith("TCPInterface[Client")):
print("")
if ifstat["status"]:
ss = "Up"
else:
ss = "Down"
if ifstat["mode"] == RNS.Interfaces.Interface.Interface.MODE_ACCESS_POINT:
modestr = "Access Point"
elif ifstat["mode"] == RNS.Interfaces.Interface.Interface.MODE_POINT_TO_POINT:
modestr = "Point-to-Point"
else:
modestr = "Full"
if ifstat["clients"] != None:
clients = ifstat["clients"]
if name.startswith("Shared Instance["):
clients_string = "Connected applications: "+str(clients-1)
cnum = max(clients-1,0)
if cnum == 1:
spec_str = " program"
else:
spec_str = " programs"
clients_string = "Serving : "+str(cnum)+spec_str
else:
clients_string = "Connected clients: "+str(clients)
clients_string = "Clients : "+str(clients)
else:
clients = None
print(" {n}".format(n=ifstat["name"]))
print("\tStatus: {ss}".format(ss=ss))
if "ifac_netname" in ifstat and ifstat["ifac_netname"] != None:
print(" Network : {nn}".format(nn=ifstat["ifac_netname"]))
print(" Status : {ss}".format(ss=ss))
if clients != None:
print("\t"+clients_string)
print("\tRX: {rxb}\n\tTX: {txb}".format(rxb=size_str(ifstat["rxb"]), txb=size_str(ifstat["txb"])))
print(" "+clients_string)
if not (name.startswith("Shared Instance[") or name.startswith("TCPInterface[Client") or name.startswith("LocalInterface[")):
print(" Mode : {mode}".format(mode=modestr))
if "bitrate" in ifstat and ifstat["bitrate"] != None:
print(" Rate : {ss}".format(ss=speed_str(ifstat["bitrate"])))
if "peers" in ifstat and ifstat["peers"] != None:
print(" Peers : {np} reachable".format(np=ifstat["peers"]))
if "ifac_signature" in ifstat and ifstat["ifac_signature"] != None:
sigstr = "<…"+RNS.hexrep(ifstat["ifac_signature"][-5:], delimit=False)+">"
print(" Access : {nb}-bit IFAC by {sig}".format(nb=ifstat["ifac_size"]*8, sig=sigstr))
if "i2p_b32" in ifstat and ifstat["i2p_b32"] != None:
print(" I2P B32 : {ep}".format(ep=str(ifstat["i2p_b32"])))
if "announce_queue" in ifstat and ifstat["announce_queue"] != None and ifstat["announce_queue"] > 0:
aqn = ifstat["announce_queue"]
if aqn == 1:
print(" Queued : {np} announce".format(np=aqn))
else:
print(" Queued : {np} announces".format(np=aqn))
print(" Traffic : {txb}\n {rxb}".format(rxb=size_str(ifstat["rxb"]), txb=size_str(ifstat["txb"])))
if "transport_id" in stats and stats["transport_id"] != None:
print("\n Reticulum Transport Instance "+RNS.prettyhexrep(stats["transport_id"])+" running")
print("")
else:
print("Could not get RNS status")
@@ -87,5 +163,21 @@ def main():
print("")
exit()
def speed_str(num, suffix='bps'):
units = ['','k','M','G','T','P','E','Z']
last_unit = 'Y'
if suffix == 'Bps':
num /= 8
units = ['','K','M','G','T','P','E','Z']
last_unit = 'Y'
for unit in units:
if abs(num) < 1000.0:
return "%3.2f %s%s" % (num, unit, suffix)
num /= 1000.0
return "%.2f %s%s" % (num, last_unit, suffix)
if __name__ == "__main__":
main()
+49 -4
View File
@@ -1,3 +1,25 @@
# MIT License
#
# Copyright (c) 2016-2022 Mark Qvist / unsigned.io
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
import os
import sys
import glob
@@ -31,12 +53,15 @@ LOG_EXTREME = 7
LOG_STDOUT = 0x91
LOG_FILE = 0x92
LOG_MAXSIZE = 5*1024*1024
loglevel = LOG_NOTICE
logfile = None
logdest = LOG_STDOUT
logtimefmt = "%Y-%m-%d %H:%M:%S"
random.seed(os.urandom(10))
instance_random = random.Random()
instance_random.seed(os.urandom(10))
_always_override_destination = False
@@ -65,12 +90,19 @@ def loglevelname(level):
def version():
return __version__
def host_os():
from .vendor.platformutils import get_platform
return get_platform()
def timestamp_str(time_s):
timestamp = time.localtime(time_s)
return time.strftime(logtimefmt, timestamp)
def log(msg, level=3, _override_destination = False):
global _always_override_destination
if loglevel >= level:
timestamp = time.time()
logstring = "["+time.strftime(logtimefmt)+"] ["+loglevelname(level)+"] "+msg
logstring = "["+timestamp_str(time.time())+"] ["+loglevelname(level)+"] "+msg
logging_lock.acquire()
if (logdest == LOG_STDOUT or _always_override_destination or _override_destination):
@@ -82,6 +114,13 @@ def log(msg, level=3, _override_destination = False):
file = open(logfile, "a")
file.write(logstring+"\n")
file.close()
if os.path.getsize(logfile) > LOG_MAXSIZE:
prevfile = logfile+".1"
if os.path.isfile(prevfile):
os.unlink(prevfile)
os.rename(logfile, prevfile)
logging_lock.release()
except Exception as e:
logging_lock.release()
@@ -92,10 +131,15 @@ def log(msg, level=3, _override_destination = False):
def rand():
result = random.random()
result = instance_random.random()
return result
def hexrep(data, delimit=True):
try:
iter(data)
except TypeError:
data = [data]
delimiter = ":"
if not delimit:
delimiter = ""
@@ -111,4 +155,5 @@ def panic():
os._exit(255)
def exit():
print("")
sys.exit(0)
+1 -1
View File
@@ -1 +1 @@
__version__ = "0.2.6"
__version__ = "0.3.5"
+20
View File
@@ -0,0 +1,20 @@
Copyright (c) 2018 Viktor Villainov
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+25
View File
@@ -0,0 +1,25 @@
"""
A modern asynchronous library for building I2P applications.
"""
from .__version__ import (
__title__, __description__, __url__, __version__,
__author__, __author_email__, __license__, __copyright__
)
from .sam import Destination, PrivateKey
from .aiosam import (
get_sam_socket, dest_lookup, new_destination,
create_session, stream_connect, stream_accept,
Session, StreamConnection, StreamAcceptor
)
from .tunnel import ClientTunnel, ServerTunnel
from .utils import get_sam_address
from .exceptions import (
CantReachPeer, DuplicatedDest, DuplicatedId, I2PError,
InvalidId, InvalidKey, KeyNotFound, PeerNotFound, Timeout,
)
+8
View File
@@ -0,0 +1,8 @@
__title__ = 'i2plib'
__description__ = 'A modern asynchronous library for building I2P applications.'
__url__ = 'https://github.com/l-n-s/i2plib'
__version__ = '0.0.14'
__author__ = 'Viktor Villainov'
__author_email__ = 'supervillain@riseup.net'
__license__ = 'MIT'
__copyright__ = 'Copyright 2018 Viktor Villainov'
+258
View File
@@ -0,0 +1,258 @@
import asyncio
from . import sam
from . import exceptions
from . import utils
from .log import logger
def parse_reply(data):
if not data:
raise ConnectionAbortedError("Empty response: SAM API went offline")
try:
msg = sam.Message(data.decode().strip())
logger.debug("SAM reply: "+str(msg))
except:
raise ConnectionAbortedError("Invalid SAM response")
return msg
async def get_sam_socket(sam_address=sam.DEFAULT_ADDRESS, loop=None):
"""A couroutine used to create a new SAM socket.
:param sam_address: (optional) SAM API address
:param loop: (optional) event loop instance
:return: A (reader, writer) pair
"""
reader, writer = await asyncio.open_connection(*sam_address, loop=loop)
writer.write(sam.hello("3.1", "3.1"))
reply = parse_reply(await reader.readline())
if reply.ok:
return (reader, writer)
else:
writer.close()
raise exceptions.SAM_EXCEPTIONS[reply["RESULT"]]()
async def dest_lookup(domain, sam_address=sam.DEFAULT_ADDRESS,
loop=None):
"""A coroutine used to lookup a full I2P destination by .i2p domain or
.b32.i2p address.
:param domain: Address to be resolved, can be a .i2p domain or a .b32.i2p
address.
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:return: An instance of :class:`Destination`
"""
reader, writer = await get_sam_socket(sam_address, loop)
writer.write(sam.naming_lookup(domain))
reply = parse_reply(await reader.readline())
writer.close()
if reply.ok:
return sam.Destination(reply["VALUE"])
else:
raise exceptions.SAM_EXCEPTIONS[reply["RESULT"]]()
async def new_destination(sam_address=sam.DEFAULT_ADDRESS, loop=None,
sig_type=sam.Destination.default_sig_type):
"""A coroutine used to generate a new destination with a private key of a
chosen signature type.
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:param sig_type: (optional) Signature type
:return: An instance of :class:`Destination`
"""
reader, writer = await get_sam_socket(sam_address, loop)
writer.write(sam.dest_generate(sig_type))
reply = parse_reply(await reader.readline())
writer.close()
return sam.Destination(reply["PRIV"], has_private_key=True)
async def create_session(session_name, sam_address=sam.DEFAULT_ADDRESS,
loop=None, style="STREAM",
signature_type=sam.Destination.default_sig_type,
destination=None, options={}):
"""A coroutine used to create a new SAM session.
:param session_name: Session nick name
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:param style: (optional) Session style, can be STREAM, DATAGRAM, RAW
:param signature_type: (optional) If the destination is TRANSIENT, this
signature type is used
:param destination: (optional) Destination to use in this session. Can be
a base64 encoded string, :class:`Destination`
instance or None. TRANSIENT destination is used when it
is None.
:param options: (optional) A dict object with i2cp options
:return: A (reader, writer) pair
"""
logger.debug("Creating session {}".format(session_name))
if destination:
if type(destination) == sam.Destination:
destination = destination
else:
destination = sam.Destination(
destination, has_private_key=True)
dest_string = destination.private_key.base64
else:
dest_string = sam.TRANSIENT_DESTINATION
options = " ".join(["{}={}".format(k, v) for k, v in options.items()])
reader, writer = await get_sam_socket(sam_address, loop)
writer.write(sam.session_create(
style, session_name, dest_string, options))
reply = parse_reply(await reader.readline())
if reply.ok:
if not destination:
destination = sam.Destination(
reply["DESTINATION"], has_private_key=True)
logger.debug(destination.base32)
logger.debug("Session created {}".format(session_name))
return (reader, writer)
else:
writer.close()
raise exceptions.SAM_EXCEPTIONS[reply["RESULT"]]()
async def stream_connect(session_name, destination,
sam_address=sam.DEFAULT_ADDRESS, loop=None):
"""A coroutine used to connect to a remote I2P destination.
:param session_name: Session nick name
:param destination: I2P destination to connect to
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:return: A (reader, writer) pair
"""
logger.debug("Connecting stream {}".format(session_name))
if isinstance(destination, str) and not destination.endswith(".i2p"):
destination = sam.Destination(destination)
elif isinstance(destination, str):
destination = await dest_lookup(destination, sam_address, loop)
reader, writer = await get_sam_socket(sam_address, loop)
writer.write(sam.stream_connect(session_name, destination.base64,
silent="false"))
reply = parse_reply(await reader.readline())
if reply.ok:
logger.debug("Stream connected {}".format(session_name))
return (reader, writer)
else:
writer.close()
raise exceptions.SAM_EXCEPTIONS[reply["RESULT"]]()
async def stream_accept(session_name, sam_address=sam.DEFAULT_ADDRESS,
loop=None):
"""A coroutine used to accept a connection from the I2P network.
:param session_name: Session nick name
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:return: A (reader, writer) pair
"""
reader, writer = await get_sam_socket(sam_address, loop)
writer.write(sam.stream_accept(session_name, silent="false"))
reply = parse_reply(await reader.readline())
if reply.ok:
return (reader, writer)
else:
writer.close()
raise exceptions.SAM_EXCEPTIONS[reply["RESULT"]]()
### Context managers
class Session:
"""Async SAM session context manager.
:param session_name: Session nick name
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:param style: (optional) Session style, can be STREAM, DATAGRAM, RAW
:param signature_type: (optional) If the destination is TRANSIENT, this
signature type is used
:param destination: (optional) Destination to use in this session. Can be
a base64 encoded string, :class:`Destination`
instance or None. TRANSIENT destination is used when it
is None.
:param options: (optional) A dict object with i2cp options
:return: :class:`Session` object
"""
def __init__(self, session_name, sam_address=sam.DEFAULT_ADDRESS,
loop=None, style="STREAM",
signature_type=sam.Destination.default_sig_type,
destination=None, options={}):
self.session_name = session_name
self.sam_address = sam_address
self.loop = loop
self.style = style
self.signature_type = signature_type
self.destination = destination
self.options = options
async def __aenter__(self):
self.reader, self.writer = await create_session(self.session_name,
sam_address=self.sam_address, loop=self.loop, style=self.style,
signature_type=self.signature_type,
destination=self.destination, options=self.options)
return self
async def __aexit__(self, exc_type, exc, tb):
### TODO handle exceptions
self.writer.close()
class StreamConnection:
"""Async stream connection context manager.
:param session_name: Session nick name
:param destination: I2P destination to connect to
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:return: :class:`StreamConnection` object
"""
def __init__(self, session_name, destination,
sam_address=sam.DEFAULT_ADDRESS, loop=None):
self.session_name = session_name
self.sam_address = sam_address
self.loop = loop
self.destination = destination
async def __aenter__(self):
self.reader, self.writer = await stream_connect(self.session_name,
self.destination, sam_address=self.sam_address, loop=self.loop)
self.read = self.reader.read
self.write = self.writer.write
return self
async def __aexit__(self, exc_type, exc, tb):
### TODO handle exceptions
self.writer.close()
class StreamAcceptor:
"""Async stream acceptor context manager.
:param session_name: Session nick name
:param sam_address: (optional) SAM API address
:param loop: (optional) Event loop instance
:return: :class:`StreamAcceptor` object
"""
def __init__(self, session_name, sam_address=sam.DEFAULT_ADDRESS,
loop=None):
self.session_name = session_name
self.sam_address = sam_address
self.loop = loop
async def __aenter__(self):
self.reader, self.writer = await stream_accept(self.session_name,
sam_address=self.sam_address, loop=self.loop)
self.read = self.reader.read
self.write = self.writer.write
return self
async def __aexit__(self, exc_type, exc, tb):
### TODO handle exceptions
self.writer.close()
+44
View File
@@ -0,0 +1,44 @@
# SAM exceptions
class SAMException(IOError):
"""Base class for SAM exceptions"""
class CantReachPeer(SAMException):
"""The peer exists, but cannot be reached"""
class DuplicatedDest(SAMException):
"""The specified Destination is already in use"""
class DuplicatedId(SAMException):
"""The nickname is already associated with a session"""
class I2PError(SAMException):
"""A generic I2P error"""
class InvalidId(SAMException):
"""STREAM SESSION ID doesn't exist"""
class InvalidKey(SAMException):
"""The specified key is not valid (bad format, etc.)"""
class KeyNotFound(SAMException):
"""The naming system can't resolve the given name"""
class PeerNotFound(SAMException):
"""The peer cannot be found on the network"""
class Timeout(SAMException):
"""The peer cannot be found on the network"""
SAM_EXCEPTIONS = {
"CANT_REACH_PEER": CantReachPeer,
"DUPLICATED_DEST": DuplicatedDest,
"DUPLICATED_ID": DuplicatedId,
"I2P_ERROR": I2PError,
"INVALID_ID": InvalidId,
"INVALID_KEY": InvalidKey,
"KEY_NOT_FOUND": KeyNotFound,
"PEER_NOT_FOUND": PeerNotFound,
"TIMEOUT": Timeout,
}
+5
View File
@@ -0,0 +1,5 @@
"""Logging configuration."""
import logging
# Name the logger after the package.
logger = logging.getLogger(__package__)
+147
View File
@@ -0,0 +1,147 @@
from base64 import b64decode, b64encode, b32encode
from hashlib import sha256
import struct
import re
I2P_B64_CHARS = "-~"
def i2p_b64encode(x):
"""Encode I2P destination"""
return b64encode(x, altchars=I2P_B64_CHARS.encode()).decode()
def i2p_b64decode(x):
"""Decode I2P destination"""
return b64decode(x, altchars=I2P_B64_CHARS, validate=True)
SAM_BUFSIZE = 4096
DEFAULT_ADDRESS = ("127.0.0.1", 7656)
DEFAULT_MIN_VER = "3.1"
DEFAULT_MAX_VER = "3.1"
TRANSIENT_DESTINATION = "TRANSIENT"
VALID_BASE32_ADDRESS = re.compile(r"^([a-zA-Z0-9]{52}).b32.i2p$")
VALID_BASE64_ADDRESS = re.compile(r"^([a-zA-Z0-9-~=]{516,528})$")
class Message(object):
"""Parse SAM message to an object"""
def __init__(self, s):
self.opts = {}
if type(s) != str:
self._reply_string = s.decode().strip()
else:
self._reply_string = s
self.cmd, self.action, opts = self._reply_string.split(" ", 2)
for v in opts.split(" "):
data = v.split("=", 1) if "=" in v else (v, True)
self.opts[data[0]] = data[1]
def __getitem__(self, key):
return self.opts[key]
@property
def ok(self):
return self["RESULT"] == "OK"
def __repr__(self):
return self._reply_string
# SAM request messages
def hello(min_version, max_version):
return "HELLO VERSION MIN={} MAX={}\n".format(min_version,
max_version).encode()
def session_create(style, session_id, destination, options=""):
return "SESSION CREATE STYLE={} ID={} DESTINATION={} {}\n".format(
style, session_id, destination, options).encode()
def stream_connect(session_id, destination, silent="false"):
return "STREAM CONNECT ID={} DESTINATION={} SILENT={}\n".format(
session_id, destination, silent).encode()
def stream_accept(session_id, silent="false"):
return "STREAM ACCEPT ID={} SILENT={}\n".format(session_id, silent).encode()
def stream_forward(session_id, port, options=""):
return "STREAM FORWARD ID={} PORT={} {}\n".format(
session_id, port, options).encode()
def naming_lookup(name):
return "NAMING LOOKUP NAME={}\n".format(name).encode()
def dest_generate(signature_type):
return "DEST GENERATE SIGNATURE_TYPE={}\n".format(signature_type).encode()
class Destination(object):
"""I2P destination
https://geti2p.net/spec/common-structures#destination
:param data: (optional) Base64 encoded data or binary data
:param path: (optional) A path to a file with binary data
:param has_private_key: (optional) Does data have a private key?
"""
ECDSA_SHA256_P256 = 1
ECDSA_SHA384_P384 = 2
ECDSA_SHA512_P521 = 3
EdDSA_SHA512_Ed25519 = 7
default_sig_type = EdDSA_SHA512_Ed25519
_pubkey_size = 256
_signkey_size = 128
_min_cert_size = 3
def __init__(self, data=None, path=None, has_private_key=False):
#: Binary destination
self.data = bytes()
#: Base64 encoded destination
self.base64 = ""
#: :class:`RNS.vendor.i2plib.PrivateKey` instance or None
self.private_key = None
if path:
with open(path, "rb") as f: data = f.read()
if data and has_private_key:
self.private_key = PrivateKey(data)
cert_len = struct.unpack("!H", self.private_key.data[385:387])[0]
data = self.private_key.data[:387+cert_len]
if not data:
raise Exception("Can't create a destination with no data")
self.data = data if type(data) == bytes else i2p_b64decode(data)
self.base64 = data if type(data) == str else i2p_b64encode(data)
def __repr__(self):
return "<Destination: {}>".format(self.base32)
@property
def base32(self):
"""Base32 destination hash of this destination"""
desthash = sha256(self.data).digest()
return b32encode(desthash).decode()[:52].lower()
class PrivateKey(object):
"""I2P private key
https://geti2p.net/spec/common-structures#keysandcert
:param data: Base64 encoded data or binary data
"""
def __init__(self, data):
#: Binary private key
self.data = data if type(data) == bytes else i2p_b64decode(data)
#: Base64 encoded private key
self.base64 = data if type(data) == str else i2p_b64encode(data)
+202
View File
@@ -0,0 +1,202 @@
import logging
import asyncio
import argparse
from . import sam
from . import aiosam
from . import utils
from .log import logger
BUFFER_SIZE = 65536
async def proxy_data(reader, writer):
"""Proxy data from reader to writer"""
try:
while True:
data = await reader.read(BUFFER_SIZE)
if not data:
break
writer.write(data)
except Exception as e:
logger.debug('proxy_data_task exception {}'.format(e))
finally:
try:
writer.close()
except RuntimeError:
pass
logger.debug('close connection')
class I2PTunnel(object):
"""Base I2P Tunnel object, not to be used directly
:param local_address: A local address to use for a tunnel.
E.g. ("127.0.0.1", 6668)
:param destination: (optional) Destination to use for this tunnel. Can be
a base64 encoded string, :class:`Destination`
instance or None. A new destination is created when it
is None.
:param session_name: (optional) Session nick name. A new session nickname is
generated if not specified.
:param options: (optional) A dict object with i2cp options
:param loop: (optional) Event loop instance
:param sam_address: (optional) SAM API address
"""
def __init__(self, local_address, destination=None, session_name=None,
options={}, loop=None, sam_address=sam.DEFAULT_ADDRESS):
self.local_address = local_address
self.destination = destination
self.session_name = session_name or utils.generate_session_id()
self.options = options
self.loop = loop
self.sam_address = sam_address
async def _pre_run(self):
if not self.destination:
self.destination = await aiosam.new_destination(
sam_address=self.sam_address, loop=self.loop)
_, self.session_writer = await aiosam.create_session(
self.session_name, style=self.style, options=self.options,
sam_address=self.sam_address,
loop=self.loop, destination=self.destination)
def stop(self):
"""Stop the tunnel"""
self.session_writer.close()
class ClientTunnel(I2PTunnel):
"""Client tunnel, a subclass of tunnel.I2PTunnel
If you run a client tunnel with a local address ("127.0.0.1", 6668) and
a remote destination "irc.echelon.i2p", all connections to 127.0.0.1:6668
will be proxied to irc.echelon.i2p.
:param remote_destination: Remote I2P destination, can be either .i2p
domain, .b32.i2p address, base64 destination or
:class:`Destination` instance
"""
def __init__(self, remote_destination, *args, **kwargs):
super().__init__(*args, **kwargs)
self.style = "STREAM"
self.remote_destination = remote_destination
async def run(self):
"""A coroutine used to run the tunnel"""
await self._pre_run()
async def handle_client(client_reader, client_writer):
"""Handle local client connection"""
remote_reader, remote_writer = await aiosam.stream_connect(
self.session_name, self.remote_destination,
sam_address=self.sam_address, loop=self.loop)
asyncio.ensure_future(proxy_data(remote_reader, client_writer),
loop=self.loop)
asyncio.ensure_future(proxy_data(client_reader, remote_writer),
loop=self.loop)
self.server = await asyncio.start_server(handle_client, *self.local_address, loop=self.loop)
def stop(self):
super().stop()
self.server.close()
class ServerTunnel(I2PTunnel):
"""Server tunnel, a subclass of tunnel.I2PTunnel
If you want to expose a local service 127.0.0.1:80 to the I2P network, run
a server tunnel with a local address ("127.0.0.1", 80). If you don't
provide a private key or a session name, it will use a TRANSIENT
destination.
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.style = "STREAM"
async def run(self):
"""A coroutine used to run the tunnel"""
await self._pre_run()
async def handle_client(incoming, client_reader, client_writer):
# data and dest may come in one chunk
dest, data = incoming.split(b"\n", 1)
remote_destination = sam.Destination(dest.decode())
logger.debug("{} client connected: {}.b32.i2p".format(
self.session_name, remote_destination.base32))
try:
remote_reader, remote_writer = await asyncio.wait_for(
asyncio.open_connection(
host=self.local_address[0],
port=self.local_address[1], loop=self.loop),
timeout=5, loop=self.loop)
if data: remote_writer.write(data)
asyncio.ensure_future(proxy_data(remote_reader, client_writer),
loop=self.loop)
asyncio.ensure_future(proxy_data(client_reader, remote_writer),
loop=self.loop)
except ConnectionRefusedError:
client_writer.close()
async def server_loop():
try:
while True:
client_reader, client_writer = await aiosam.stream_accept(
self.session_name, sam_address=self.sam_address,
loop=self.loop)
incoming = await client_reader.read(BUFFER_SIZE)
asyncio.ensure_future(handle_client(
incoming, client_reader, client_writer), loop=self.loop)
except asyncio.CancelledError:
pass
self.server_loop = asyncio.ensure_future(server_loop(), loop=self.loop)
def stop(self):
super().stop()
self.server_loop.cancel()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('type', metavar="TYPE", choices=('server', 'client'),
help="Tunnel type (server or client)")
parser.add_argument('address', metavar="ADDRESS",
help="Local address (e.g. 127.0.0.1:8000)")
parser.add_argument('--debug', '-d', action='store_true',
help='Debugging')
parser.add_argument('--key', '-k', default='', metavar='PRIVATE_KEY',
help='Path to private key file')
parser.add_argument('--destination', '-D', default='',
metavar='DESTINATION', help='Remote destination')
args = parser.parse_args()
SAM_ADDRESS = utils.get_sam_address()
logging.basicConfig(level=logging.DEBUG if args.debug else logging.INFO)
loop = asyncio.get_event_loop()
loop.set_debug(args.debug)
if args.key:
destination = sam.Destination(path=args.key, has_private_key=True)
else:
destination = None
local_address = utils.address_from_string(args.address)
if args.type == "client":
tunnel = ClientTunnel(args.destination, local_address, loop=loop,
destination=destination, sam_address=SAM_ADDRESS)
elif args.type == "server":
tunnel = ServerTunnel(local_address, loop=loop, destination=destination,
sam_address=SAM_ADDRESS)
asyncio.ensure_future(tunnel.run(), loop=loop)
try:
loop.run_forever()
except KeyboardInterrupt:
tunnel.stop()
finally:
loop.stop()
loop.close()
+42
View File
@@ -0,0 +1,42 @@
import socket
import os
import random
import string
from . import sam
def get_free_port():
"""Get a free port on your local host"""
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
s.bind(('', 0))
free_port = s.getsockname()[1]
s.close()
return free_port
def is_address_accessible(address):
"""Check if address is accessible or down"""
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
is_accessible = s.connect_ex(address) == 0
s.close()
return is_accessible
def address_from_string(address_string):
"""Address tuple from host:port string"""
address = address_string.split(":")
return (address[0], int(address[1]))
def get_sam_address():
"""
Get SAM address from environment variable I2P_SAM_ADDRESS, or use a default
value
"""
value = os.getenv("I2P_SAM_ADDRESS")
return address_from_string(value) if value else sam.DEFAULT_ADDRESS
def generate_session_id(length=6):
"""Generate random session id"""
rand = random.SystemRandom()
sid = [rand.choice(string.ascii_letters) for _ in range(length)]
return "reticulum-" + "".join(sid)
+45
View File
@@ -0,0 +1,45 @@
def get_platform():
from os import environ
if "ANDROID_ARGUMENT" in environ:
return "android"
elif "ANDROID_ROOT" in environ:
return "android"
else:
import sys
return sys.platform
def is_darwin():
if get_platform() == "darwin":
return True
else:
return False
def is_android():
if get_platform() == "android":
return True
else:
return False
def is_windows():
if str(get_platform()).startswith("win"):
return True
else:
return False
def platform_checks():
if is_windows():
import sys
if sys.version_info.major >= 3 and sys.version_info.minor >= 8:
pass
else:
import RNS
RNS.log("On Windows, Reticulum requires Python 3.8 or higher.", RNS.LOG_ERROR)
RNS.log("Please update Python to run Reticulum.", RNS.LOG_ERROR)
RNS.panic()
def cryptography_old_api():
import cryptography
if cryptography.__version__ == "2.8":
return True
else:
return False
+252 -128
View File
@@ -1,4 +1,4 @@
# u-msgpack-python v2.5.0 - v at sergeev.io
# u-msgpack-python v2.7.1 - v at sergeev.io
# https://github.com/vsergeev/u-msgpack-python
#
# u-msgpack-python is a lightweight MessagePack serializer and deserializer
@@ -10,7 +10,7 @@
#
# MIT License
#
# Copyright (c) 2013-2016 vsergeev / Ivan (Vanya) A. Sergeev
# Copyright (c) 2013-2020 vsergeev / Ivan (Vanya) A. Sergeev
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
@@ -31,7 +31,7 @@
# THE SOFTWARE.
#
"""
u-msgpack-python v2.5.0 - v at sergeev.io
u-msgpack-python v2.7.1 - v at sergeev.io
https://github.com/vsergeev/u-msgpack-python
u-msgpack-python is a lightweight MessagePack serializer and deserializer
@@ -49,10 +49,15 @@ import datetime
import sys
import io
__version__ = "2.5.0"
if sys.version_info[0:2] >= (3, 3):
from collections.abc import Hashable
else:
from collections import Hashable
__version__ = "2.7.1"
"Module version string"
version = (2, 5, 0)
version = (2, 7, 1)
"Module version tuple"
@@ -61,7 +66,7 @@ version = (2, 5, 0)
##############################################################################
# Extension type for application-defined types and data
class Ext:
class Ext(object):
"""
The Ext class facilitates creating a serializable extension object to store
an application-defined type and data byte array.
@@ -75,23 +80,33 @@ class Ext:
type: application-defined type integer
data: application-defined data byte array
TypeError:
Type is not an integer.
ValueError:
Type is out of range of -128 to 127.
TypeError::
Data is not type 'bytes' (Python 3) or not type 'str' (Python 2).
Example:
>>> foo = umsgpack.Ext(0x05, b"\x01\x02\x03")
>>> foo = umsgpack.Ext(5, b"\x01\x02\x03")
>>> umsgpack.packb({u"special stuff": foo, u"awesome": True})
'\x82\xa7awesome\xc3\xadspecial stuff\xc7\x03\x05\x01\x02\x03'
>>> bar = umsgpack.unpackb(_)
>>> print(bar["special stuff"])
Ext Object (Type: 0x05, Data: 01 02 03)
Ext Object (Type: 5, Data: 01 02 03)
>>>
"""
# Check type is type int
# Check type is type int and in range
if not isinstance(type, int):
raise TypeError("ext type is not type integer")
# Check data is type bytes
elif not (-2**7 <= type <= 2**7 - 1):
raise ValueError("ext type value {:d} is out of range (-128 to 127)".format(type))
# Check data is type bytes or str
elif sys.version_info[0] == 3 and not isinstance(data, bytes):
raise TypeError("ext data is not type \'bytes\'")
elif sys.version_info[0] == 2 and not isinstance(data, str):
raise TypeError("ext data is not type \'str\'")
self.type = type
self.data = data
@@ -99,9 +114,8 @@ class Ext:
"""
Compare this Ext object with another for equality.
"""
return (isinstance(other, self.__class__) and
self.type == other.type and
self.data == other.data)
return isinstance(other, self.__class__) \
and self.type == other.type and self.data == other.data
def __ne__(self, other):
"""
@@ -113,8 +127,8 @@ class Ext:
"""
String representation of this Ext object.
"""
s = "Ext Object (Type: 0x%02x, Data: " % self.type
s += " ".join(["0x%02x" % ord(self.data[i:i + 1])
s = "Ext Object (Type: {:d}, Data: ".format(self.type)
s += " ".join(["0x{:02}".format(ord(self.data[i:i + 1]))
for i in xrange(min(len(self.data), 8))])
if len(self.data) > 8:
s += " ..."
@@ -130,7 +144,52 @@ class Ext:
class InvalidString(bytes):
"""Subclass of bytes to hold invalid UTF-8 strings."""
pass
##############################################################################
# Ext Serializable Decorator
##############################################################################
_ext_class_to_type = {}
_ext_type_to_class = {}
def ext_serializable(ext_type):
"""
Return a decorator to register a class for automatic packing and unpacking
with the specified Ext type code. The application class should implement a
`packb()` method that returns serialized bytes, and an `unpackb()` class
method or static method that accepts serialized bytes and returns an
instance of the application class.
Args:
ext_type: application-defined Ext type code
Raises:
TypeError:
Ext type is not an integer.
ValueError:
Ext type is out of range of -128 to 127.
ValueError:
Ext type or class already registered.
"""
def wrapper(cls):
if not isinstance(ext_type, int):
raise TypeError("Ext type is not type integer")
elif not (-2**7 <= ext_type <= 2**7 - 1):
raise ValueError("Ext type value {:d} is out of range of -128 to 127".format(ext_type))
elif ext_type in _ext_type_to_class:
raise ValueError("Ext type {:d} already registered with class {:s}".format(ext_type, repr(_ext_type_to_class[ext_type])))
elif cls in _ext_class_to_type:
raise ValueError("Class {:s} already registered with Ext type {:d}".format(repr(cls), ext_type))
_ext_type_to_class[ext_type] = cls
_ext_class_to_type[cls] = ext_type
return cls
return wrapper
##############################################################################
# Exceptions
@@ -140,39 +199,32 @@ class InvalidString(bytes):
# Base Exception classes
class PackException(Exception):
"Base class for exceptions encountered during packing."
pass
class UnpackException(Exception):
"Base class for exceptions encountered during unpacking."
pass
# Packing error
class UnsupportedTypeException(PackException):
"Object type not supported for packing."
pass
# Unpacking error
class InsufficientDataException(UnpackException):
"Insufficient data to unpack the serialized object."
pass
class InvalidStringException(UnpackException):
"Invalid UTF-8 string encountered during unpacking."
pass
class UnsupportedTimestampException(UnpackException):
"Unsupported timestamp format encountered during unpacking."
pass
class ReservedCodeException(UnpackException):
"Reserved code encountered during unpacking."
pass
class UnhashableKeyException(UnpackException):
@@ -180,12 +232,10 @@ class UnhashableKeyException(UnpackException):
Unhashable key encountered during map unpacking.
The serialized map cannot be deserialized into a Python dictionary.
"""
pass
class DuplicateKeyException(UnpackException):
"Duplicate key encountered during map unpacking."
pass
# Backwards compatibility
@@ -250,15 +300,15 @@ def _pack_integer(obj, fp, options):
else:
raise UnsupportedTypeException("huge signed int")
else:
if obj <= 127:
if obj < 128:
fp.write(struct.pack("B", obj))
elif obj <= 2**8 - 1:
elif obj < 2**8:
fp.write(b"\xcc" + struct.pack("B", obj))
elif obj <= 2**16 - 1:
elif obj < 2**16:
fp.write(b"\xcd" + struct.pack(">H", obj))
elif obj <= 2**32 - 1:
elif obj < 2**32:
fp.write(b"\xce" + struct.pack(">I", obj))
elif obj <= 2**64 - 1:
elif obj < 2**64:
fp.write(b"\xcf" + struct.pack(">Q", obj))
else:
raise UnsupportedTypeException("huge unsigned int")
@@ -285,94 +335,99 @@ def _pack_float(obj, fp, options):
def _pack_string(obj, fp, options):
obj = obj.encode('utf-8')
if len(obj) <= 31:
fp.write(struct.pack("B", 0xa0 | len(obj)) + obj)
elif len(obj) <= 2**8 - 1:
fp.write(b"\xd9" + struct.pack("B", len(obj)) + obj)
elif len(obj) <= 2**16 - 1:
fp.write(b"\xda" + struct.pack(">H", len(obj)) + obj)
elif len(obj) <= 2**32 - 1:
fp.write(b"\xdb" + struct.pack(">I", len(obj)) + obj)
obj_len = len(obj)
if obj_len < 32:
fp.write(struct.pack("B", 0xa0 | obj_len) + obj)
elif obj_len < 2**8:
fp.write(b"\xd9" + struct.pack("B", obj_len) + obj)
elif obj_len < 2**16:
fp.write(b"\xda" + struct.pack(">H", obj_len) + obj)
elif obj_len < 2**32:
fp.write(b"\xdb" + struct.pack(">I", obj_len) + obj)
else:
raise UnsupportedTypeException("huge string")
def _pack_binary(obj, fp, options):
if len(obj) <= 2**8 - 1:
fp.write(b"\xc4" + struct.pack("B", len(obj)) + obj)
elif len(obj) <= 2**16 - 1:
fp.write(b"\xc5" + struct.pack(">H", len(obj)) + obj)
elif len(obj) <= 2**32 - 1:
fp.write(b"\xc6" + struct.pack(">I", len(obj)) + obj)
obj_len = len(obj)
if obj_len < 2**8:
fp.write(b"\xc4" + struct.pack("B", obj_len) + obj)
elif obj_len < 2**16:
fp.write(b"\xc5" + struct.pack(">H", obj_len) + obj)
elif obj_len < 2**32:
fp.write(b"\xc6" + struct.pack(">I", obj_len) + obj)
else:
raise UnsupportedTypeException("huge binary string")
def _pack_oldspec_raw(obj, fp, options):
if len(obj) <= 31:
fp.write(struct.pack("B", 0xa0 | len(obj)) + obj)
elif len(obj) <= 2**16 - 1:
fp.write(b"\xda" + struct.pack(">H", len(obj)) + obj)
elif len(obj) <= 2**32 - 1:
fp.write(b"\xdb" + struct.pack(">I", len(obj)) + obj)
obj_len = len(obj)
if obj_len < 32:
fp.write(struct.pack("B", 0xa0 | obj_len) + obj)
elif obj_len < 2**16:
fp.write(b"\xda" + struct.pack(">H", obj_len) + obj)
elif obj_len < 2**32:
fp.write(b"\xdb" + struct.pack(">I", obj_len) + obj)
else:
raise UnsupportedTypeException("huge raw string")
def _pack_ext(obj, fp, options):
if len(obj.data) == 1:
obj_len = len(obj.data)
if obj_len == 1:
fp.write(b"\xd4" + struct.pack("B", obj.type & 0xff) + obj.data)
elif len(obj.data) == 2:
elif obj_len == 2:
fp.write(b"\xd5" + struct.pack("B", obj.type & 0xff) + obj.data)
elif len(obj.data) == 4:
elif obj_len == 4:
fp.write(b"\xd6" + struct.pack("B", obj.type & 0xff) + obj.data)
elif len(obj.data) == 8:
elif obj_len == 8:
fp.write(b"\xd7" + struct.pack("B", obj.type & 0xff) + obj.data)
elif len(obj.data) == 16:
elif obj_len == 16:
fp.write(b"\xd8" + struct.pack("B", obj.type & 0xff) + obj.data)
elif len(obj.data) <= 2**8 - 1:
fp.write(b"\xc7" +
struct.pack("BB", len(obj.data), obj.type & 0xff) + obj.data)
elif len(obj.data) <= 2**16 - 1:
fp.write(b"\xc8" +
struct.pack(">HB", len(obj.data), obj.type & 0xff) + obj.data)
elif len(obj.data) <= 2**32 - 1:
fp.write(b"\xc9" +
struct.pack(">IB", len(obj.data), obj.type & 0xff) + obj.data)
elif obj_len < 2**8:
fp.write(b"\xc7" + struct.pack("BB", obj_len, obj.type & 0xff) + obj.data)
elif obj_len < 2**16:
fp.write(b"\xc8" + struct.pack(">HB", obj_len, obj.type & 0xff) + obj.data)
elif obj_len < 2**32:
fp.write(b"\xc9" + struct.pack(">IB", obj_len, obj.type & 0xff) + obj.data)
else:
raise UnsupportedTypeException("huge ext data")
def _pack_ext_timestamp(obj, fp, options):
delta = obj - _epoch
if not obj.tzinfo:
# Object is naive datetime, convert to aware date time,
# assuming UTC timezone
delta = obj.replace(tzinfo=_utc_tzinfo) - _epoch
else:
# Object is aware datetime
delta = obj - _epoch
seconds = delta.seconds + delta.days * 86400
microseconds = delta.microseconds
if microseconds == 0 and 0 <= seconds <= 2**32 - 1:
# 32-bit timestamp
fp.write(b"\xd6\xff" +
struct.pack(">I", seconds))
fp.write(b"\xd6\xff" + struct.pack(">I", seconds))
elif 0 <= seconds <= 2**34 - 1:
# 64-bit timestamp
value = ((microseconds * 1000) << 34) | seconds
fp.write(b"\xd7\xff" +
struct.pack(">Q", value))
fp.write(b"\xd7\xff" + struct.pack(">Q", value))
elif -2**63 <= abs(seconds) <= 2**63 - 1:
# 96-bit timestamp
fp.write(b"\xc7\x0c\xff" +
struct.pack(">I", microseconds * 1000) +
struct.pack(">q", seconds))
fp.write(b"\xc7\x0c\xff" + struct.pack(">Iq", microseconds * 1000, seconds))
else:
raise UnsupportedTypeException("huge timestamp")
def _pack_array(obj, fp, options):
if len(obj) <= 15:
fp.write(struct.pack("B", 0x90 | len(obj)))
elif len(obj) <= 2**16 - 1:
fp.write(b"\xdc" + struct.pack(">H", len(obj)))
elif len(obj) <= 2**32 - 1:
fp.write(b"\xdd" + struct.pack(">I", len(obj)))
obj_len = len(obj)
if obj_len < 16:
fp.write(struct.pack("B", 0x90 | obj_len))
elif obj_len < 2**16:
fp.write(b"\xdc" + struct.pack(">H", obj_len))
elif obj_len < 2**32:
fp.write(b"\xdd" + struct.pack(">I", obj_len))
else:
raise UnsupportedTypeException("huge array")
@@ -381,12 +436,13 @@ def _pack_array(obj, fp, options):
def _pack_map(obj, fp, options):
if len(obj) <= 15:
fp.write(struct.pack("B", 0x80 | len(obj)))
elif len(obj) <= 2**16 - 1:
fp.write(b"\xde" + struct.pack(">H", len(obj)))
elif len(obj) <= 2**32 - 1:
fp.write(b"\xdf" + struct.pack(">I", len(obj)))
obj_len = len(obj)
if obj_len < 16:
fp.write(struct.pack("B", 0x80 | obj_len))
elif obj_len < 2**16:
fp.write(b"\xde" + struct.pack(">H", obj_len))
elif obj_len < 2**32:
fp.write(b"\xdf" + struct.pack(">I", obj_len))
else:
raise UnsupportedTypeException("huge array")
@@ -435,9 +491,14 @@ def _pack2(obj, fp, **options):
_pack_nil(obj, fp, options)
elif ext_handlers and obj.__class__ in ext_handlers:
_pack_ext(ext_handlers[obj.__class__](obj), fp, options)
elif obj.__class__ in _ext_class_to_type:
try:
_pack_ext(Ext(_ext_class_to_type[obj.__class__], obj.packb()), fp, options)
except AttributeError:
raise NotImplementedError("Ext serializable class {:s} is missing implementation of packb()".format(repr(obj.__class__)))
elif isinstance(obj, bool):
_pack_boolean(obj, fp, options)
elif isinstance(obj, int) or isinstance(obj, long):
elif isinstance(obj, (int, long)):
_pack_integer(obj, fp, options)
elif isinstance(obj, float):
_pack_float(obj, fp, options)
@@ -449,7 +510,7 @@ def _pack2(obj, fp, **options):
_pack_string(obj, fp, options)
elif isinstance(obj, str):
_pack_binary(obj, fp, options)
elif isinstance(obj, list) or isinstance(obj, tuple):
elif isinstance(obj, (list, tuple)):
_pack_array(obj, fp, options)
elif isinstance(obj, dict):
_pack_map(obj, fp, options)
@@ -464,9 +525,19 @@ def _pack2(obj, fp, **options):
_pack_ext(ext_handlers[t](obj), fp, options)
else:
raise UnsupportedTypeException(
"unsupported type: %s" % str(type(obj)))
"unsupported type: {:s}".format(str(type(obj))))
elif _ext_class_to_type:
# Linear search for superclass
t = next((t for t in _ext_class_to_type if isinstance(obj, t)), None)
if t:
try:
_pack_ext(Ext(_ext_class_to_type[t], obj.packb()), fp, options)
except AttributeError:
raise NotImplementedError("Ext serializable class {:s} is missing implementation of packb()".format(repr(t)))
else:
raise UnsupportedTypeException("unsupported type: {:s}".format(str(type(obj))))
else:
raise UnsupportedTypeException("unsupported type: %s" % str(type(obj)))
raise UnsupportedTypeException("unsupported type: {:s}".format(str(type(obj))))
# Pack for Python 3, with unicode 'str' type, 'bytes' type, and no 'long' type
@@ -507,6 +578,11 @@ def _pack3(obj, fp, **options):
_pack_nil(obj, fp, options)
elif ext_handlers and obj.__class__ in ext_handlers:
_pack_ext(ext_handlers[obj.__class__](obj), fp, options)
elif obj.__class__ in _ext_class_to_type:
try:
_pack_ext(Ext(_ext_class_to_type[obj.__class__], obj.packb()), fp, options)
except AttributeError:
raise NotImplementedError("Ext serializable class {:s} is missing implementation of packb()".format(repr(obj.__class__)))
elif isinstance(obj, bool):
_pack_boolean(obj, fp, options)
elif isinstance(obj, int):
@@ -521,7 +597,7 @@ def _pack3(obj, fp, **options):
_pack_string(obj, fp, options)
elif isinstance(obj, bytes):
_pack_binary(obj, fp, options)
elif isinstance(obj, list) or isinstance(obj, tuple):
elif isinstance(obj, (list, tuple)):
_pack_array(obj, fp, options)
elif isinstance(obj, dict):
_pack_map(obj, fp, options)
@@ -536,10 +612,20 @@ def _pack3(obj, fp, **options):
_pack_ext(ext_handlers[t](obj), fp, options)
else:
raise UnsupportedTypeException(
"unsupported type: %s" % str(type(obj)))
"unsupported type: {:s}".format(str(type(obj))))
elif _ext_class_to_type:
# Linear search for superclass
t = next((t for t in _ext_class_to_type if isinstance(obj, t)), None)
if t:
try:
_pack_ext(Ext(_ext_class_to_type[t], obj.packb()), fp, options)
except AttributeError:
raise NotImplementedError("Ext serializable class {:s} is missing implementation of packb()".format(repr(t)))
else:
raise UnsupportedTypeException("unsupported type: {:s}".format(str(type(obj))))
else:
raise UnsupportedTypeException(
"unsupported type: %s" % str(type(obj)))
"unsupported type: {:s}".format(str(type(obj))))
def _packb2(obj, **options):
@@ -613,9 +699,20 @@ def _packb3(obj, **options):
def _read_except(fp, n):
if n == 0:
return b""
data = fp.read(n)
if len(data) < n:
if len(data) == 0:
raise InsufficientDataException()
while len(data) < n:
chunk = fp.read(n - len(data))
if len(chunk) == 0:
raise InsufficientDataException()
data += chunk
return data
@@ -640,21 +737,21 @@ def _unpack_integer(code, fp, options):
return struct.unpack(">I", _read_except(fp, 4))[0]
elif code == b'\xcf':
return struct.unpack(">Q", _read_except(fp, 8))[0]
raise Exception("logic error, not int: 0x%02x" % ord(code))
raise Exception("logic error, not int: 0x{:02x}".format(ord(code)))
def _unpack_reserved(code, fp, options):
if code == b'\xc1':
raise ReservedCodeException(
"encountered reserved code: 0x%02x" % ord(code))
"encountered reserved code: 0x{:02x}".format(ord(code)))
raise Exception(
"logic error, not reserved code: 0x%02x" % ord(code))
"logic error, not reserved code: 0x{:02x}".format(ord(code)))
def _unpack_nil(code, fp, options):
if code == b'\xc0':
return None
raise Exception("logic error, not nil: 0x%02x" % ord(code))
raise Exception("logic error, not nil: 0x{:02x}".format(ord(code)))
def _unpack_boolean(code, fp, options):
@@ -662,7 +759,7 @@ def _unpack_boolean(code, fp, options):
return False
elif code == b'\xc3':
return True
raise Exception("logic error, not boolean: 0x%02x" % ord(code))
raise Exception("logic error, not boolean: 0x{:02x}".format(ord(code)))
def _unpack_float(code, fp, options):
@@ -670,7 +767,7 @@ def _unpack_float(code, fp, options):
return struct.unpack(">f", _read_except(fp, 4))[0]
elif code == b'\xcb':
return struct.unpack(">d", _read_except(fp, 8))[0]
raise Exception("logic error, not float: 0x%02x" % ord(code))
raise Exception("logic error, not float: 0x{:02x}".format(ord(code)))
def _unpack_string(code, fp, options):
@@ -683,7 +780,7 @@ def _unpack_string(code, fp, options):
elif code == b'\xdb':
length = struct.unpack(">I", _read_except(fp, 4))[0]
else:
raise Exception("logic error, not string: 0x%02x" % ord(code))
raise Exception("logic error, not string: 0x{:02x}".format(ord(code)))
# Always return raw bytes in compatibility mode
global compatibility
@@ -707,7 +804,7 @@ def _unpack_binary(code, fp, options):
elif code == b'\xc6':
length = struct.unpack(">I", _read_except(fp, 4))[0]
else:
raise Exception("logic error, not binary: 0x%02x" % ord(code))
raise Exception("logic error, not binary: 0x{:02x}".format(ord(code)))
return _read_except(fp, length)
@@ -730,43 +827,48 @@ def _unpack_ext(code, fp, options):
elif code == b'\xc9':
length = struct.unpack(">I", _read_except(fp, 4))[0]
else:
raise Exception("logic error, not ext: 0x%02x" % ord(code))
raise Exception("logic error, not ext: 0x{:02x}".format(ord(code)))
ext_type = struct.unpack("b", _read_except(fp, 1))[0]
ext_data = _read_except(fp, length)
# Create extension object
ext = Ext(ext_type, ext_data)
# Unpack with ext handler, if we have one
ext_handlers = options.get("ext_handlers")
if ext_handlers and ext.type in ext_handlers:
return ext_handlers[ext.type](ext)
if ext_handlers and ext_type in ext_handlers:
return ext_handlers[ext_type](Ext(ext_type, ext_data))
# Unpack with ext classes, if type is registered
if ext_type in _ext_type_to_class:
try:
return _ext_type_to_class[ext_type].unpackb(ext_data)
except AttributeError:
raise NotImplementedError("Ext serializable class {:s} is missing implementation of unpackb()".format(repr(_ext_type_to_class[ext_type])))
# Timestamp extension
if ext.type == -1:
return _unpack_ext_timestamp(ext, options)
if ext_type == -1:
return _unpack_ext_timestamp(ext_data, options)
return ext
return Ext(ext_type, ext_data)
def _unpack_ext_timestamp(ext, options):
if len(ext.data) == 4:
def _unpack_ext_timestamp(ext_data, options):
obj_len = len(ext_data)
if obj_len == 4:
# 32-bit timestamp
seconds = struct.unpack(">I", ext.data)[0]
seconds = struct.unpack(">I", ext_data)[0]
microseconds = 0
elif len(ext.data) == 8:
elif obj_len == 8:
# 64-bit timestamp
value = struct.unpack(">Q", ext.data)[0]
value = struct.unpack(">Q", ext_data)[0]
seconds = value & 0x3ffffffff
microseconds = (value >> 34) // 1000
elif len(ext.data) == 12:
elif obj_len == 12:
# 96-bit timestamp
seconds = struct.unpack(">q", ext.data[4:12])[0]
microseconds = struct.unpack(">I", ext.data[0:4])[0] // 1000
seconds = struct.unpack(">q", ext_data[4:12])[0]
microseconds = struct.unpack(">I", ext_data[0:4])[0] // 1000
else:
raise UnsupportedTimestampException(
"unsupported timestamp with data length %d" % len(ext.data))
"unsupported timestamp with data length {:d}".format(len(ext_data)))
return _epoch + datetime.timedelta(seconds=seconds,
microseconds=microseconds)
@@ -780,7 +882,10 @@ def _unpack_array(code, fp, options):
elif code == b'\xdd':
length = struct.unpack(">I", _read_except(fp, 4))[0]
else:
raise Exception("logic error, not array: 0x%02x" % ord(code))
raise Exception("logic error, not array: 0x{:02x}".format(ord(code)))
if options.get('use_tuple'):
return tuple((_unpack(fp, options) for i in xrange(length)))
return [_unpack(fp, options) for i in xrange(length)]
@@ -799,10 +904,9 @@ def _unpack_map(code, fp, options):
elif code == b'\xdf':
length = struct.unpack(">I", _read_except(fp, 4))[0]
else:
raise Exception("logic error, not map: 0x%02x" % ord(code))
raise Exception("logic error, not map: 0x{:02x}".format(ord(code)))
d = {} if not options.get('use_ordered_dict') \
else collections.OrderedDict()
d = {} if not options.get('use_ordered_dict') else collections.OrderedDict()
for _ in xrange(length):
# Unpack key
k = _unpack(fp, options)
@@ -810,12 +914,12 @@ def _unpack_map(code, fp, options):
if isinstance(k, list):
# Attempt to convert list into a hashable tuple
k = _deep_list_to_tuple(k)
elif not isinstance(k, collections.Hashable):
elif not isinstance(k, Hashable):
raise UnhashableKeyException(
"encountered unhashable key: %s, %s" % (str(k), str(type(k))))
"encountered unhashable key: \"{:s}\" ({:s})".format(str(k), str(type(k))))
elif k in d:
raise DuplicateKeyException(
"encountered duplicate key: %s, %s" % (str(k), str(type(k))))
"encountered duplicate key: \"{:s}\" ({:s})".format(str(k), str(type(k))))
# Unpack value
v = _unpack(fp, options)
@@ -824,7 +928,7 @@ def _unpack_map(code, fp, options):
d[k] = v
except TypeError:
raise UnhashableKeyException(
"encountered unhashable key: %s" % str(k))
"encountered unhashable key: \"{:s}\"".format(str(k)))
return d
@@ -848,6 +952,8 @@ def _unpack2(fp, **options):
Ext into an object
use_ordered_dict (bool): unpack maps into OrderedDict, instead of
unordered dict (default False)
use_tuple (bool): unpacks arrays into tuples, instead of lists (default
False)
allow_invalid_utf8 (bool): unpack invalid strings into instances of
InvalidString, for access to the bytes
(default False)
@@ -892,6 +998,8 @@ def _unpack3(fp, **options):
Ext into an object
use_ordered_dict (bool): unpack maps into OrderedDict, instead of
unordered dict (default False)
use_tuple (bool): unpacks arrays into tuples, instead of lists (default
False)
allow_invalid_utf8 (bool): unpack invalid strings into instances of
InvalidString, for access to the bytes
(default False)
@@ -937,6 +1045,8 @@ def _unpackb2(s, **options):
Ext into an object
use_ordered_dict (bool): unpack maps into OrderedDict, instead of
unordered dict (default False)
use_tuple (bool): unpacks arrays into tuples, instead of lists (default
False)
allow_invalid_utf8 (bool): unpack invalid strings into instances of
InvalidString, for access to the bytes
(default False)
@@ -985,6 +1095,8 @@ def _unpackb3(s, **options):
Ext into an object
use_ordered_dict (bool): unpack maps into OrderedDict, instead of
unordered dict (default False)
use_tuple (bool): unpacks arrays into tuples, instead of lists (default
False)
allow_invalid_utf8 (bool): unpack invalid strings into instances of
InvalidString, for access to the bytes
(default False)
@@ -1045,9 +1157,21 @@ def __init():
if sys.version_info[0] == 3:
_utc_tzinfo = datetime.timezone.utc
else:
_utc_tzinfo = None
class UTC(datetime.tzinfo):
ZERO = datetime.timedelta(0)
# Calculate epoch datetime
def utcoffset(self, dt):
return UTC.ZERO
def tzname(self, dt):
return "UTC"
def dst(self, dt):
return UTC.ZERO
_utc_tzinfo = UTC()
# Calculate an aware epoch datetime
_epoch = datetime.datetime(1970, 1, 1, tzinfo=_utc_tzinfo)
# Auto-detect system float precision
Binary file not shown.
+1 -1
View File
@@ -1,4 +1,4 @@
# Sphinx build info version 1
# This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
config: bf2e68cefd79a49afe077549bac593bf
config: d4939f555bda9c488f47cdcede85949d
tags: 645f666f9bcd5a90fca523b33c5a78b7
Binary file not shown.

After

Width:  |  Height:  |  Size: 79 KiB

+228 -9
View File
@@ -6,17 +6,37 @@ The best way to get started with the Reticulum Network Stack depends on what
you want to do. This guide will outline sensible starting paths for different
scenarios.
Try Using a Reticulum-based Program
=============================================
If you simply want to try using a program built with Reticulum, you can take
a look at `Nomad Network <https://github.com/markqvist/nomadnet>`_, which
provides a basic encrypted communications suite built completely on Reticulum.
If you simply want to try using a program built with Reticulum, a few different
programs exist that allow basic communication and a range of other useful functions
over even extremely low-bandwidth Reticulum networks.
These programs will let you get a feel for how Reticulum works. They have been designed
to run well over networks based on LoRa or packet radio, but can also be used completely
over local WiFi, wired ethernet, the Internet, or any combination.
As such, it is easy to get started experimenting, without having to set up any radio
transceivers or infrastructure just to try it out. Launching the programs on separate
devices connected to the same WiFi network is enough to get started, and physical
radio interfaces can then be added later.
Nomad Network
^^^^^^^^^^^^^
The terminal-based program `Nomad Network <https://github.com/markqvist/nomadnet>`_
provides a complete encrypted communications suite built with Reticulum. It features
encrypted messaging (both direct and delayed-delivery for offline users), file sharing,
and has a built-in text-browser and page server with support for dynamically rendered pages,
user authentication and more.
.. image:: screenshots/nomadnet_3.png
:target: _images/nomadnet_3.png
`Nomad Network <https://github.com/markqvist/nomadnet>`_ is a user-facing client
in the development for the messaging and information-sharing protocol
for the messaging and information-sharing protocol
`LXMF <https://github.com/markqvist/lxmf>`_, another project built with Reticulum.
You can install Nomad Network via pip:
@@ -29,28 +49,136 @@ You can install Nomad Network via pip:
# ... and run
nomadnet
**Please Note**: If this is the very first time you use pip to install a program
on your system, you might need to reboot your system for your program to become
available. If you get a "command not found" error or similar when running the
program, reboot your system and try again.
Sideband
^^^^^^^^
If you would rather use a program with a graphical user interface, you can take
a look at `Sideband <https://unsigned.io/sideband>`_, which is available for Android,
Linux and macOS.
.. image:: screenshots/sideband_1.png
:width: 400px
:align: center
:target: _images/sideband_1.png
Sideband is currently in the early stages of development, but already provides basic
communication features, and interoperates with Nomad Network, or any other LXMF client.
Using the Included Utilities
=============================================
Reticulum comes with a range of included utilities that make it easier to
manage your network, check connectivity and make Reticulum available to other
programs on your system.
You can use ``rnsd`` to run Reticulum as a background or foreground service,
and the ``rnstatus``, ``rnpath`` and ``rnprobe`` utilities to view and query
network status and connectivity.
To learn more about these utility programs, have a look at the
:ref:`Using Reticulum on Your System<using-main>` chapter of this manual.
Creating a Network With Reticulum
=============================================
To create a network, you will need to specify one or more *interfaces* for
Reticulum to use. This is done in the Reticulum configuration file, which by
default is located at ``~/.reticulum/config``.
default is located at ``~/.reticulum/config``. You can edit this file by hand,
or use the interactive ``rnsconfig`` utility.
When Reticulum is started for the first time, it will create a default
configuration file, with one active interface. This default interface uses
your existing ethernet network (if there is one), and only allows you to
communicate with other Reticulum peers within your local broadcast domain.
your existing ethernet and WiFi networks (if any), and only allows you to
communicate with other Reticulum peers within your local broadcast domains.
To communicate further, you will have to add one or more interfaces. The default
configuration includes a number of examples, ranging from using TCP over the
internet, to LoRa and Packet Radio interfaces.
With Reticulum, you only need to configure what interfaces you want to communicate
over. There is no need to configure address spaces, subnets, routing tables,
or other things you might be used to from other network types.
Once Reticulums knows which interfaces it should use, it will automatically
discover topography and configure transport of data to any destinations it
knows about.
In situations where you already have an established WiFi or ethernet network, and
many devices that want to utilise the same external Reticulum network (for example over
LoRa), it will often be sufficient to let one system act as a Reticulum gateway, by
adding any external interfaces to this systems configuration, and enabling transport. Any
other device on your local WiFi will then be able to connect to this wider Reticulum
network just using the default interface configuration.
Possibly, the examples in the config file are enough to get you started. If
you want more information, you can read the :ref:`Building Networks<networks-main>`
and :ref:`Interfaces<interfaces-main>` chapters of this manual.
Connecting Reticulum Instances Over the Internet
================================================
Reticulum currently offers two interfaces suitable for connecting instances over the Internet: :ref:`TCP<interfaces-tcps>`
and :ref:`I2P<interfaces-i2p>`. Each interface offers a different set of features, and Reticulum
users should carefully choose the interface which best suites their needs.
The ``TCPServerInterface`` allows users to host an instance accessible over TCP/IP. This
method is generally faster, lower latency, and more energy efficient than using ``I2PInterface``,
however it also leaks more data about the server host.
TCP connections reveal the IP address of both your instance and the server to anyone who can
inspect the connection. Someone could use this information to determine your location or identity. Adversaries
inspecting your packets may be able to record packet metadata like time of transmission and packet size.
Even though Reticulum encrypts traffic, TCP does not, so an adversary may be able to use
packet inspection to learn that a system is running Reticulum, and what other IP adresses connect to it.
Hosting a publicly reachable instance over TCP also requires a publicly reachable IP address,
which most Internet connections don't offer anymore.
The ``I2PInterface`` routes messages through the `Invisible Internet Protocol
(I2P) <https://geti2p.net/en/>`_. To properly use this interface, users must also run an I2P daemon in
parallel to ``rnsd``. For always-on I2P nodes it is recommended to use `i2pd <https://i2pd.website/>`_.
By default, I2P will encrypt and mix all traffic sent over the Internet, and
hide both the sender and receiver Reticulum instance IP addresses. Running an I2P node
will also relay other I2P user's encrypted packets, which will use extra
bandwidth and compute power, but also makes timing attacks and other forms of
deep-packet-inspection much more difficult.
I2P also allows users to host globally available Reticulum instances from non-public IPs and behind firewalls.
In general it is recommended to use an I2P node if you want to host a publically accessible
instance, while preserving anonymity. If you care more about performance, and a slightly
easier setup, use TCP.
Connect to the Public Testnet
===========================================
An experimental public testnet has been made accessible over both I2P and TCP. You can join it
by adding one of the following interfaces to your ``.reticulum/config`` file:
.. code::
# For connecting over TCP/IP:
[[RNS Testnet Frankfurt]]
type = TCPClientInterface
interface_enabled = yes
outgoing = True
target_host = frankfurt.rns.unsigned.io
target_port = 4965
# For connecting over I2P:
[[RNS Testnet I2P Node A]]
type = I2PInterface
interface_enabled = yes
peers = ykzlw5ujbaqc2xkec4cpvgyxj257wcrmmgkuxqmqcur7cq3w3lha.b32.i2p
Many other Reticulum instances are connecting to this testnet, and you can also join it
via other entry points if you know them. There is absolutely no control over the network
topography, usage or what types of instances connect. It will also occasionally be used
to test various failure scenarios, and there are no availability or service guarantees.
Develop a Program with Reticulum
===========================================
@@ -65,6 +193,13 @@ The above command will install Reticulum and dependencies, and you will be
ready to import and use RNS in your own programs. The next step will most
likely be to look at some :ref:`Example Programs<examples-main>`.
For extended functionality, you can install optional dependencies:
.. code::
pip3 install pyserial netifaces
Further information can be found in the :ref:`API Reference<api-main>`.
@@ -108,4 +243,88 @@ don't use pip, but try this recipe:
python3 Examples/Filetransfer.py -h
When you have experimented with the basic examples, it's time to go read the
:ref:`Understanding Reticulum<understanding-main>` chapter.
:ref:`Understanding Reticulum<understanding-main>` chapter.
Reticulum on ARM64
==============================================
On some architectures, including ARM64, not all dependencies have precompiled
binaries. On such systems, you will need to install ``python3-dev`` before
installing Reticulum or programs that depend on Reticulum.
.. code::
# Install Python and development packages
sudo apt update
sudo apt install python3 python3-pip python3-dev
# Install Reticulum
python3 -m pip install rns
Reticulum on Android
==============================================
Reticulum can be used on Android in different ways. The easiest way to get
started is using an app like `Sideband <https://unsigned.io/sideband>`_.
For more control and features, you can use Reticulum and related programs via
the `Termux app <https://termux.com/>`_, at the time of writing available on
`F-droid <https://f-droid.org>`_.
Termux is a terminal emulator and Linux environment for Android based devices,
which includes the ability to use many different programs and libraries,
including Reticulum.
Since the Python cryptography.io module does not offer pre-built wheels for
Android, the standard one-line install of Reticulum does not work on Android,
and a few extra commands are required.
From within Termux, execute the following:
.. code::
# First, make sure indexes and packages are up to date.
pkg update
pkg upgrade
# Then install dependencies for the cryptography library.
pkg install python build-essential openssl libffi rust
# Make sure pip is up to date, and install the wheel module.
pip3 install wheel pip --upgrade
# To allow the installer to build the cryptography module,
# we need to let it know what platform we are compiling for:
export CARGO_BUILD_TARGET="aarch64-linux-android"
# Start the install process for the cryptography module.
# Depending on your device, this can take several minutes,
# since the module must be compiled locally on your device.
pip3 install cryptography
# If the above installation succeeds, you can now install
# Reticulum and any related software
pip3 install rns
It is also possible to include Reticulum in apps compiled and distributed as
Android APKs. A detailed tutorial and example source code will be included
here at a later point.
Adding Radio Interfaces
==============================================
Once you have Reticulum installed and working, you can add radio interfaces with
any compatible hardware you have available. For information on how to configure
this, see the :ref:`Interfaces<interfaces-main>` section of this manual.
A range of common LoRa development boards and transceiver modules can be used
as interfaces with Reticulum. You can refer to the following external resources
for more information:
* `How To Make Your Own RNodes <https://unsigned.io/how-to-make-your-own-rnodes/>`_
* `Installing RNode Firmware on Compatible LoRa Devices <https://unsigned.io/installing-rnode-firmware-on-t-beam-and-lora32-devices/>`_
* `Private, Secure and Uncensorable Messaging Over a LoRa Mesh <https://unsigned.io/private-messaging-over-lora/>`_
* `RNode Firmware <https://github.com/markqvist/RNode_Firmware/>`_
If you have communications hardware that you think would be suitable for use with Reticulum,
you are welcome to head over to the `GitHub discussion pages <https://github.com/markqvist/Reticulum/discussions>`_
and propose adding an interface for the hardware.
+311 -73
View File
@@ -14,6 +14,308 @@ for Reticulum to use.
The following sections describe the interfaces currently available in Reticulum,
and gives example configurations for the respective interface types.
For a high-level overview of how networks can be formed over different interface
types, have a look at the :ref:`Building Networks<networks-main>` chapter of this
manual.
.. _interfaces-options:
Common Interface Options
========================
A number of general configuration options are available on most interfaces.
These can be used to control various aspects of interface behaviour.
* | The ``enabled`` option tells Reticulum whether or not
to bring up the interface. Defaults to ``False``. For any
interface to be brought up, the ``enabled`` option
must be set to ``True`` or ``Yes``.
* | The ``mode`` option allows selecting the high-level behaviour
of the interface from a number of options.
- The default value is ``full``. In this mode, all discovery,
meshing and transport functionality is available.
- In the ``access_point`` (or shorthand ``ap``) mode, the
interface will operate as a network access point. In this
mode, announces will not be automatically broadcasted on
the interface, and paths to destinations on the interface
will have a much shorter expiry time. This mode is useful
for creating interfaces that are mostly quiet, unless when
someone is actually using them. An example of this could
be a radio interface serving a wide area, where users are
expected to connect momentarily, use the network, and then
disappear again.
* | The ``outgoing`` option sets whether an interface is allowed
to transmit. Defaults to ``True``. If set to ``False`` or ``No``
the interface will only receive data, and never transmit.
* | The ``network_name`` option sets the virtual network name for
the interface. This allows multiple separate network segments
to exist on the same physical channel or medium.
* | The ``passphrase`` option sets an authentication passphrase on
the interface. This option can be used in conjunction with the
``network_name`` option, or be used alone.
* | The ``ifac_size`` option allows customising the length of the
Interface Authentication Codes carried by each packet on named
and/or authenticated network segments. It is set by default to
a size suitable for the interface in question, but can be set
to a custom size between 8 and 512 bits by using this option.
In normal usage, this option should not be changed from the
default.
* | The ``announce_cap`` option lets you configure the maximum
bandwidth to allocate, at any given time, to propagating
announces and other network upkeep traffic. It is configured at
2% by default, and should normally not need to be changed. Can
be set to any value between ``1`` and ``100``.
* | The ``bitrate`` option configures the interface bitrate.
Reticulum will use interface speeds reported by hardware, or
try to guess a suitable rate when the hardware doesn't report
any. In most cases, the automatically found rate should be
sufficient, but it can be configured by using the ``bitrate``
option, to set the interface speed in *bits per second*.
.. _interfaces-auto:
Auto Interface
==============
The Auto Interface enables communication with other discoverable Reticulum
nodes over autoconfigured IPv6 and UDP. It does not need any functional IP
infrastructure like routers or DHCP servers, but will require at least some
sort of switching medium between peers (a wired switch, a hub, a WiFi access
point or similar), and that link-local IPv6 is enabled in your operating
system, which should be enabled by default in almost all OSes.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[Default Interface]]
type = AutoInterface
interface_enabled = True
# You can create multiple isolated Reticulum
# networks on the same physical LAN by
# specifying different Group IDs.
group_id = reticulum
# You can also select specifically which
# kernel networking devices to use.
devices = wlan0,eth1
# Or let AutoInterface use all suitable
# devices except for a list of ignored ones.
ignored_devices = tun0,eth0
If you are connected to the Internet with IPv6, and your provider will route
IPv6 multicast, you can potentially configure the Auto Interface to globally
autodiscover other Reticulum nodes within your selected Group ID. You can specify
the discovery scope by setting it to one of ``link``, ``admin``, ``site``,
``organisation`` or ``global``.
.. code::
[[Default Interface]]
type = AutoInterface
interface_enabled = True
# Configure global discovery
group_id = custom_network_name
discovery_scope = global
# Other configuration options
discovery_port = 48555
data_port = 49555
.. _interfaces-i2p:
I2P Interface
=============
The I2P interface lets you connect Reticulum instances over the
`Invisible Internet Protocol <https://i2pd.website>`_. This can be
especially useful in cases where you want to host a globally reachable
Reticulum instance, but do not have access to any public IP addresses,
have a frequently changing IP address, or have firewalls blocking
inbound traffic.
Using the I2P interface, you will get a globally reachable, portable
and persistent I2P address that your Reticulum instance can be reached
at.
To use the I2P interface, you must have an I2P router running
on your system. The easiest way to acheive this is to download and
install the `latest release <https://github.com/PurpleI2P/i2pd/releases/latest>`_
of the ``ì2pd`` package. For more details about I2P, see the
`geti2p.net website <https://geti2p.net/en/about/intro>`_.`
When an I2P router is running on your system, you can simply add
an I2P interface to reticulum:
.. code::
[[I2P]]
type = I2PInterface
interface_enabled = yes
connectable = yes
On the first start, Reticulum will generate a new I2P address for the
interface and start listening for inbound traffic on it. This can take
a while the first time, especially if your I2P router was also just
started, and is not yet well-connected to the I2P network. When ready,
you should see I2P base32 address printed to your log file. You can
also inspect the status of the interface using the ``rnstatus`` utility.
To connect to other Reticulum instances over I2P, just add a comma-separated
list of I2P base32 addresses to the ``peers`` option of the interface:
.. code::
[[I2P]]
type = I2PInterface
interface_enabled = yes
connectable = yes
peers = 5urvjicpzi7q3ybztsef4i5ow2aq4soktfj7zedz53s47r54jnqq.b32.i2p
It can take anywhere from a few seconds to a few minutes to establish
I2P connections to the desired peers, so Reticulum handles the process
in the background, and will output relevant events to the log.
**Please Note!** While the I2P interface is the simplest way to use
Reticulum over I2P, it is also possible to tunnel the TCP server and
client interfaces over I2P manually. This can be useful in situations
where more control is needed, but requires manual tunnel setup through
the I2P daemon configuration.
It is important to note that the two methods are *interchangably compatible*.
You can use the I2PInterface to connect to a TCPServerInterface that
was manually tunneled over I2P, for example. This offers a high degree
of flexibility in network setup, while retaining ease of use in simpler
use-cases.
.. _interfaces-tcps:
TCP Server Interface
====================
The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[TCP Server Interface]]
type = TCPServerInterface
interface_enabled = True
# This configuration will listen on all IP
# interfaces on port 4242
listen_ip = 0.0.0.0
listen_port = 4242
# Alternatively you can bind to a specific IP
# listen_ip = 10.0.0.88
# listen_port = 4242
# Or a specific network device
# device = eth0
# port = 4242
**Please Note!** The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:
.. code::
[[TCP Server on I2P]]
type = TCPServerInterface
interface_enabled = yes
listen_ip = 127.0.0.1
listen_port = 5001
i2p_tunneled = yes
.. _interfaces-tcpc:
TCP Client Interface
====================
To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.
.. code::
# Here's an example of a TCP Client interface. The
# target_host can either be an IP address or a hostname.
[[TCP Client Interface]]
type = TCPClientInterface
interface_enabled = True
target_host = 127.0.0.1
target_port = 4242
It is also possible to use this interface type to connect via other programs
or hardware devices that expose a KISS interface on a TCP port, for example
software-based soundmodems. To do this, use the ``kiss_framing`` option:
.. code::
# Here's an example of a TCP Client interface that connects
# to a software TNC soundmodem on a KISS over TCP port.
[[TCP KISS Interface]]
type = TCPClientInterface
interface_enabled = True
kiss_framing = True
target_host = 127.0.0.1
target_port = 8001
**Caution!** Only use the KISS framing option when connecting to external devices
and programs like soundmodems and similar over TCP. When using the
``TCPClientInterface`` in conjunction with the ``TCPServerInterface`` you should
never enable ``kiss_framing``, since this will disable internal reliability and
recovery mechanisms that greatly improves performance over unreliable and
intermittent TCP links.
**Please Note!** The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:
.. code::
[[TCP Client over I2P]]
type = TCPClientInterface
interface_enabled = yes
target_host = 127.0.0.1
target_port = 5001
i2p_tunneled = yes
.. _interfaces-udp:
UDP Interface
@@ -24,6 +326,12 @@ private and the internet. It can also allow broadcast communication
over IP networks, so it can provide an easy way to enable connectivity
with all other peers on a local area network.
*Please Note!* Using broadcast UDP traffic has performance implications,
especially on WiFi. If your goal is simply to enable easy communication
with all peers in your local ethernet broadcast domain, the
:ref:`Auto Interface<interfaces-auto>` performs better, and is just as
easy to use.
The below example is enabled by default on new Reticulum installations,
as it provides an easy way to get started and to test Reticulum on a
pre-existing LAN.
@@ -36,7 +344,7 @@ pre-existing LAN.
[[Default UDP Interface]]
type = UDPInterface
interface_enabled = True
outgoing = True
listen_ip = 0.0.0.0
listen_port = 4242
forward_ip = 255.255.255.255
@@ -44,9 +352,7 @@ pre-existing LAN.
# The above configuration will allow communication
# within the local broadcast domains of all local
# IP interfaces. This is enabled by default as an
# easy way to get started, but you might want to
# consider altering it to something more specific.
# IP interfaces.
# Instead of specifying listen_ip, listen_port,
# forward_ip and forward_port, you can also bind
@@ -74,64 +380,6 @@ pre-existing LAN.
# forward_ip = 10.55.0.16
# forward_port = 4242
.. _interfaces-tcps:
TCP Server Interface
====================
The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[TCP Server Interface]]
type = TCPServerInterface
interface_enabled = True
outgoing = True
# This configuration will listen on all IP
# interfaces on port 4242
listen_ip = 0.0.0.0
listen_port = 4242
# Alternatively you can bind to a specific IP
# listen_ip = 10.0.0.88
# listen_port = 4242
# Or a specific network device
# device = eth0
# port = 4242
.. _interfaces-tcpc:
TCP Client Interface
====================
To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.
.. code::
# Here's an example of a TCP Client interface. The
# target_host can either be an IP address or a hostname.
[[TCP Client Interface]]
type = TCPClientInterface
interface_enabled = True
outgoing = True
target_host = 127.0.0.1
target_port = 4242
.. _interfaces-rnode:
@@ -152,11 +400,6 @@ can be used, and offers full control over LoRa parameters.
# Enable interface if you want use it!
interface_enabled = True
# Allow transmit on interface. Setting
# this to false will create a listen-
# only interface.
outgoing = true
# Serial port for the device
port = /dev/ttyUSB0
@@ -207,7 +450,6 @@ directly over a wire-pair, or for using devices such as data radios and lasers.
[[Serial Interface]]
type = SerialInterface
interface_enabled = True
outgoing = True
# Serial port for the device
port = /dev/ttyUSB0
@@ -234,7 +476,6 @@ for station identification purposes.
[[Packet Radio KISS Interface]]
type = KISSInterface
interface_enabled = True
outgoing = true
# Serial port for the device
port = /dev/ttyUSB1
@@ -305,9 +546,6 @@ beaconing functionality described above.
# Enable interface if you want use it!
interface_enabled = True
# Allow transmit on interface.
outgoing = True
# Serial port for the device
port = /dev/ttyUSB2
@@ -339,4 +577,4 @@ beaconing functionality described above.
# Whether to use KISS flow-control.
# This is useful for modems with a
# small internal packet buffer.
flow_control = false
flow_control = false
+25 -5
View File
@@ -27,6 +27,11 @@ with Reticulum:
While the adress space can support billions of endpoints, Reticulum is
also very useful when just a few devices needs to communicate.
* | Low-bandwidth networks, like LoRa and packet radio, can interoperate and
interconnect with much larger and higher bandwidth networks without issue.
Reticulum automatically manages the flow of information to and from various
network segments, and when bandwidth is limited, local traffic is prioritised.
* | Reticulum provides sender/initiator anonymity by default. There is no way
to filter traffic or discriminate it based on the source of the traffic.
@@ -47,23 +52,38 @@ with Reticulum:
transport node. Letting every node be a transport node will in most cases
degrade the performance and reliability of the network.
In general terms, if a node is stationary, well-connected and kept running
*In general terms, if a node is stationary, well-connected and kept running
most of the time, it is a good candidate to be a transport node. For optimal
performance, a network should contain the amount of transport nodes that
provides connectivity to the intended area / topography, and not many more
than that.
than that.*
* | Reticulum is designed to work reliably in open, trustless environments. This
means you can use it to create open-access networks, where participants can
join and leave in an free and unorganised manner. This property allows an
entirely new, and so far, mostly unexplored class of networked applications,
where networks, and the information flow within them can form and dissolve
organically.
* | You can just as easily create closed networks, since Reticulum allows you to
add authentication to any interface. This means you can restrict access on
any interface type, even when using legacy devices, such as modems. You can
also mix authenticated and open interfaces on the same system. See the
:ref:`Common Interface Options<interfaces-options>` section of the :ref:`Interfaces<interfaces-main>`
chapter of this manual for information on how to set up interface authentication.
Reticulum allows you to mix very different kinds of networking mediums into a
unified mesh, or to keep everything within one medium. You could build a "virtual
network" running entirely over the Internet, where all nodes communicate over TCP
and UDP "channels". You could also build such a network using MQTT or ZeroMQ as
the underlying carrier for Reticulum.
and UDP "channels". You could also build such a network using other already-established
communications channels as the underlying carrier for Reticulum.
However, most real-world networks will probably involve either some form of
wireless or direct hardline communications. To allow Reticulum to communicate
over any type of medium, you must specify it in the configuration file, by default
located at ``~/.reticulum/config``.
located at ``~/.reticulum/config``. See the :ref:`Supported Interfaces<interfaces-main>`
chapter of this manual for interface configuration examples.
Any number of interfaces can be configured, and Reticulum will automatically
decide which are suitable to use in any given situation, depending on where
+285 -186
View File
@@ -3,20 +3,21 @@
***********************
Understanding Reticulum
***********************
This chapter will briefly describe the overall purpose and operating principles of Reticulum, a
networking stack designed for reliable and secure communication over high-latency, low-bandwidth
links. It should give you an overview of how the stack works, and an understanding of how to
This chapter will briefly describe the overall purpose and operating principles of Reticulum.
It should give you an overview of how the stack works, and an understanding of how to
develop networked applications using Reticulum.
This document is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the best place to go for such information is the Python reference implementation of Reticulum, along
with the code examples and API reference. It is however an essential resource to understanding the
general principles of Reticulum, how to apply them when creating your own networks or software.
This chapter is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the only complete repository, and final authority on how Reticulum actually functions, is the Python
reference implementation and API reference. That being said, this chapter is an essential resource in
understanding how Reticulum works from a high-level perspective, along with the general principles of
Reticulum, and how to apply them when creating your own networks or software.
After reading this document, you should be well-equipped to understand how a Reticulum network
operates, what it can achieve, and how you can use it yourself. If you want to help out with the
development, this is also the place to start, since it will provide a pretty clear overview of the
sentiments and the philosophy behind Reticulum.
sentiments and the philosophy behind Reticulum, what problems it seeks to solve, and how it
approaches those solutions.
.. _understanding-motivation:
@@ -25,34 +26,42 @@ Motivation
The primary motivation for designing and implementing Reticulum has been the current lack of
reliable, functional and secure minimal-infrastructure modes of digital communication. It is my
belief that it is highly desirable to create a cheap and reliable way to set up a wide-range digital
communication network that can securely allow exchange of information between people and
belief that it is highly desirable to create a reliable and efficient way to set up long-range digital
communication networks that can securely allow exchange of information between people and
machines, with no central point of authority, control, censorship or barrier to entry.
Almost all of the various networking systems in use today share a common limitation, namely that they
require large amounts of coordination and trust to work, and to join the networks you need approval
Almost all of the various networking systems in use today share a common limitation: They
require large amounts of coordination and centralised trust and power to function. To join such networks, you need approval
of gatekeepers in control. This need for coordination and trust inevitably leads to an environment of
central control, where it's very easy for infrastructure operators or governments to control or alter
traffic, and censor or persecute unwanted actors.
traffic, and censor or persecute unwanted actors. It also makes it completely impossible to freely deploy
and use networks at will, like one would use other common tools that enhance individual agency and freedom.
Reticulum aims to require as little coordination and trust as possible. In fact, the only
“coordination” required is to know the characteristics of physical medium carrying Reticulum traffic.
Reticulum aims to require as little coordination and trust as possible. It aims to make secure,
anonymous and permissionless networking and information exchange a tool that anyone can just pick up and use.
Since Reticulum is completely medium agnostic, this could be whatever is best suited to the situation.
In some cases, this might be 1200 baud packet radio links over VHF frequencies, in other cases it might
be a microwave network using off-the-shelf radios. At the time of release of this document, the
recommended setup for development and testing is using LoRa radio modules with an open source firmware
(see the section :ref:`Reference System Setup<understanding-referencesystem>`), connected to a small
computer like a Raspberry Pi. As an example, the default reference setup provides a channel capacity
of 5.4 Kbps, and a usable direct node-to-node range of around 15 kilometers (indefinitely extendable
by using multiple hops).
Since Reticulum is completely medium agnostic, it can be used to build networks on whatever is best
suited to the situation, or whatever you have available. In some cases, this might be packet radio
links over VHF frequencies, in other cases it might be a 2.4 GHz
network using off-the-shelf radios, or it might be using common LoRa development boards.
At the time of release of this document, the fastest and easiest setup for development and testing is using
LoRa radio modules with an open source firmware (see the section :ref:`Reference Setup<understanding-referencesystem>`),
connected to any kind of computer or mobile device that Reticulum can run on.
The ultimate aim of Reticulum is to allow anyone to be their own network operator, and to make it
cheap and easy to cover vast areas with a myriad of independent, interconnectable and autonomous networks.
Reticulum **is not** *one network*, it **is a tool** to build *thousands of networks*.
Networks without kill-switches, surveillance, censorship and control. Networks that can freely interoperate, associate and disassociate
with each other, and require no central oversight. Networks for human beings. *Networks for the people*.
.. _understanding-goals:
Goals
=====
To be as widely usable and easy to use as possible, the following goals have been used to
To be as widely usable and efficient to deploy as possible, the following goals have been used to
guide the design of Reticulum:
@@ -60,31 +69,34 @@ guide the design of Reticulum:
Reticulum must be implemented with, and be able to run using only open source software. This is
critical to ensuring the availability, security and transparency of the system.
* **Hardware layer agnosticism**
Reticulum shall be fully hardware agnostic, and shall be useable over a wide range
Reticulum must be fully hardware agnostic, and shall be useable over a wide range of
physical networking layers, such as data radios, serial lines, modems, handheld transceivers,
wired ethernet, wifi, or anything else that can carry a digital data stream. Hardware made for
dedicated Reticulum use shall be as cheap as possible and use off-the-shelf components, so
it can be easily replicated.
it can be easily modified and replicated by anyone interested in doing so.
* **Very low bandwidth requirements**
Reticulum should be able to function reliably over links with a transmission capacity as low
as *1,000 bps*.
as *500 bits per second*.
* **Encryption by default**
Reticulum must use encryption by default where possible and applicable.
Reticulum must use strong encryption by default for all communication.
* **Initiator Anonymity**
It must be possible to communicate over a Reticulum network without revealing any identifying
information about oneself.
* **Unlicensed use**
Reticulum shall be functional over physical communication mediums that do not require any
form of license to use. Reticulum must be designed in a way, so it is usable over ISM radio
frequency bands, and can provide functional long distance links in such conditions, for example
by connecting a modem to a PMR or CB radio, or by using LoRa or WiFi modules.
* **Supplied software**
Apart from the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic communication suite using Reticulum must be
implemented and released at the same time as Reticulum itself. This shall serve both as a
functional communication suite, and as an example and learning resource to others wishing
In addition to the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic set of Reticulum-based communication tools must be
implemented and released along with Reticulum itself. These shall serve both as a
functional, basic communication suite, and as an example and learning resource to others wishing
to build applications with Reticulum.
* **Ease of use**
The reference implementation of Reticulum is written in Python, to make it easy to use
and understand. A programmer with only basic experience should be able to use
Reticulum in their own applications.
Reticulum to write networked applications.
* **Low cost**
It shall be as cheap as possible to deploy a communication system based on Reticulum. This
should be achieved by using cheap off-the-shelf hardware that potential users might already
@@ -99,27 +111,36 @@ Introduction & Basic Functionality
Reticulum is a networking stack suited for high-latency, low-bandwidth links. Reticulum is at its
core a *message oriented* system. It is suited for both local point-to-point or point-to-multipoint
scenarios where alle nodes are within range of each other, as well as scenarios where packets need
to be transported over multiple hops to reach the recipient.
to be transported over multiple hops in a complex network to reach the recipient.
Reticulum does away with the idea of addresses and ports known from IP, TCP and UDP. Instead
Reticulum uses the singular concept of *destinations*. Any application using Reticulum as its
networking stack will need to create one or more destinations to receive data, and know the
destinations it needs to send data to.
All destinations in Reticulum are represented internally as 10 bytes, derived from truncating a full
All destinations in Reticulum are represented as a 10 byte hash, derived from truncating a full
SHA-256 hash of identifying characteristics of the destination. To users, the destination addresses
will be displayed as 10 bytes in hexadecimal representation, as in the following example: ``<80e29bf7cccaf31431b3>``.
By default Reticulum encrypts all data using public-key cryptography. Any message sent to a
destination is encrypted with that destinations public key. Reticulum can also set up an encrypted
channel to a destination with *Perfect Forward Secrecy* and *Initiator Anonymity* using a elliptic
The truncation size of 10 bytes (80 bits) for destinations has been choosen as a reasonable tradeoff between address space
and packet overhead. The address space accomodated by this size can support many billions of
simultaneously active devices on the same network, while keeping packet overhead low, which is
essential on low-bandwidth networks. In the very unlikely case that this address space nears
congestion, a one-line code change can upgrade the Reticulum address space all the way up to 256
bits, ensuring the Reticulum address space could potentially support galactic-scale networks.
This is obviusly complete and ridiculous over-allocation, and as such, the current 80 bits should
be sufficient, even far into the future.
By default Reticulum encrypts all data using elliptic curve cryptography. Any packet sent to a
destination is encrypted with a derived ephemeral key. Reticulum can also set up an encrypted
channel to a destination with *Forward Secrecy* and *Initiator Anonymity* using a elliptic
curve cryptography and ephemeral keys derived from a Diffie Hellman exchange on Curve25519. In
Reticulum terminology, this is called a *Link*.
Reticulum terminology, this is called a *Link*. The multi-hop transport, coordination, verification
and reliability layers are fully autonomous and also based on elliptic curve cryptography.
Reticulum also offers symmetric key encryption for group-oriented communications, as well as
unencrypted packets for broadcast purposes, or situations where you need the communication to be in
plain text. The multi-hop transport, coordination, verification and reliability layers are fully
autonomous and based on public key cryptography.
plain text.
Reticulum can connect to a variety of interfaces such as radio modems, data radios and serial ports,
and offers the possibility to easily tunnel Reticulum traffic over IP links such as the Internet or
@@ -135,22 +156,30 @@ destinations. Reticulum uses three different basic destination types, and one sp
* **Single**
The *single* destination type defines a public-key encrypted destination. Any data sent to this
destination will be encrypted with the destinations public key, and will only be readable by
the creator of the destination.
* **Group**
The *group* destination type defines a symmetrically encrypted destination. Data sent to this
destination will be encrypted with a symmetric key, and will be readable by anyone in
possession of the key. The *group* destination can be used just as well by only two peers, as it
can by many.
The *single* destination type is the most common type in Reticulum, and should be used for
most purposes. It is always identified by a unique public key. Any data sent to this
destination will be encrypted using ephemeral keys derived from an ECDH key exchange, and will
only be readable by the creator of the destination, who holds the corresponding private key.
* **Plain**
A *plain* destination type is unencrypted, and suited for traffic that should be broadcast to a
number of users, or should be readable by anyone. Traffic to a *plain* destination is not encrypted.
Generally, *plain* destinations can be used for broadcast information intended to be public.
Plain destinations are only reachable directly, and packets adressed to plain destinations are
never transported over multiple hops in the network. To be transportable over multiple hops in Reticulum, information
*must* be encrypted, since Reticulum uses the per-packet encryption to verify routing paths and
keep them alive.
* **Group**
The *group* special destination type, that defines a symmetrically encrypted virtual destination.
Data sent to this destination will be encrypted with a symmetric key, and will be readable by
anyone in possession of the key, but as with the *plain* destination type, packets to this type
of destination are not currently transported over multiple hops, although a planned upgrade
to Reticulum will allow globally reachable *group* destinations.
* **Link**
A *link* is a special destination type, that serves as an abstract channel to a *single*
destination, directly connected or over multiple hops. The *link* also offers reliability and
more efficient encryption, forward secrecy, initiator anonymity, and as such can be useful even
when a node is directly reachable.
when a node is directly reachable. It also offers a more capable API and allows easily carrying
out requests and responses, large data transfers and more.
.. _understanding-destinationnaming:
@@ -192,7 +221,7 @@ packet.
In actual use of *single* destination naming, it is advisable not to use any uniquely identifying
features in aspect naming. Aspect names should be general terms describing what kind of destination
is represented. The uniquely identifying aspect is always acheived by the appending the public key,
which expands the destination into a uniquely identifyable one.
which expands the destination into a uniquely identifyable one. Reticulum does this automatically.
Any destination on a Reticulum network can be addressed and reached simply by knowning its
destination hash (and public key, but if the public key is not known, it can be requested from the
@@ -208,30 +237,32 @@ To recap, the different destination types should be used in the following situat
When private communication between two or more endpoints is needed. Supports multiple hops
indirectly, but must first be established through a *single* destination.
* **Plain**
When plain-text communication is desirable, for example when broadcasting information.
When plain-text communication is desirable, for example when broadcasting information, or for local discovery purposes.
To communicate with a *single* destination, you need to know its public key. Any method for
obtaining the public key is valid, but Reticulum includes a simple mechanism for making other
nodes aware of your destinations public key, called the *announce*. It is also possible to request
an unknown public key from the network, as all participating nodes serve as a distributed ledger
an unknown public key from the network, as all transport instances serve as a distributed ledger
of public keys.
Note that public key information can be shared and verified in many other ways than using the
built-in *announce* functionality, and that it is therefore not required to use the announce/request
Note that public key information can be shared and verified in other ways than using the
built-in *announce* functionality, and that it is therefore not required to use the *announce* and *path request*
functionality to obtain public keys. It is by far the easiest though, and should definitely be used
if there is not a good reason for doing it differently.
if there is not a very good reason for doing it differently.
.. _understanding-keyannouncements:
Public Key Announcements
------------------------
An *announce* will send a special packet over any configured interfaces, containing all needed
An *announce* will send a special packet over any relevant interfaces, containing all needed
information about the destination hash and public key, and can also contain some additional,
application specific data. The entire packet is signed by the sender to ensure authenticity. It is not
required to use the announce functionality, but in many cases it will be the simplest way to share
public keys on the network. As an example, an announce in a simple messenger application might
contain the following information:
public keys on the network. The announce mechanism also serves to establish end-to-end connectivity
to the announced destination, as the announce propagates through the network.
As an example, an announce in a simple messenger application might contain the following information:
* The announcers destination hash
@@ -244,13 +275,21 @@ With this information, any Reticulum node that receives it will be able to recon
destination to securely communicate with that destination. You might have noticed that there is one
piece of information lacking to reconstruct full knowledge of the announced destination, and that is
the aspect names of the destination. These are intentionally left out to save bandwidth, since they
will be implicit in almost all cases. If a destination name is not entirely implicit, information can be
included in the application specific data part that will allow the receiver to infer the naming.
will be implicit in almost all cases. The receiving application will already know them. If a destination
name is not entirely implicit, information can be included in the application specific data part that
will allow the receiver to infer the naming.
It is important to note that announces will be forwarded throughout the network according to a
certain pattern. This will be detailed in the section
:ref:`The Announce Mechanism in Detail<understanding-announce>`.
In Reticulum, destinations are allowed to move around the network at will. This is very different from
protocols such as IP, where an address is always expected to stay within the network segment it was assigned in.
This limitation does not exist in Reticulum, and any destination is *completely portable* over the entire topography
of the network, and *can even be moved to other Reticulum networks* than the one it was created in, and
still become reachable. To update it's reachability, a destination simply needs to send an announce on any
networks it is part of. After a short while, it will be globally reachable in the network.
Seeing how *single* destinations are always tied to a private/public key pair leads us to the next topic.
.. _understanding-identities:
@@ -259,21 +298,22 @@ Identities
----------
In Reticulum, an *identity* does not necessarily represent a personal identity, but is an abstraction that
can represent any kind of *verified entity*. This could very well be a person, but it could also be the
can represent any kind of *verifiable entity*. This could very well be a person, but it could also be the
control interface of a machine, a program, robot, computer, sensor or something else entirely. In
general, any kind of agent that can act, or be acted upon, or store or manipulate information, can be
represented as an identity.
represented as an identity. An *identity* can be used to create any number of destinations.
As we have seen, a *single* destination will always have an *identity* tied to it, but not *plain* or *group*
A *single* destination will always have an *identity* tied to it, but not *plain* or *group*
destinations. Destinations and identities share a multilateral connection. You can create a
destination, and if it is not connected to an identity upon creation, it will just create a new one to use
automatically. This may be desirable in some situations, but often you will probably want to create
the identity first, and then link it to created destinations.
the identity first, and then use it to create new destinations.
Building upon the simple messenger example, we could use an identity to represent the user of the
application. Destinations created will then be linked to this identity to allow communication to
reach the user. In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately.
As an example, we could use an identity to represent the user of a messaging application.
Destinations can then be created by this identity to allow communication to reach the user.
In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately, since obtaining access to the identity keys equals
obtaining access and controlling reachability to any destinations created by that identity.
.. _understanding-gettingfurther:
@@ -292,57 +332,73 @@ In the following sections, two concepts that allow this will be introduced, *pat
Reticulum Transport
===================
The term routing has been purposefully avoided until now. The current methods of routing used in IP-based
networks are fundamentally incompatible with the physical link types that Reticulum was designed to handle.
These routing methodologies assume trust at the physical layer, and often needs a lot more bandwidth than
Reticulum can assume is available.
The methods of routing used in traditional networks are fundamentally incompatible with the physical medium
types and circumstances that Reticulum was designed to handle. These mechanisms mostly assume trust at the physical layer,
and often needs a lot more bandwidth than Reticulum can assume is available. Since Reticulum is designed to
survive running over open radio spectrum, no such trust can be assumed, and bandwidth is often very limited.
Since Reticulum is designed to run over open radio spectrum, no such trust exists, and bandwidth is often
very limited. Existing routing protocols like BGP or OSPF carry too much overhead to be practically
useable over bandwidth-limited, high-latency links.
To overcome such challenges, Reticulums *Transport* system uses public-key cryptography to
implement the concept of *paths* that allow discovery of how to get information to a certain
To overcome such challenges, Reticulums *Transport* system uses asymmetric elliptic curve cryptography to
implement the concept of *paths* that allow discovery of how to get information closer to a certain
destination. It is important to note that no single node in a Reticulum network knows the complete
path to a destination. Every Transport node participating in a Reticulum network will only
know what the most direct way to get a packet one hop closer to it's destination is.
know the most direct way to get a packet one hop closer to it's destination.
.. _understanding-nodetypes:
Node Types
----------
Currently, Reticulum distinguishes between two types of network nodes. All nodes on a Reticulum network
are *Reticulum Instances*, and some are alo *Transport Nodes*. If a system running Reticulum is fixed in
one place, and is intended to be kept available most of the time, it is a good contender to be a *Transport Node*.
Any Reticulum Instance can become a Transport Node by enabling it in the configuration.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for wider connectivity.
If a node is an *Instance* it should be given the configuration directive ``enable_transport = No``, which
is the default setting.
If it is a *Transport Node*, it should be given the configuration directive ``enable_transport = Yes``.
.. _understanding-announce:
The Announce Mechanism in Detail
--------------------------------
When an *announce* is transmitted by a node, it will be forwarded by any node receiving it, but
according to some specific rules:
When an *announce* for a destination is transmitted by from a Reticulum instance, it will be forwarded by
any transport node receiving it, but according to some specific rules:
* | If this exact announce has already been received before, ignore it.
* | If not, record into a table which node the announce was received from, and how many times in
* | If not, record into a table which Transport Node the announce was received from, and how many times in
total it has been retransmitted to get here.
* | If the announce has been retransmitted *m+1* times, it will not be forwarded. By default, *m* is
set to 18.
* | If the announce has been retransmitted *m+1* times, it will not be forwarded any more. By default, *m* is
set to 128.
* | The announce will be assigned a delay *d* = c\ :sup:`h` seconds, where *c* is a decay constant, and *h* is the amount of times this packet has already been forwarded.
* | After a randomised delay, the announce will be retransmitted on all interfaces that have bandwidth
available for processing announces. By default, the maximum bandwidth allocation for processing
announces is set at 2%, but can be configured on a per-interface basis.
* | The packet will be given a priority *p = 1/d*.
* | If any given interface does not have enough bandwidth available for retransmitting the announce,
the announce will be assigned a priority inversely proportional to it's hop count, and be inserted
into a queue managed by the interface.
* | If at least *d* seconds has passed since the announce was received, and no other packets with a
priority higher than *p* are waiting in the queue (see Packet Prioritisation), and the channel is
not utilized by other traffic, the announce will be forwarded.
* | When the interface has bandwidth available for processing an announce, it will prioritise announces
for destinations that are closest in terms of hops, thus prioritising reachability and connectivity
of local nodes, even on slow networks that connect to wider and faster networks.
* | If no other nodes are heard retransmitting the announce with a greater hop count than when
it left this node, transmitting it will be retried *r* times. By default, *r* is set to 1. Retries
follow same rules as above, with the exception that it must wait for at least *d* = c\ :sup:`h+1` +
t + rand(0, rw) seconds. This amount of time is equal to the amount of time it would take the next
node to retransmit the packet, plus a random window. By default, *t* is set to 10 seconds, and the
random window *rw* is set to 10 seconds.
* | After the announce has been re-transmitted, and if no other nodes are heard retransmitting the announce
with a greater hop count than when it left this node, transmitting it will be retried *r* times. By default,
*r* is set to 1.
* | If a newer announce from the same destination arrives, while an identical one is already in
the queue, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce, but will use *d* and *p* of the old
announce.
* | If a newer announce from the same destination arrives, while an identical one is already waiting
to be transmitted, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce.
Once an announce has reached a node in the network, any other node in direct contact with that
node will be able to reach the destination the announce originated from, simply by sending a packet
@@ -350,11 +406,16 @@ addressed to that destination. Any node with knowledge of the announce will be a
packet towards the destination by looking up the next node with the shortest amount of hops to the
destination.
According to these rules and default constants, an announce will propagate throughout the network
in a predictable way. In an example network utilising the default constants, and with an average link
distance of *Lavg =* 15 kilometers, an announce will be able to propagate outwards to a radius of 180
kilometers in 34 minutes, and a *maximum announce radius* of 270 kilometers in approximately 3
days.
According to these rules, an announce will propagate throughout the network in a predictable way,
and make the announced destination reachable in a short amount of time. Fast networks that have the
capacity to process many announces can reach full convergence very quickly, even when constantly adding
new destinations. Slower segments of such networks might take a bit longer to gain full knowledge about
the wide and fast networks they are connected to, but can still do so over time, while prioritising full
and quickly converging end-to-end connectivity for their local, slower segments.
In general, even extremely complex networks, that utilize the maximum 128 hops will converge to full
end-to-end connectivity in about one minute, given there is enough bandwidth available to process
the required amount of announces.
.. _understanding-paths:
@@ -401,7 +462,7 @@ For exchanges of small amounts of information, Reticulum offers the *Packet* API
For exchanges of larger amounts of data, or when longer sessions of bidirectional communication is desired, Reticulum offers the *Link* API. To establish a *link*, the following process is employed:
* | First, the node that wishes to establish a link will send out a special packet, that
traverses the network and locates the desired destination. Along the way, the nodes that
traverses the network and locates the desired destination. Along the way, the Transport Nodes that
forward the packet will take note of this *link request*.
* | Second, if the destination accepts the *link request* , it will send back a packet that proves the
@@ -412,15 +473,19 @@ For exchanges of larger amounts of data, or when longer sessions of bidirectiona
* | When the validity of the *link* has been accepted by forwarding nodes, these nodes will
remember the *link* , and it can subsequently be used by referring to a hash representing it.
* | As a part of the *link request* , a Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes, using elliptic curve cryptography. As such,
this mode of communication is preferred, even for situations when nodes can directly communicate,
when the amount of data to be exchanged numbers in the tens of packets.
* | As a part of the *link request*, an Elliptic Curve Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes. As such, this mode of communication is preferred,
even for situations when nodes can directly communicate, when the amount of data to be exchanged numbers
in the tens of packets, or whenever the use of the more advanced API functions is desired.
* | When a *link* has been set up, it automatically provides message receipt functionality, through
the same *proof* mechanism discussed before, so the sending node can obtain verified confirmation
that the information reached the intended recipient.
* | Once the *link* has been set up, the initiator can remain anonymous, or choose to authenticate towards
the destination using a Reticulum Identity. This authentication is happening inside the encrypted
link, and is only revealed to the verified destination, and no intermediaries.
In a moment, we will discuss the details of how this methodology is implemented, but lets first
recap what purposes this methodology serves. We first ensure that the node answering our request
is actually the one we want to communicate with, and not a malicious actor pretending to be so.
@@ -507,7 +572,7 @@ the transfer is needed.
This is the purpose of the Reticulum :ref:`Resource<api-resource>`. A *Resource* can automatically
handle the reliable transfer of an arbitrary amount of data over an established :ref:`Link<api-link>`.
Resources can auto-compress data, will handle breaking the data into individual packets, sequencing
the transfer and reassembling the data on the other end.
the transfer, integrity verification and reassembling the data on the other end.
:ref:`Resources<api-resource>` are programmatically very simple to use, and only requires a few lines
of codes to reliably transfer any amount of data. They can be used to transfer data stored in memory,
@@ -515,57 +580,62 @@ or stream data directly from files.
.. _understanding-referencesystem:
Reference System Setup
Reference Setup
======================
This section will detail the recommended *Reference System Setup* for Reticulum. It is important to
note that Reticulum is designed to be usable over more or less any medium that allows you to send
and receive data in a digital form, and satisfies some very low minimum requirements. The
communication channel must support at least half-duplex operation, and provide an average
throughput of around 1000 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum software should be able to run on more or less any hardware that can provide a Python 3.x
This section will detail a recommended *Reference Setup* for Reticulum. It is important to
note that Reticulum is designed to be usable on more or less any computing device, and over more
or less any medium that allows you to send and receive data, which satisfies some very low
minimum requirements.
The communication channel must support at least half-duplex operation, and provide an average
throughput of around 500 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum stack should be able to run on more or less any hardware that can provide a Python 3.x
runtime environment.
That being said, the reference setup has been outlined to provide a common platform for anyone
That being said, this reference setup has been outlined to provide a common platform for anyone
who wants to help in the development of Reticulum, and for everyone who wants to know a
recommended setup to get started. A reference system consists of three parts:
recommended setup to get started experimenting. A reference system consists of three parts:
* **A channel access device**
Or *CAD* , in short, provides access to the physical medium whereupon the communication
* **An Interface Device**
Which provides access to the physical medium whereupon the communication
takes place, for example a radio with an integrated modem. A setup with a separate modem
connected to a radio would also be termed a “channel access device.
* **A host device**
Some sort of computing device that can run the necessary software, communicates with the
channel access device, and provides user interaction.
* **A software stack**
connected to a radio would also be an interface device.
* **A Host Device**
Some sort of computing device that can run the necessary software, communicate with the
interface device, and provide user interaction.
* **A Software Stack**
The software implementing the Reticulum protocol and applications using it.
The reference setup can be considered a relatively stable platform to develop on, and also to start
building networks on. While details of the implementation might change at the current stage of
building networks or applications on. While details of the implementation might change at the current stage of
development, it is the goal to maintain hardware compatibility for as long as entirely possible, and
the current reference setup has been determined to provide a functional platform for many years
into the future. The current Reference System Setup is as follows:
* **Channel Access Device**
* **Interface Device**
A data radio consisting of a LoRa radio module, and a microcontroller with open source
firmware, that can connect to host devices via USB. It operates in either the 430, 868 or 900
MHz frequency bands. More details can be found on the `RNode Page <https://unsigned.io/rnode>`_.
* **Host device**
* **Host Device**
Any computer device running Linux and Python. A Raspberry Pi with a Debian based OS is
recommended.
* **Software stack**
The current Reference Implementation Release of Reticulum, running on a Debian based
* **Software Stack**
The most recently released Python Implementation of Reticulum, running on a Debian based
operating system.
It is very important to note, that the reference channel access device **does not** use the LoRaWAN
standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an MCU with the correct firmware. Full details on how to
To avoid confusion, it is very important to note, that the reference interface device **does not**
use the LoRaWAN standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an controller with the correct firmware. Full details on how to
get or make such a device is available on the `RNode Page <https://unsigned.io/rnode>`_.
With the current reference setup, it should be possible to get on a Reticulum network for around 100$
even if you have none of the hardware already, and need to purchase everything.
This reference setup is of course just a recommendation for getting started easily, and you should
tailor it to your own specific needs, or whatever hardware you have available.
.. _understanding-protocolspecifics:
Protocol Specifics
@@ -576,19 +646,6 @@ Reticulum, but non critical in understanding how the protocol works on a general
treated more as a reference than as essential reading.
Node Types
----------
Currently Reticulum defines two node types, the *Station* and the *Peer*. A node is a *station* if it fixed
in one place, and if it is intended to be kept online most of the time. Otherwise the node is a *peer*.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for connectivity.
If a node is a *Peer* it should be given the configuration directive ``enable_transport = No``.
If it is a *Station*, it should be given the configuration directive ``enable_transport = Yes``.
Packet Prioritisation
---------------------
@@ -596,15 +653,30 @@ Currently, Reticulum is completely priority-agnostic regarding general traffic.
on a first-come, first-serve basis. Announce re-transmission are handled according to the re-transmission
times and priorities described earlier in this chapter.
It is possible that a prioritisation engine could be added to Reticulum in the future, but in
the light of Reticulums goal of equal access, doing so would need to be the subject of careful
investigation of the consequences first.
Interface Access Codes
----------------------
Reticulum can create named virtual networks, and networks that are only accessible by knowing a preshared
passphrase. The configuration of this is detailed in the :ref:`Common Interface Options<interfaces-options>`
section. To implement these feature, Reticulum uses the concept of Interface Access Codes, that are calculated
and verified per packet.
An interface with a named virtual network or passphrase authentication enabled will derive a shared Ed25519
signing identity, and for every outbound packet generate a signature of the entire packet. This signature is
then inserted into the packet as an Interface Access Code before transmission. Depending on the speed and
capabilities of the interface, the IFAC can be the full 512-bit Ed25519 signature, or a truncated version.
Configured IFAC length can be inspected for all interfaces with the ``rnstatus`` utility.
Upon receipt, the interface will check that the signature matches the expected value, and drop the packet if it
does not. This ensures that only packets sent with the correct naming and/or passphrase parameters are allowed to
pass onto the network.
.. _understanding-packetformat:
Binary Packet Format
--------------------
Wire Format
-----------
.. code-block:: text
@@ -615,9 +687,14 @@ Binary Packet Format
[HEADER 2 bytes] [ADDRESSES 10/20 bytes] [CONTEXT 1 byte] [DATA 0-477 bytes]
* The HEADER field is 2 bytes long.
* Byte 1: [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 1: [IFAC Flag], [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 2: Number of hops
* Interface Access Code field if the IFAC flag was set.
* The length of the Interface Access Code can vary from
1 to 64 bytes according to physical interface
capabilities and configuration.
* The ADDRESSES field contains either 1 or 2 addresses.
* Each address is 10 bytes long.
* The Header Type flag in the HEADER field determines
@@ -630,12 +707,16 @@ Binary Packet Format
* The DATA field is between 0 and 477 bytes.
* It contains the packets data payload.
IFAC Flag
-----------------
open 0 Packet for publically accessible interface
authenticated 1 Interface authentication is included in packet
Header Types
-----------------
type 1 00 Two byte header, one 10 byte address field
type 2 01 Two byte header, two 10 byte address fields
type 3 10 Reserved
type 4 11 Reserved
type 1 0 Two byte header, one 10 byte address field
type 2 1 Two byte header, two 10 byte address fields
Propagation Types
@@ -664,42 +745,60 @@ Binary Packet Format
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELDS CONTEXT FIELD DATA FIELD
_______|_______ ________________|________________ ________|______ __|_
| | | | | | | |
01010000 00000100 [ADDR1, 10 bytes] [ADDR2, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 4
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = TRANSPORT
+------------- Header Type = HEADER_2 (two byte header, two address fields)
01010000 00000100 [HASH1, 10 bytes] [HASH2, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 4
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = TRANSPORT
|+------------- Header Type = HEADER_2 (two byte header, two address fields)
+-------------- Access Codes = DISABLED
+- Packet Example -+
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ _______|_______ ________|______ __|_
| | | | | | | |
00000000 00000111 [ADDR1, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 7
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = BROADCAST
+------------- Header Type = HEADER_1 (two byte header, one address field)
00000000 00000111 [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = DISABLED
Size examples of different packet types
---------------------------------------
+- Packet Example -+
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size including all fields.
HEADER FIELD IFAC FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ ______|______ _______|_______ ________|______ __|_
| | | | | | | | | |
10000000 00000111 [IFAC, N bytes] [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = ENABLED
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
Size examples of different packet types
---------------------------------------
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size counting all fields including headers,
but excluding any interface access codes.
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
+142 -22
View File
@@ -6,16 +6,18 @@ Using Reticulum on Your System
Reticulum is not installed as a driver or kernel module, as one might expect
of a networking stack. Instead, Reticulum is distributed as a Python module.
This means that no special privileges are required to install or use it.
This means that no special privileges are required to install or use it. It
is also very light-weight, and easy to transfer to and install on new systems.
Any program or application that uses Reticulum will automatically load and
initialise Reticulum when it starts.
In many cases, this approach is sufficient. When any program needs to use
Reticulum, it is loaded, initialised, interfaces are brought up, and the
program can now communicate over Reticulum. If another program starts up
and also wants access to the same Reticulum network, the instance is simply
shared. This works for any number of programs running concurrently, and is
very easy to use, but depending on your use case, there are other options.
program can now communicate over any Reticulum networks available. If another
program starts up and also wants access to the same Reticulum network, the
instance is simply shared. This works for any number of programs running
concurrently, and is very easy to use, but depending on your use case, there
are other options.
Included Utility Programs
-------------------------
@@ -57,6 +59,7 @@ the same system.
-q, --quiet
--version show program's version number and exit
You can easily add ``rnsd`` as an always-on service by :ref:`configuring a service<using-systemd>`.
The rnstatus Utility
====================
@@ -71,33 +74,49 @@ interfaces, similar to the ``ifconfig`` program.
# Example output
Shared Instance[37428]
Status: Up
Connected applications: 1
RX: 1.13 KB
TX: 1.07 KB
Status : Up
Serving : 1 program
Rate : 1.00 Gbps
Traffic : 83.13 KB
86.10 KB↓
UDPInterface[Default UDP Interface/0.0.0.0:4242]
Status: Up
RX: 1.01 KB
TX: 1.01 KB
AutoInterface[Local]
Status : Up
Mode : Full
Rate : 10.00 Mbps
Peers : 1 reachable
Traffic : 63.23 KB↑
80.17 KB↓
TCPInterface[RNS Testnet Frankfurt/frankfurt.rns.unsigned.io:4965]
Status: Up
RX: 1.37 KB
TX: 9.02 KB
Status : Up
Mode : Full
Rate : 10.00 Mbps
Traffic : 187.27 KB↑
74.17 KB↓
RNodeInterface[RNode UHF]
Status : Up
Mode : Access Point
Rate : 1.30 kbps
Access : 64-bit IFAC by <…e702c42ba8>
Traffic : 8.49 KB↑
9.23 KB↓
Reticulum Transport Instance <5245a8efe1788c6a70e1> running
.. code:: text
usage: rnsd [-h] [--config CONFIG] [-v] [-q] [--version]
usage: rnstatus [-h] [--config CONFIG] [--version] [-a] [-v]
Reticulum Network Stack Daemon
Reticulum Network Stack Status
optional arguments:
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
-v, --verbose
-q, --quiet
--version show program's version number and exit
-a, --all show all interfaces
-v, --verbose
The rnpath Utility
@@ -116,7 +135,8 @@ destinations on the Reticulum network.
.. code:: text
usage: rnpath.py [-h] [--config CONFIG] [--version] [-v] [destination]
usage: rnpath [-h] [--config CONFIG] [--version] [-t] [-d] [-w seconds] [-v]
[destination]
Reticulum Path Discovery Utility
@@ -127,6 +147,9 @@ destinations on the Reticulum network.
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
--version show program's version number and exit
-t, --table show all known paths
-d, --drop remove the path to a destination
-w seconds timeout before giving up
-v, --verbose
@@ -162,4 +185,101 @@ destinations will not have this option enabled, and will not be probable.
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
--version show program's version number and exit
-v, --verbose
-v, --verbose
Improving System Configuration
------------------------------
If you are setting up a system for permanent use with Reticulum, there is a
few system configuration changes that can make this easier to administrate.
These changes will be detailed here.
Fixed Serial Port Names
=======================
On a Reticulum instance with several serial port based interfaces, it can be
beneficial to use the fixed device names for the serial ports, instead
of the dynamically allocated shorthands such as ``/dev/ttyUSB0``. Under most
Debian-based distributions, including Ubuntu and Raspberry Pi OS, these nodes
can be found under ``/dev/serial/by-id``.
You can use such a device path directly in place of the numbered shorthands.
Here is an example of a packet radio TNC configured as such:
.. code:: text
[[Packet Radio KISS Interface]]
type = KISSInterface
interface_enabled = True
outgoing = true
port = /dev/serial/by-id/usb-FTDI_FT230X_Basic_UART_43891CKM-if00-port0
speed = 115200
databits = 8
parity = none
stopbits = 1
preamble = 150
txtail = 10
persistence = 200
slottime = 20
Using this methodology avoids potential naming mix-ups where physical devices
might be plugged and unplugged in different orders, or when device name
assignment varies from one boot to another.
.. _using-systemd:
Reticulum as a System Service
=============================
Instead of starting Reticulum manually, you can install ``rnsd`` as a system
service and have it start automatically at boot.
If you installed Reticulum with ``pip``, the ``rnsd`` program will most likely
be located in a user-local installation path only, which means ``systemd`` will not
be able to execute it. In this case, you can simply symlink the ``rnsd`` program
into a directory that is in systemd's path:
.. code:: text
sudo ln -s $(which rnsd) /usr/local/bin/
You can then create the service file ``/etc/systemd/system/rnsd.service`` with the
following content:
.. code:: text
[Unit]
Description=Reticulum Network Stack Daemon
After=multi-user.target
[Service]
# If you run Reticulum on WiFi devices,
# or other devices that need some extra
# time to initialise, you might want to
# add a short delay before Reticulum is
# started by systemd:
# ExecStartPre=/bin/sleep 10
Type=simple
Restart=always
RestartSec=3
User=USERNAMEHERE
ExecStart=rnsd --service
[Install]
WantedBy=multi-user.target
Be sure to replace ``USERNAMEHERE`` with the user you want to run ``rnsd`` as.
To manually start ``rnsd`` run:
.. code:: text
sudo systemctl start rnsd
If you want to automatically start ``rnsd`` at boot, run:
.. code:: text
sudo systemctl enable rnsd
+45 -21
View File
@@ -2,9 +2,9 @@
What is Reticulum?
******************
Reticulum is a cryptography-based networking stack for wide-area networks built on readily available hardware, and can operate even with very high latency and extremely low bandwidth.
Reticulum is a cryptography-based networking stack for building wide-area networks with readily available hardware, that can continue to operate even with extremely low bandwidth and very high latency.
Reticulum allows you to build very wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.
Reticulum allows you to build wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.
Reticulum is a complete networking stack, and does not need IP or higher layers, although it is easy to utilise IP (with TCP or UDP) as the underlying carrier for Reticulum. It is therefore trivial to tunnel Reticulum over the Internet or private IP networks. Reticulum is built directly on cryptographic principles, allowing resilience and stable functionality in open and trustless networks.
@@ -13,12 +13,7 @@ No kernel modules or drivers are required. Reticulum runs completely in userland
Current Status
==============
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered relatively stable at the moment, but could change if warranted.
Caveat Emptor
==============
Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered stable at the moment, but could change if absolutely warranted.
What does Reticulum Offer?
@@ -27,9 +22,11 @@ What does Reticulum Offer?
* Fully self-configuring multi-hop routing
* Asymmetric X25519 encryption and Ed25519 signatures as a basis for all communication
* Complete initiator anonymity, communicate without revealing your identity
* Forward Secrecy with ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
* Asymmetric encryption based on X25519, and Ed25519 signatures as a basis for all communication
* Forward Secrecy by using ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
* Reticulum uses the `Fernet <https://github.com/fernet/spec/blob/master/Spec.md>`_ specification for on-the-wire / over-the-air encryption
@@ -47,6 +44,12 @@ What does Reticulum Offer?
* An intuitive and developer-friendly API
* Efficient link establishment
* Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes
* Low cost of keeping links open at only 0.62 bits per second
* Reliable and efficient transfer of arbritrary amounts of data
* Reticulum can handle a few bytes of data or files of many gigabytes
@@ -55,25 +58,23 @@ What does Reticulum Offer?
* The API is very easy to use, and provides transfer progress
* Efficient link establishment
* Authentication and virtual network segmentation on all supported interface types
* Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes
* Low cost of keeping links open at only 0.62 bits per second
* Flexible scalability allowing extremely low-bandwidth networks to co-exist and interoperate with large, high-bandwidth networks
Where can Reticulum be Used?
============================
Over practically any medium that can support at least a half-duplex channel
with 1.000 bits per second throughput, and an MTU of 500 bytes. Data radios,
with 500 bits per second throughput, and an MTU of 500 bytes. Data radios,
modems, LoRa radios, serial lines, AX.25 TNCs, amateur radio digital modes,
ad-hoc WiFi, free-space optical links and similar systems are all examples
of the types of interfaces Reticulum was designed for.
An open-source LoRa-based interface called `RNode <https://unsigned.io/rnode>`_
has been designed specifically for use with Reticulum. It is possible to build
yourself, or it can be purchased as a complete transceiver that just needs a
USB connection to the host.
has been designed as an example transceiver that is very suitable for
Reticulum. It is possible to build it yourself, to transform a common LoRa
development board into one, or it can be purchased as a complete transceiver.
Reticulum can also be encapsulated over existing IP networks, so there's
nothing stopping you from using it over wired ethernet or your local WiFi
@@ -83,24 +84,47 @@ self-configuring, resilient and encrypted mesh.
As an example, it's possible to set up a Raspberry Pi connected to both a
LoRa radio, a packet radio TNC and a WiFi network. Once the interfaces are
configured, Reticulum will take care of the rest, and any device on the WiFi
added, Reticulum will take care of the rest, and any device on the WiFi
network can communicate with nodes on the LoRa and packet radio sides of the
network, and vice versa.
Interface Types and Devices
===========================
Reticulum implements a range of generalised interface types that covers most of the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. Currently, the following interfaces are supported:
Reticulum implements a range of generalised interface types that covers the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. Currently, Reticulum can use the following devices and communication mediums:
* Any ethernet device
* WiFi devices
* Wired ethernet devices
* Fibre-optic transceivers
* Data radios with ethernet ports
* LoRa using `RNode <https://unsigned.io/rnode>`_
* Can be installed on `many popular LoRa boards <https://github.com/markqvist/rnodeconfigutil#supported-devices>`_
* Can be purchased as a `ready to use transceiver <https://unsigned.io/rnode>`_
* Packet Radio TNCs, such as `OpenModem <https://unsigned.io/openmodem>`_
* Any packet radio TNC in KISS mode
* Ideal for VHF and UHF radio
* Any device with a serial port
* The I2P network
* TCP over IP networks
* UDP over IP networks
For a full list and more details, see the :ref:`Supported Interfaces<interfaces-main>` chapter.
For a full list and more details, see the :ref:`Supported Interfaces<interfaces-main>` chapter.
Caveat Emptor
==============
Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.
+1 -1
View File
@@ -1,6 +1,6 @@
var DOCUMENTATION_OPTIONS = {
URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
VERSION: '0.2.6 beta',
VERSION: '0.3.5 beta',
LANGUAGE: 'None',
COLLAPSE_INDEX: false,
BUILDER: 'html',
+53 -6
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Code Examples &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Code Examples &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -27,7 +27,7 @@
<li class="right" >
<a href="reference.html" title="API Reference"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Code Examples</a></li>
</ul>
</div>
@@ -174,7 +174,7 @@ notifications about announces from relevant destinations.</p>
<span class="n">APP_NAME</span> <span class="o">=</span> <span class="s2">&quot;example_utilities&quot;</span>
<span class="c1"># We initialise two lists of strings to use as app_data</span>
<span class="n">fruits</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Peach&quot;</span><span class="p">,</span> <span class="s2">&quot;Quince&quot;</span><span class="p">,</span> <span class="s2">&quot;Date palm&quot;</span><span class="p">,</span> <span class="s2">&quot;Tangerine&quot;</span><span class="p">,</span> <span class="s2">&quot;Pomelo&quot;</span><span class="p">,</span> <span class="s2">&quot;Carambola&quot;</span><span class="p">,</span> <span class="s2">&quot;Grape&quot;</span><span class="p">]</span>
<span class="n">fruits</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Peach&quot;</span><span class="p">,</span> <span class="s2">&quot;Quince&quot;</span><span class="p">,</span> <span class="s2">&quot;Date&quot;</span><span class="p">,</span> <span class="s2">&quot;Tangerine&quot;</span><span class="p">,</span> <span class="s2">&quot;Pomelo&quot;</span><span class="p">,</span> <span class="s2">&quot;Carambola&quot;</span><span class="p">,</span> <span class="s2">&quot;Grape&quot;</span><span class="p">]</span>
<span class="n">noble_gases</span> <span class="o">=</span> <span class="p">[</span><span class="s2">&quot;Helium&quot;</span><span class="p">,</span> <span class="s2">&quot;Neon&quot;</span><span class="p">,</span> <span class="s2">&quot;Argon&quot;</span><span class="p">,</span> <span class="s2">&quot;Krypton&quot;</span><span class="p">,</span> <span class="s2">&quot;Xenon&quot;</span><span class="p">,</span> <span class="s2">&quot;Radon&quot;</span><span class="p">,</span> <span class="s2">&quot;Oganesson&quot;</span><span class="p">]</span>
<span class="c1"># This initialisation is executed when the program is started</span>
@@ -488,6 +488,8 @@ the Packet interface.</p>
<span class="c1"># This initialisation is executed when the users chooses</span>
<span class="c1"># to run as a server</span>
<span class="k">def</span> <span class="nf">server</span><span class="p">(</span><span class="n">configpath</span><span class="p">):</span>
<span class="k">global</span> <span class="n">reticulum</span>
<span class="c1"># We must first initialise Reticulum</span>
<span class="n">reticulum</span> <span class="o">=</span> <span class="n">RNS</span><span class="o">.</span><span class="n">Reticulum</span><span class="p">(</span><span class="n">configpath</span><span class="p">)</span>
@@ -544,11 +546,32 @@ the Packet interface.</p>
<span class="k">def</span> <span class="nf">server_callback</span><span class="p">(</span><span class="n">message</span><span class="p">,</span> <span class="n">packet</span><span class="p">):</span>
<span class="k">global</span> <span class="n">reticulum</span>
<span class="c1"># Tell the user that we received an echo request, and</span>
<span class="c1"># that we are going to send a reply to the requester.</span>
<span class="c1"># Sending the proof is handled automatically, since we</span>
<span class="c1"># set up the destination to prove all incoming packets.</span>
<span class="n">RNS</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="s2">&quot;Received packet from echo client, proof sent&quot;</span><span class="p">)</span>
<span class="n">reception_stats</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
<span class="k">if</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">is_connected_to_shared_instance</span><span class="p">:</span>
<span class="n">reception_rssi</span> <span class="o">=</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">get_packet_rssi</span><span class="p">(</span><span class="n">packet</span><span class="o">.</span><span class="n">packet_hash</span><span class="p">)</span>
<span class="n">reception_snr</span> <span class="o">=</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">get_packet_snr</span><span class="p">(</span><span class="n">packet</span><span class="o">.</span><span class="n">packet_hash</span><span class="p">)</span>
<span class="k">if</span> <span class="n">reception_rssi</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [RSSI &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">reception_rssi</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dBm]&quot;</span>
<span class="k">if</span> <span class="n">reception_snr</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [SNR &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">reception_snr</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dBm]&quot;</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">if</span> <span class="n">packet</span><span class="o">.</span><span class="n">rssi</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [RSSI &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">packet</span><span class="o">.</span><span class="n">rssi</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dBm]&quot;</span>
<span class="k">if</span> <span class="n">packet</span><span class="o">.</span><span class="n">snr</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [SNR &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">packet</span><span class="o">.</span><span class="n">snr</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dB]&quot;</span>
<span class="n">RNS</span><span class="o">.</span><span class="n">log</span><span class="p">(</span><span class="s2">&quot;Received packet from echo client, proof sent&quot;</span><span class="o">+</span><span class="n">reception_stats</span><span class="p">)</span>
<span class="c1">##########################################################</span>
@@ -558,6 +581,8 @@ the Packet interface.</p>
<span class="c1"># This initialisation is executed when the users chooses</span>
<span class="c1"># to run as a client</span>
<span class="k">def</span> <span class="nf">client</span><span class="p">(</span><span class="n">destination_hexhash</span><span class="p">,</span> <span class="n">configpath</span><span class="p">,</span> <span class="n">timeout</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<span class="k">global</span> <span class="n">reticulum</span>
<span class="c1"># We need a binary representation of the destination</span>
<span class="c1"># hash that was entered on the command line</span>
<span class="k">try</span><span class="p">:</span>
@@ -654,6 +679,8 @@ the Packet interface.</p>
<span class="c1"># This function is called when our reply destination</span>
<span class="c1"># receives a proof packet.</span>
<span class="k">def</span> <span class="nf">packet_delivered</span><span class="p">(</span><span class="n">receipt</span><span class="p">):</span>
<span class="k">global</span> <span class="n">reticulum</span>
<span class="k">if</span> <span class="n">receipt</span><span class="o">.</span><span class="n">status</span> <span class="o">==</span> <span class="n">RNS</span><span class="o">.</span><span class="n">PacketReceipt</span><span class="o">.</span><span class="n">DELIVERED</span><span class="p">:</span>
<span class="n">rtt</span> <span class="o">=</span> <span class="n">receipt</span><span class="o">.</span><span class="n">get_rtt</span><span class="p">()</span>
<span class="k">if</span> <span class="p">(</span><span class="n">rtt</span> <span class="o">&gt;=</span> <span class="mi">1</span><span class="p">):</span>
@@ -663,10 +690,30 @@ the Packet interface.</p>
<span class="n">rtt</span> <span class="o">=</span> <span class="nb">round</span><span class="p">(</span><span class="n">rtt</span><span class="o">*</span><span class="mi">1000</span><span class="p">,</span> <span class="mi">3</span><span class="p">)</span>
<span class="n">rttstring</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="n">rtt</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; milliseconds&quot;</span>
<span class="n">reception_stats</span> <span class="o">=</span> <span class="s2">&quot;&quot;</span>
<span class="k">if</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">is_connected_to_shared_instance</span><span class="p">:</span>
<span class="n">reception_rssi</span> <span class="o">=</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">get_packet_rssi</span><span class="p">(</span><span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">packet_hash</span><span class="p">)</span>
<span class="n">reception_snr</span> <span class="o">=</span> <span class="n">reticulum</span><span class="o">.</span><span class="n">get_packet_snr</span><span class="p">(</span><span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">packet_hash</span><span class="p">)</span>
<span class="k">if</span> <span class="n">reception_rssi</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [RSSI &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">reception_rssi</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dBm]&quot;</span>
<span class="k">if</span> <span class="n">reception_snr</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [SNR &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">reception_snr</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dB]&quot;</span>
<span class="k">else</span><span class="p">:</span>
<span class="k">if</span> <span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="k">if</span> <span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">rssi</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [RSSI &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">rssi</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dBm]&quot;</span>
<span class="k">if</span> <span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">snr</span> <span class="o">!=</span> <span class="kc">None</span><span class="p">:</span>
<span class="n">reception_stats</span> <span class="o">+=</span> <span class="s2">&quot; [SNR &quot;</span><span class="o">+</span><span class="nb">str</span><span class="p">(</span><span class="n">receipt</span><span class="o">.</span><span class="n">proof_packet</span><span class="o">.</span><span class="n">snr</span><span class="p">)</span><span class="o">+</span><span class="s2">&quot; dB]&quot;</span>
<span class="n">RNS</span><span class="o">.</span><span class="n">log</span><span class="p">(</span>
<span class="s2">&quot;Valid reply received from &quot;</span><span class="o">+</span>
<span class="n">RNS</span><span class="o">.</span><span class="n">prettyhexrep</span><span class="p">(</span><span class="n">receipt</span><span class="o">.</span><span class="n">destination</span><span class="o">.</span><span class="n">hash</span><span class="p">)</span><span class="o">+</span>
<span class="s2">&quot;, round-trip time is &quot;</span><span class="o">+</span><span class="n">rttstring</span>
<span class="s2">&quot;, round-trip time is &quot;</span><span class="o">+</span><span class="n">rttstring</span><span class="o">+</span>
<span class="n">reception_stats</span>
<span class="p">)</span>
<span class="c1"># This function is called if a packet times out.</span>
@@ -2319,7 +2366,7 @@ interface to efficiently pass files of any size over a Reticulum <a class="refer
<li class="right" >
<a href="reference.html" title="API Reference"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Code Examples</a></li>
</ul>
</div>
+6 -4
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Index &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Index &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -23,7 +23,7 @@
<li class="right" style="margin-right: 10px">
<a href="#" title="General Index"
accesskey="I">index</a></li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Index</a></li>
</ul>
</div>
@@ -60,10 +60,12 @@
<table style="width: 100%" class="indextable genindextable"><tr>
<td style="width: 33%; vertical-align: top;"><ul>
<li><a href="reference.html#RNS.Resource.advertise">advertise() (RNS.Resource method)</a>
</li>
<li><a href="reference.html#RNS.Destination.announce">announce() (RNS.Destination method)</a>
</li>
</ul></td>
<td style="width: 33%; vertical-align: top;"><ul>
<li><a href="reference.html#RNS.Destination.announce">announce() (RNS.Destination method)</a>
<li><a href="reference.html#RNS.Reticulum.ANNOUNCE_CAP">ANNOUNCE_CAP (RNS.Reticulum attribute)</a>
</li>
<li><a href="reference.html#RNS.Destination.app_and_aspects_from_name">app_and_aspects_from_name() (RNS.Destination static method)</a>
</li>
@@ -416,7 +418,7 @@
<li class="right" style="margin-right: 10px">
<a href="#" title="General Index"
>index</a></li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Index</a></li>
</ul>
</div>
+204 -11
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Getting Started Fast &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Getting Started Fast &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -31,7 +31,7 @@
<li class="right" >
<a href="whatis.html" title="What is Reticulum?"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Getting Started Fast</a></li>
</ul>
</div>
@@ -48,12 +48,26 @@ you want to do. This guide will outline sensible starting paths for different
scenarios.</p>
<div class="section" id="try-using-a-reticulum-based-program">
<h2>Try Using a Reticulum-based Program<a class="headerlink" href="#try-using-a-reticulum-based-program" title="Permalink to this headline"></a></h2>
<p>If you simply want to try using a program built with Reticulum, you can take
a look at <a class="reference external" href="https://github.com/markqvist/nomadnet">Nomad Network</a>, which
provides a basic encrypted communications suite built completely on Reticulum.</p>
<p>If you simply want to try using a program built with Reticulum, a few different
programs exist that allow basic communication and a range of other useful functions
over even extremely low-bandwidth Reticulum networks.</p>
<p>These programs will let you get a feel for how Reticulum works. They have been designed
to run well over networks based on LoRa or packet radio, but can also be used completely
over local WiFi, wired ethernet, the Internet, or any combination.</p>
<p>As such, it is easy to get started experimenting, without having to set up any radio
transceivers or infrastructure just to try it out. Launching the programs on separate
devices connected to the same WiFi network is enough to get started, and physical
radio interfaces can then be added later.</p>
<div class="section" id="nomad-network">
<h3>Nomad Network<a class="headerlink" href="#nomad-network" title="Permalink to this headline"></a></h3>
<p>The terminal-based program <a class="reference external" href="https://github.com/markqvist/nomadnet">Nomad Network</a>
provides a complete encrypted communications suite built with Reticulum. It features
encrypted messaging (both direct and delayed-delivery for offline users), file sharing,
and has a built-in text-browser and page server with support for dynamically rendered pages,
user authentication and more.</p>
<a class="reference external image-reference" href="_images/nomadnet_3.png"><img alt="_images/nomadnet_3.png" src="_images/nomadnet_3.png" /></a>
<p><a class="reference external" href="https://github.com/markqvist/nomadnet">Nomad Network</a> is a user-facing client
in the development for the messaging and information-sharing protocol
for the messaging and information-sharing protocol
<a class="reference external" href="https://github.com/markqvist/lxmf">LXMF</a>, another project built with Reticulum.</p>
<p>You can install Nomad Network via pip:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Install ...</span>
@@ -63,23 +77,114 @@ in the development for the messaging and information-sharing protocol
<span class="n">nomadnet</span>
</pre></div>
</div>
<p><strong>Please Note</strong>: If this is the very first time you use pip to install a program
on your system, you might need to reboot your system for your program to become
available. If you get a “command not found” error or similar when running the
program, reboot your system and try again.</p>
</div>
<div class="section" id="sideband">
<h3>Sideband<a class="headerlink" href="#sideband" title="Permalink to this headline"></a></h3>
<p>If you would rather use a program with a graphical user interface, you can take
a look at <a class="reference external" href="https://unsigned.io/sideband">Sideband</a>, which is available for Android,
Linux and macOS.</p>
<a class="reference external image-reference" href="_images/sideband_1.png"><img alt="_images/sideband_1.png" class="align-center" src="_images/sideband_1.png" style="width: 400px;" /></a>
<p>Sideband is currently in the early stages of development, but already provides basic
communication features, and interoperates with Nomad Network, or any other LXMF client.</p>
</div>
</div>
<div class="section" id="using-the-included-utilities">
<h2>Using the Included Utilities<a class="headerlink" href="#using-the-included-utilities" title="Permalink to this headline"></a></h2>
<p>Reticulum comes with a range of included utilities that make it easier to
manage your network, check connectivity and make Reticulum available to other
programs on your system.</p>
<p>You can use <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> to run Reticulum as a background or foreground service,
and the <code class="docutils literal notranslate"><span class="pre">rnstatus</span></code>, <code class="docutils literal notranslate"><span class="pre">rnpath</span></code> and <code class="docutils literal notranslate"><span class="pre">rnprobe</span></code> utilities to view and query
network status and connectivity.</p>
<p>To learn more about these utility programs, have a look at the
<a class="reference internal" href="using.html#using-main"><span class="std std-ref">Using Reticulum on Your System</span></a> chapter of this manual.</p>
</div>
<div class="section" id="creating-a-network-with-reticulum">
<h2>Creating a Network With Reticulum<a class="headerlink" href="#creating-a-network-with-reticulum" title="Permalink to this headline"></a></h2>
<p>To create a network, you will need to specify one or more <em>interfaces</em> for
Reticulum to use. This is done in the Reticulum configuration file, which by
default is located at <code class="docutils literal notranslate"><span class="pre">~/.reticulum/config</span></code>.</p>
default is located at <code class="docutils literal notranslate"><span class="pre">~/.reticulum/config</span></code>. You can edit this file by hand,
or use the interactive <code class="docutils literal notranslate"><span class="pre">rnsconfig</span></code> utility.</p>
<p>When Reticulum is started for the first time, it will create a default
configuration file, with one active interface. This default interface uses
your existing ethernet network (if there is one), and only allows you to
communicate with other Reticulum peers within your local broadcast domain.</p>
your existing ethernet and WiFi networks (if any), and only allows you to
communicate with other Reticulum peers within your local broadcast domains.</p>
<p>To communicate further, you will have to add one or more interfaces. The default
configuration includes a number of examples, ranging from using TCP over the
internet, to LoRa and Packet Radio interfaces.</p>
<p>With Reticulum, you only need to configure what interfaces you want to communicate
over. There is no need to configure address spaces, subnets, routing tables,
or other things you might be used to from other network types.</p>
<p>Once Reticulums knows which interfaces it should use, it will automatically
discover topography and configure transport of data to any destinations it
knows about.</p>
<p>In situations where you already have an established WiFi or ethernet network, and
many devices that want to utilise the same external Reticulum network (for example over
LoRa), it will often be sufficient to let one system act as a Reticulum gateway, by
adding any external interfaces to this systems configuration, and enabling transport. Any
other device on your local WiFi will then be able to connect to this wider Reticulum
network just using the default interface configuration.</p>
<p>Possibly, the examples in the config file are enough to get you started. If
you want more information, you can read the <a class="reference internal" href="networks.html#networks-main"><span class="std std-ref">Building Networks</span></a>
and <a class="reference internal" href="interfaces.html#interfaces-main"><span class="std std-ref">Interfaces</span></a> chapters of this manual.</p>
</div>
<div class="section" id="connecting-reticulum-instances-over-the-internet">
<h2>Connecting Reticulum Instances Over the Internet<a class="headerlink" href="#connecting-reticulum-instances-over-the-internet" title="Permalink to this headline"></a></h2>
<p>Reticulum currently offers two interfaces suitable for connecting instances over the Internet: <a class="reference internal" href="interfaces.html#interfaces-tcps"><span class="std std-ref">TCP</span></a>
and <a class="reference internal" href="interfaces.html#interfaces-i2p"><span class="std std-ref">I2P</span></a>. Each interface offers a different set of features, and Reticulum
users should carefully choose the interface which best suites their needs.</p>
<p>The <code class="docutils literal notranslate"><span class="pre">TCPServerInterface</span></code> allows users to host an instance accessible over TCP/IP. This
method is generally faster, lower latency, and more energy efficient than using <code class="docutils literal notranslate"><span class="pre">I2PInterface</span></code>,
however it also leaks more data about the server host.</p>
<p>TCP connections reveal the IP address of both your instance and the server to anyone who can
inspect the connection. Someone could use this information to determine your location or identity. Adversaries
inspecting your packets may be able to record packet metadata like time of transmission and packet size.
Even though Reticulum encrypts traffic, TCP does not, so an adversary may be able to use
packet inspection to learn that a system is running Reticulum, and what other IP adresses connect to it.
Hosting a publicly reachable instance over TCP also requires a publicly reachable IP address,
which most Internet connections dont offer anymore.</p>
<p>The <code class="docutils literal notranslate"><span class="pre">I2PInterface</span></code> routes messages through the <a class="reference external" href="https://geti2p.net/en/">Invisible Internet Protocol
(I2P)</a>. To properly use this interface, users must also run an I2P daemon in
parallel to <code class="docutils literal notranslate"><span class="pre">rnsd</span></code>. For always-on I2P nodes it is recommended to use <a class="reference external" href="https://i2pd.website/">i2pd</a>.</p>
<p>By default, I2P will encrypt and mix all traffic sent over the Internet, and
hide both the sender and receiver Reticulum instance IP addresses. Running an I2P node
will also relay other I2P users encrypted packets, which will use extra
bandwidth and compute power, but also makes timing attacks and other forms of
deep-packet-inspection much more difficult.</p>
<p>I2P also allows users to host globally available Reticulum instances from non-public IPs and behind firewalls.</p>
<p>In general it is recommended to use an I2P node if you want to host a publically accessible
instance, while preserving anonymity. If you care more about performance, and a slightly
easier setup, use TCP.</p>
</div>
<div class="section" id="connect-to-the-public-testnet">
<h2>Connect to the Public Testnet<a class="headerlink" href="#connect-to-the-public-testnet" title="Permalink to this headline"></a></h2>
<p>An experimental public testnet has been made accessible over both I2P and TCP. You can join it
by adding one of the following interfaces to your <code class="docutils literal notranslate"><span class="pre">.reticulum/config</span></code> file:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># For connecting over TCP/IP:</span>
<span class="p">[[</span><span class="n">RNS</span> <span class="n">Testnet</span> <span class="n">Frankfurt</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPClientInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">target_host</span> <span class="o">=</span> <span class="n">frankfurt</span><span class="o">.</span><span class="n">rns</span><span class="o">.</span><span class="n">unsigned</span><span class="o">.</span><span class="n">io</span>
<span class="n">target_port</span> <span class="o">=</span> <span class="mi">4965</span>
<span class="c1"># For connecting over I2P:</span>
<span class="p">[[</span><span class="n">RNS</span> <span class="n">Testnet</span> <span class="n">I2P</span> <span class="n">Node</span> <span class="n">A</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">I2PInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">peers</span> <span class="o">=</span> <span class="n">ykzlw5ujbaqc2xkec4cpvgyxj257wcrmmgkuxqmqcur7cq3w3lha</span><span class="o">.</span><span class="n">b32</span><span class="o">.</span><span class="n">i2p</span>
</pre></div>
</div>
<p>Many other Reticulum instances are connecting to this testnet, and you can also join it
via other entry points if you know them. There is absolutely no control over the network
topography, usage or what types of instances connect. It will also occasionally be used
to test various failure scenarios, and there are no availability or service guarantees.</p>
</div>
<div class="section" id="develop-a-program-with-reticulum">
<h2>Develop a Program with Reticulum<a class="headerlink" href="#develop-a-program-with-reticulum" title="Permalink to this headline"></a></h2>
<p>If you want to develop programs that use Reticulum, the easiest way to get
@@ -90,6 +195,10 @@ started is to install the latest release of Reticulum via pip:</p>
<p>The above command will install Reticulum and dependencies, and you will be
ready to import and use RNS in your own programs. The next step will most
likely be to look at some <a class="reference internal" href="examples.html#examples-main"><span class="std std-ref">Example Programs</span></a>.</p>
<p>For extended functionality, you can install optional dependencies:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pip3</span> <span class="n">install</span> <span class="n">pyserial</span> <span class="n">netifaces</span>
</pre></div>
</div>
<p>Further information can be found in the <a class="reference internal" href="reference.html#api-main"><span class="std std-ref">API Reference</span></a>.</p>
</div>
<div class="section" id="participate-in-reticulum-development">
@@ -132,6 +241,80 @@ dont use pip, but try this recipe:</p>
<p>When you have experimented with the basic examples, its time to go read the
<a class="reference internal" href="understanding.html#understanding-main"><span class="std std-ref">Understanding Reticulum</span></a> chapter.</p>
</div>
<div class="section" id="reticulum-on-arm64">
<h2>Reticulum on ARM64<a class="headerlink" href="#reticulum-on-arm64" title="Permalink to this headline"></a></h2>
<p>On some architectures, including ARM64, not all dependencies have precompiled
binaries. On such systems, you will need to install <code class="docutils literal notranslate"><span class="pre">python3-dev</span></code> before
installing Reticulum or programs that depend on Reticulum.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Install Python and development packages</span>
<span class="n">sudo</span> <span class="n">apt</span> <span class="n">update</span>
<span class="n">sudo</span> <span class="n">apt</span> <span class="n">install</span> <span class="n">python3</span> <span class="n">python3</span><span class="o">-</span><span class="n">pip</span> <span class="n">python3</span><span class="o">-</span><span class="n">dev</span>
<span class="c1"># Install Reticulum</span>
<span class="n">python3</span> <span class="o">-</span><span class="n">m</span> <span class="n">pip</span> <span class="n">install</span> <span class="n">rns</span>
</pre></div>
</div>
</div>
<div class="section" id="reticulum-on-android">
<h2>Reticulum on Android<a class="headerlink" href="#reticulum-on-android" title="Permalink to this headline"></a></h2>
<p>Reticulum can be used on Android in different ways. The easiest way to get
started is using an app like <a class="reference external" href="https://unsigned.io/sideband">Sideband</a>.</p>
<p>For more control and features, you can use Reticulum and related programs via
the <a class="reference external" href="https://termux.com/">Termux app</a>, at the time of writing available on
<a class="reference external" href="https://f-droid.org">F-droid</a>.</p>
<p>Termux is a terminal emulator and Linux environment for Android based devices,
which includes the ability to use many different programs and libraries,
including Reticulum.</p>
<p>Since the Python cryptography.io module does not offer pre-built wheels for
Android, the standard one-line install of Reticulum does not work on Android,
and a few extra commands are required.</p>
<p>From within Termux, execute the following:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># First, make sure indexes and packages are up to date.</span>
<span class="n">pkg</span> <span class="n">update</span>
<span class="n">pkg</span> <span class="n">upgrade</span>
<span class="c1"># Then install dependencies for the cryptography library.</span>
<span class="n">pkg</span> <span class="n">install</span> <span class="n">python</span> <span class="n">build</span><span class="o">-</span><span class="n">essential</span> <span class="n">openssl</span> <span class="n">libffi</span> <span class="n">rust</span>
<span class="c1"># Make sure pip is up to date, and install the wheel module.</span>
<span class="n">pip3</span> <span class="n">install</span> <span class="n">wheel</span> <span class="n">pip</span> <span class="o">--</span><span class="n">upgrade</span>
<span class="c1"># To allow the installer to build the cryptography module,</span>
<span class="c1"># we need to let it know what platform we are compiling for:</span>
<span class="n">export</span> <span class="n">CARGO_BUILD_TARGET</span><span class="o">=</span><span class="s2">&quot;aarch64-linux-android&quot;</span>
<span class="c1"># Start the install process for the cryptography module.</span>
<span class="c1"># Depending on your device, this can take several minutes,</span>
<span class="c1"># since the module must be compiled locally on your device.</span>
<span class="n">pip3</span> <span class="n">install</span> <span class="n">cryptography</span>
<span class="c1"># If the above installation succeeds, you can now install</span>
<span class="c1"># Reticulum and any related software</span>
<span class="n">pip3</span> <span class="n">install</span> <span class="n">rns</span>
</pre></div>
</div>
<p>It is also possible to include Reticulum in apps compiled and distributed as
Android APKs. A detailed tutorial and example source code will be included
here at a later point.</p>
</div>
<div class="section" id="adding-radio-interfaces">
<h2>Adding Radio Interfaces<a class="headerlink" href="#adding-radio-interfaces" title="Permalink to this headline"></a></h2>
<p>Once you have Reticulum installed and working, you can add radio interfaces with
any compatible hardware you have available. For information on how to configure
this, see the <a class="reference internal" href="interfaces.html#interfaces-main"><span class="std std-ref">Interfaces</span></a> section of this manual.</p>
<p>A range of common LoRa development boards and transceiver modules can be used
as interfaces with Reticulum. You can refer to the following external resources
for more information:</p>
<ul class="simple">
<li><p><a class="reference external" href="https://unsigned.io/how-to-make-your-own-rnodes/">How To Make Your Own RNodes</a></p></li>
<li><p><a class="reference external" href="https://unsigned.io/installing-rnode-firmware-on-t-beam-and-lora32-devices/">Installing RNode Firmware on Compatible LoRa Devices</a></p></li>
<li><p><a class="reference external" href="https://unsigned.io/private-messaging-over-lora/">Private, Secure and Uncensorable Messaging Over a LoRa Mesh</a></p></li>
<li><p><a class="reference external" href="https://github.com/markqvist/RNode_Firmware/">RNode Firmware</a></p></li>
</ul>
<p>If you have communications hardware that you think would be suitable for use with Reticulum,
you are welcome to head over to the <a class="reference external" href="https://github.com/markqvist/Reticulum/discussions">GitHub discussion pages</a>
and propose adding an interface for the hardware.</p>
</div>
</div>
@@ -144,10 +327,20 @@ dont use pip, but try this recipe:</p>
<h3><a href="index.html">Table of Contents</a></h3>
<ul>
<li><a class="reference internal" href="#">Getting Started Fast</a><ul>
<li><a class="reference internal" href="#try-using-a-reticulum-based-program">Try Using a Reticulum-based Program</a></li>
<li><a class="reference internal" href="#try-using-a-reticulum-based-program">Try Using a Reticulum-based Program</a><ul>
<li><a class="reference internal" href="#nomad-network">Nomad Network</a></li>
<li><a class="reference internal" href="#sideband">Sideband</a></li>
</ul>
</li>
<li><a class="reference internal" href="#using-the-included-utilities">Using the Included Utilities</a></li>
<li><a class="reference internal" href="#creating-a-network-with-reticulum">Creating a Network With Reticulum</a></li>
<li><a class="reference internal" href="#connecting-reticulum-instances-over-the-internet">Connecting Reticulum Instances Over the Internet</a></li>
<li><a class="reference internal" href="#connect-to-the-public-testnet">Connect to the Public Testnet</a></li>
<li><a class="reference internal" href="#develop-a-program-with-reticulum">Develop a Program with Reticulum</a></li>
<li><a class="reference internal" href="#participate-in-reticulum-development">Participate in Reticulum Development</a></li>
<li><a class="reference internal" href="#reticulum-on-arm64">Reticulum on ARM64</a></li>
<li><a class="reference internal" href="#reticulum-on-android">Reticulum on Android</a></li>
<li><a class="reference internal" href="#adding-radio-interfaces">Adding Radio Interfaces</a></li>
</ul>
</li>
</ul>
@@ -191,7 +384,7 @@ dont use pip, but try this recipe:</p>
<li class="right" >
<a href="whatis.html" title="What is Reticulum?"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Getting Started Fast</a></li>
</ul>
</div>
+28 -9
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Reticulum Network Stack Manual &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Reticulum Network Stack Manual &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -27,7 +27,7 @@
<li class="right" >
<a href="whatis.html" title="What is Reticulum?"
accesskey="N">next</a> |</li>
<li class="nav-item nav-item-0"><a href="#">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="#">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Reticulum Network Stack Manual</a></li>
</ul>
</div>
@@ -46,17 +46,27 @@ to participate in the development of Reticulum itself.</p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="whatis.html">What is Reticulum?</a><ul>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#current-status">Current Status</a></li>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#caveat-emptor">Caveat Emptor</a></li>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#what-does-reticulum-offer">What does Reticulum Offer?</a></li>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#where-can-reticulum-be-used">Where can Reticulum be Used?</a></li>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#interface-types-and-devices">Interface Types and Devices</a></li>
<li class="toctree-l2"><a class="reference internal" href="whatis.html#caveat-emptor">Caveat Emptor</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="gettingstartedfast.html">Getting Started Fast</a><ul>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#try-using-a-reticulum-based-program">Try Using a Reticulum-based Program</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#try-using-a-reticulum-based-program">Try Using a Reticulum-based Program</a><ul>
<li class="toctree-l3"><a class="reference internal" href="gettingstartedfast.html#nomad-network">Nomad Network</a></li>
<li class="toctree-l3"><a class="reference internal" href="gettingstartedfast.html#sideband">Sideband</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#using-the-included-utilities">Using the Included Utilities</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#creating-a-network-with-reticulum">Creating a Network With Reticulum</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#connecting-reticulum-instances-over-the-internet">Connecting Reticulum Instances Over the Internet</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#connect-to-the-public-testnet">Connect to the Public Testnet</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#develop-a-program-with-reticulum">Develop a Program with Reticulum</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#participate-in-reticulum-development">Participate in Reticulum Development</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#reticulum-on-arm64">Reticulum on ARM64</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#reticulum-on-android">Reticulum on Android</a></li>
<li class="toctree-l2"><a class="reference internal" href="gettingstartedfast.html#adding-radio-interfaces">Adding Radio Interfaces</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="using.html">Using Reticulum on Your System</a><ul>
@@ -67,6 +77,11 @@ to participate in the development of Reticulum itself.</p>
<li class="toctree-l3"><a class="reference internal" href="using.html#the-rnprobe-utility">The rnprobe Utility</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="using.html#improving-system-configuration">Improving System Configuration</a><ul>
<li class="toctree-l3"><a class="reference internal" href="using.html#fixed-serial-port-names">Fixed Serial Port Names</a></li>
<li class="toctree-l3"><a class="reference internal" href="using.html#reticulum-as-a-system-service">Reticulum as a System Service</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="networks.html">Building Networks</a><ul>
@@ -80,9 +95,12 @@ to participate in the development of Reticulum itself.</p>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="interfaces.html">Supported Interfaces</a><ul>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#udp-interface">UDP Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#common-interface-options">Common Interface Options</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#auto-interface">Auto Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#i2p-interface">I2P Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#tcp-server-interface">TCP Server Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#tcp-client-interface">TCP Client Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#udp-interface">UDP Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#rnode-lora-interface">RNode LoRa Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#serial-interface">Serial Interface</a></li>
<li class="toctree-l2"><a class="reference internal" href="interfaces.html#kiss-interface">KISS Interface</a></li>
@@ -100,16 +118,17 @@ to participate in the development of Reticulum itself.</p>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="understanding.html#reticulum-transport">Reticulum Transport</a><ul>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#node-types">Node Types</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#the-announce-mechanism-in-detail">The Announce Mechanism in Detail</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#reaching-the-destination">Reaching the Destination</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#resources">Resources</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="understanding.html#reference-system-setup">Reference System Setup</a></li>
<li class="toctree-l2"><a class="reference internal" href="understanding.html#reference-setup">Reference Setup</a></li>
<li class="toctree-l2"><a class="reference internal" href="understanding.html#protocol-specifics">Protocol Specifics</a><ul>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#node-types">Node Types</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#packet-prioritisation">Packet Prioritisation</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#binary-packet-format">Binary Packet Format</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#interface-access-codes">Interface Access Codes</a></li>
<li class="toctree-l3"><a class="reference internal" href="understanding.html#wire-format">Wire Format</a></li>
</ul>
</li>
</ul>
@@ -199,7 +218,7 @@ to participate in the development of Reticulum itself.</p>
<li class="right" >
<a href="whatis.html" title="What is Reticulum?"
>next</a> |</li>
<li class="nav-item nav-item-0"><a href="#">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="#">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Reticulum Network Stack Manual</a></li>
</ul>
</div>
+293 -67
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Supported Interfaces &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Supported Interfaces &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -31,7 +31,7 @@
<li class="right" >
<a href="networks.html" title="Building Networks"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Supported Interfaces</a></li>
</ul>
</div>
@@ -50,12 +50,296 @@ common to them all is that you will need to define one or more <em>interfaces</e
for Reticulum to use.</p>
<p>The following sections describe the interfaces currently available in Reticulum,
and gives example configurations for the respective interface types.</p>
<p>For a high-level overview of how networks can be formed over different interface
types, have a look at the <a class="reference internal" href="networks.html#networks-main"><span class="std std-ref">Building Networks</span></a> chapter of this
manual.</p>
<div class="section" id="common-interface-options">
<span id="interfaces-options"></span><h2>Common Interface Options<a class="headerlink" href="#common-interface-options" title="Permalink to this headline"></a></h2>
<p>A number of general configuration options are available on most interfaces.
These can be used to control various aspects of interface behaviour.</p>
<blockquote>
<div><ul>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">enabled</span></code> option tells Reticulum whether or not
to bring up the interface. Defaults to <code class="docutils literal notranslate"><span class="pre">False</span></code>. For any
interface to be brought up, the <code class="docutils literal notranslate"><span class="pre">enabled</span></code> option
must be set to <code class="docutils literal notranslate"><span class="pre">True</span></code> or <code class="docutils literal notranslate"><span class="pre">Yes</span></code>.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">mode</span></code> option allows selecting the high-level behaviour
of the interface from a number of options.</div>
</div>
<blockquote>
<div><ul class="simple">
<li><p>The default value is <code class="docutils literal notranslate"><span class="pre">full</span></code>. In this mode, all discovery,
meshing and transport functionality is available.</p></li>
<li><p>In the <code class="docutils literal notranslate"><span class="pre">access_point</span></code> (or shorthand <code class="docutils literal notranslate"><span class="pre">ap</span></code>) mode, the
interface will operate as a network access point. In this
mode, announces will not be automatically broadcasted on
the interface, and paths to destinations on the interface
will have a much shorter expiry time. This mode is useful
for creating interfaces that are mostly quiet, unless when
someone is actually using them. An example of this could
be a radio interface serving a wide area, where users are
expected to connect momentarily, use the network, and then
disappear again.</p></li>
</ul>
</div></blockquote>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">outgoing</span></code> option sets whether an interface is allowed
to transmit. Defaults to <code class="docutils literal notranslate"><span class="pre">True</span></code>. If set to <code class="docutils literal notranslate"><span class="pre">False</span></code> or <code class="docutils literal notranslate"><span class="pre">No</span></code>
the interface will only receive data, and never transmit.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">network_name</span></code> option sets the virtual network name for
the interface. This allows multiple separate network segments
to exist on the same physical channel or medium.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">passphrase</span></code> option sets an authentication passphrase on
the interface. This option can be used in conjunction with the
<code class="docutils literal notranslate"><span class="pre">network_name</span></code> option, or be used alone.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">ifac_size</span></code> option allows customising the length of the
Interface Authentication Codes carried by each packet on named
and/or authenticated network segments. It is set by default to
a size suitable for the interface in question, but can be set
to a custom size between 8 and 512 bits by using this option.
In normal usage, this option should not be changed from the
default.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">announce_cap</span></code> option lets you configure the maximum
bandwidth to allocate, at any given time, to propagating
announces and other network upkeep traffic. It is configured at
2% by default, and should normally not need to be changed. Can
be set to any value between <code class="docutils literal notranslate"><span class="pre">1</span></code> and <code class="docutils literal notranslate"><span class="pre">100</span></code>.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The <code class="docutils literal notranslate"><span class="pre">bitrate</span></code> option configures the interface bitrate.
Reticulum will use interface speeds reported by hardware, or
try to guess a suitable rate when the hardware doesnt report
any. In most cases, the automatically found rate should be
sufficient, but it can be configured by using the <code class="docutils literal notranslate"><span class="pre">bitrate</span></code>
option, to set the interface speed in <em>bits per second</em>.</div>
</div>
</li>
</ul>
</div></blockquote>
</div>
<div class="section" id="auto-interface">
<span id="interfaces-auto"></span><h2>Auto Interface<a class="headerlink" href="#auto-interface" title="Permalink to this headline"></a></h2>
<p>The Auto Interface enables communication with other discoverable Reticulum
nodes over autoconfigured IPv6 and UDP. It does not need any functional IP
infrastructure like routers or DHCP servers, but will require at least some
sort of switching medium between peers (a wired switch, a hub, a WiFi access
point or similar), and that link-local IPv6 is enabled in your operating
system, which should be enabled by default in almost all OSes.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># This example demonstrates a TCP server interface.</span>
<span class="c1"># It will listen for incoming connections on the</span>
<span class="c1"># specified IP address and port number.</span>
<span class="p">[[</span><span class="n">Default</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">AutoInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># You can create multiple isolated Reticulum</span>
<span class="c1"># networks on the same physical LAN by</span>
<span class="c1"># specifying different Group IDs.</span>
<span class="n">group_id</span> <span class="o">=</span> <span class="n">reticulum</span>
<span class="c1"># You can also select specifically which</span>
<span class="c1"># kernel networking devices to use.</span>
<span class="n">devices</span> <span class="o">=</span> <span class="n">wlan0</span><span class="p">,</span><span class="n">eth1</span>
<span class="c1"># Or let AutoInterface use all suitable</span>
<span class="c1"># devices except for a list of ignored ones.</span>
<span class="n">ignored_devices</span> <span class="o">=</span> <span class="n">tun0</span><span class="p">,</span><span class="n">eth0</span>
</pre></div>
</div>
<p>If you are connected to the Internet with IPv6, and your provider will route
IPv6 multicast, you can potentially configure the Auto Interface to globally
autodiscover other Reticulum nodes within your selected Group ID. You can specify
the discovery scope by setting it to one of <code class="docutils literal notranslate"><span class="pre">link</span></code>, <code class="docutils literal notranslate"><span class="pre">admin</span></code>, <code class="docutils literal notranslate"><span class="pre">site</span></code>,
<code class="docutils literal notranslate"><span class="pre">organisation</span></code> or <code class="docutils literal notranslate"><span class="pre">global</span></code>.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">Default</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">AutoInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># Configure global discovery</span>
<span class="n">group_id</span> <span class="o">=</span> <span class="n">custom_network_name</span>
<span class="n">discovery_scope</span> <span class="o">=</span> <span class="k">global</span>
<span class="c1"># Other configuration options</span>
<span class="n">discovery_port</span> <span class="o">=</span> <span class="mi">48555</span>
<span class="n">data_port</span> <span class="o">=</span> <span class="mi">49555</span>
</pre></div>
</div>
</div>
<div class="section" id="i2p-interface">
<span id="interfaces-i2p"></span><h2>I2P Interface<a class="headerlink" href="#i2p-interface" title="Permalink to this headline"></a></h2>
<p>The I2P interface lets you connect Reticulum instances over the
<a class="reference external" href="https://i2pd.website">Invisible Internet Protocol</a>. This can be
especially useful in cases where you want to host a globally reachable
Reticulum instance, but do not have access to any public IP addresses,
have a frequently changing IP address, or have firewalls blocking
inbound traffic.</p>
<p>Using the I2P interface, you will get a globally reachable, portable
and persistent I2P address that your Reticulum instance can be reached
at.</p>
<p>To use the I2P interface, you must have an I2P router running
on your system. The easiest way to acheive this is to download and
install the <a class="reference external" href="https://github.com/PurpleI2P/i2pd/releases/latest">latest release</a>
of the <code class="docutils literal notranslate"><span class="pre">ì2pd</span></code> package. For more details about I2P, see the
<a class="reference external" href="https://geti2p.net/en/about/intro">geti2p.net website</a>.`</p>
<p>When an I2P router is running on your system, you can simply add
an I2P interface to reticulum:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">I2P</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">I2PInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">connectable</span> <span class="o">=</span> <span class="n">yes</span>
</pre></div>
</div>
<p>On the first start, Reticulum will generate a new I2P address for the
interface and start listening for inbound traffic on it. This can take
a while the first time, especially if your I2P router was also just
started, and is not yet well-connected to the I2P network. When ready,
you should see I2P base32 address printed to your log file. You can
also inspect the status of the interface using the <code class="docutils literal notranslate"><span class="pre">rnstatus</span></code> utility.</p>
<p>To connect to other Reticulum instances over I2P, just add a comma-separated
list of I2P base32 addresses to the <code class="docutils literal notranslate"><span class="pre">peers</span></code> option of the interface:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">I2P</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">I2PInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">connectable</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">peers</span> <span class="o">=</span> <span class="mi">5</span><span class="n">urvjicpzi7q3ybztsef4i5ow2aq4soktfj7zedz53s47r54jnqq</span><span class="o">.</span><span class="n">b32</span><span class="o">.</span><span class="n">i2p</span>
</pre></div>
</div>
<p>It can take anywhere from a few seconds to a few minutes to establish
I2P connections to the desired peers, so Reticulum handles the process
in the background, and will output relevant events to the log.</p>
<p><strong>Please Note!</strong> While the I2P interface is the simplest way to use
Reticulum over I2P, it is also possible to tunnel the TCP server and
client interfaces over I2P manually. This can be useful in situations
where more control is needed, but requires manual tunnel setup through
the I2P daemon configuration.</p>
<p>It is important to note that the two methods are <em>interchangably compatible</em>.
You can use the I2PInterface to connect to a TCPServerInterface that
was manually tunneled over I2P, for example. This offers a high degree
of flexibility in network setup, while retaining ease of use in simpler
use-cases.</p>
</div>
<div class="section" id="tcp-server-interface">
<span id="interfaces-tcps"></span><h2>TCP Server Interface<a class="headerlink" href="#tcp-server-interface" title="Permalink to this headline"></a></h2>
<p>The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># This example demonstrates a TCP server interface.</span>
<span class="c1"># It will listen for incoming connections on the</span>
<span class="c1"># specified IP address and port number.</span>
<span class="p">[[</span><span class="n">TCP</span> <span class="n">Server</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPServerInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># This configuration will listen on all IP</span>
<span class="c1"># interfaces on port 4242</span>
<span class="n">listen_ip</span> <span class="o">=</span> <span class="mf">0.0</span><span class="o">.</span><span class="mf">0.0</span>
<span class="n">listen_port</span> <span class="o">=</span> <span class="mi">4242</span>
<span class="c1"># Alternatively you can bind to a specific IP</span>
<span class="c1"># listen_ip = 10.0.0.88</span>
<span class="c1"># listen_port = 4242</span>
<span class="c1"># Or a specific network device</span>
<span class="c1"># device = eth0</span>
<span class="c1"># port = 4242</span>
</pre></div>
</div>
<p><strong>Please Note!</strong> The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">TCP</span> <span class="n">Server</span> <span class="n">on</span> <span class="n">I2P</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPServerInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">listen_ip</span> <span class="o">=</span> <span class="mf">127.0</span><span class="o">.</span><span class="mf">0.1</span>
<span class="n">listen_port</span> <span class="o">=</span> <span class="mi">5001</span>
<span class="n">i2p_tunneled</span> <span class="o">=</span> <span class="n">yes</span>
</pre></div>
</div>
</div>
<div class="section" id="tcp-client-interface">
<span id="interfaces-tcpc"></span><h2>TCP Client Interface<a class="headerlink" href="#tcp-client-interface" title="Permalink to this headline"></a></h2>
<p>To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Here&#39;s an example of a TCP Client interface. The</span>
<span class="c1"># target_host can either be an IP address or a hostname.</span>
<span class="p">[[</span><span class="n">TCP</span> <span class="n">Client</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPClientInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">target_host</span> <span class="o">=</span> <span class="mf">127.0</span><span class="o">.</span><span class="mf">0.1</span>
<span class="n">target_port</span> <span class="o">=</span> <span class="mi">4242</span>
</pre></div>
</div>
<p>It is also possible to use this interface type to connect via other programs
or hardware devices that expose a KISS interface on a TCP port, for example
software-based soundmodems. To do this, use the <code class="docutils literal notranslate"><span class="pre">kiss_framing</span></code> option:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Here&#39;s an example of a TCP Client interface that connects</span>
<span class="c1"># to a software TNC soundmodem on a KISS over TCP port.</span>
<span class="p">[[</span><span class="n">TCP</span> <span class="n">KISS</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPClientInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">kiss_framing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">target_host</span> <span class="o">=</span> <span class="mf">127.0</span><span class="o">.</span><span class="mf">0.1</span>
<span class="n">target_port</span> <span class="o">=</span> <span class="mi">8001</span>
</pre></div>
</div>
<p><strong>Caution!</strong> Only use the KISS framing option when connecting to external devices
and programs like soundmodems and similar over TCP. When using the
<code class="docutils literal notranslate"><span class="pre">TCPClientInterface</span></code> in conjunction with the <code class="docutils literal notranslate"><span class="pre">TCPServerInterface</span></code> you should
never enable <code class="docutils literal notranslate"><span class="pre">kiss_framing</span></code>, since this will disable internal reliability and
recovery mechanisms that greatly improves performance over unreliable and
intermittent TCP links.</p>
<p><strong>Please Note!</strong> The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">TCP</span> <span class="n">Client</span> <span class="n">over</span> <span class="n">I2P</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPClientInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="n">yes</span>
<span class="n">target_host</span> <span class="o">=</span> <span class="mf">127.0</span><span class="o">.</span><span class="mf">0.1</span>
<span class="n">target_port</span> <span class="o">=</span> <span class="mi">5001</span>
<span class="n">i2p_tunneled</span> <span class="o">=</span> <span class="n">yes</span>
</pre></div>
</div>
</div>
<div class="section" id="udp-interface">
<span id="interfaces-udp"></span><h2>UDP Interface<a class="headerlink" href="#udp-interface" title="Permalink to this headline"></a></h2>
<p>A UDP interface can be useful for communicating over IP networks, both
private and the internet. It can also allow broadcast communication
over IP networks, so it can provide an easy way to enable connectivity
with all other peers on a local area network.</p>
<p><em>Please Note!</em> Using broadcast UDP traffic has performance implications,
especially on WiFi. If your goal is simply to enable easy communication
with all peers in your local ethernet broadcast domain, the
<a class="reference internal" href="#interfaces-auto"><span class="std std-ref">Auto Interface</span></a> performs better, and is just as
easy to use.</p>
<p>The below example is enabled by default on new Reticulum installations,
as it provides an easy way to get started and to test Reticulum on a
pre-existing LAN.</p>
@@ -65,7 +349,7 @@ pre-existing LAN.</p>
<span class="p">[[</span><span class="n">Default</span> <span class="n">UDP</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">UDPInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">listen_ip</span> <span class="o">=</span> <span class="mf">0.0</span><span class="o">.</span><span class="mf">0.0</span>
<span class="n">listen_port</span> <span class="o">=</span> <span class="mi">4242</span>
<span class="n">forward_ip</span> <span class="o">=</span> <span class="mf">255.255</span><span class="o">.</span><span class="mf">255.255</span>
@@ -73,9 +357,7 @@ pre-existing LAN.</p>
<span class="c1"># The above configuration will allow communication</span>
<span class="c1"># within the local broadcast domains of all local</span>
<span class="c1"># IP interfaces. This is enabled by default as an</span>
<span class="c1"># easy way to get started, but you might want to</span>
<span class="c1"># consider altering it to something more specific.</span>
<span class="c1"># IP interfaces.</span>
<span class="c1"># Instead of specifying listen_ip, listen_port,</span>
<span class="c1"># forward_ip and forward_port, you can also bind</span>
@@ -105,55 +387,6 @@ pre-existing LAN.</p>
</pre></div>
</div>
</div>
<div class="section" id="tcp-server-interface">
<span id="interfaces-tcps"></span><h2>TCP Server Interface<a class="headerlink" href="#tcp-server-interface" title="Permalink to this headline"></a></h2>
<p>The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># This example demonstrates a TCP server interface.</span>
<span class="c1"># It will listen for incoming connections on the</span>
<span class="c1"># specified IP address and port number.</span>
<span class="p">[[</span><span class="n">TCP</span> <span class="n">Server</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPServerInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># This configuration will listen on all IP</span>
<span class="c1"># interfaces on port 4242</span>
<span class="n">listen_ip</span> <span class="o">=</span> <span class="mf">0.0</span><span class="o">.</span><span class="mf">0.0</span>
<span class="n">listen_port</span> <span class="o">=</span> <span class="mi">4242</span>
<span class="c1"># Alternatively you can bind to a specific IP</span>
<span class="c1"># listen_ip = 10.0.0.88</span>
<span class="c1"># listen_port = 4242</span>
<span class="c1"># Or a specific network device</span>
<span class="c1"># device = eth0</span>
<span class="c1"># port = 4242</span>
</pre></div>
</div>
</div>
<div class="section" id="tcp-client-interface">
<span id="interfaces-tcpc"></span><h2>TCP Client Interface<a class="headerlink" href="#tcp-client-interface" title="Permalink to this headline"></a></h2>
<p>To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Here&#39;s an example of a TCP Client interface. The</span>
<span class="c1"># target_host can either be an IP address or a hostname.</span>
<span class="p">[[</span><span class="n">TCP</span> <span class="n">Client</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">TCPClientInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">target_host</span> <span class="o">=</span> <span class="mf">127.0</span><span class="o">.</span><span class="mf">0.1</span>
<span class="n">target_port</span> <span class="o">=</span> <span class="mi">4242</span>
</pre></div>
</div>
</div>
<div class="section" id="rnode-lora-interface">
<span id="interfaces-rnode"></span><h2>RNode LoRa Interface<a class="headerlink" href="#rnode-lora-interface" title="Permalink to this headline"></a></h2>
<p>To use Reticulum over LoRa, the <a class="reference external" href="https://unsigned.io/rnode/">RNode</a> interface
@@ -167,11 +400,6 @@ can be used, and offers full control over LoRa parameters.</p>
<span class="c1"># Enable interface if you want use it!</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># Allow transmit on interface. Setting</span>
<span class="c1"># this to false will create a listen-</span>
<span class="c1"># only interface.</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="n">true</span>
<span class="c1"># Serial port for the device</span>
<span class="n">port</span> <span class="o">=</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">ttyUSB0</span>
@@ -218,7 +446,6 @@ directly over a wire-pair, or for using devices such as data radios and lasers.<
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">Serial</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">SerialInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># Serial port for the device</span>
<span class="n">port</span> <span class="o">=</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">ttyUSB0</span>
@@ -241,7 +468,6 @@ for station identification purposes.</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="p">[[</span><span class="n">Packet</span> <span class="n">Radio</span> <span class="n">KISS</span> <span class="n">Interface</span><span class="p">]]</span>
<span class="nb">type</span> <span class="o">=</span> <span class="n">KISSInterface</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="n">true</span>
<span class="c1"># Serial port for the device</span>
<span class="n">port</span> <span class="o">=</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">ttyUSB1</span>
@@ -306,9 +532,6 @@ beaconing functionality described above.</p>
<span class="c1"># Enable interface if you want use it!</span>
<span class="n">interface_enabled</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># Allow transmit on interface.</span>
<span class="n">outgoing</span> <span class="o">=</span> <span class="kc">True</span>
<span class="c1"># Serial port for the device</span>
<span class="n">port</span> <span class="o">=</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">ttyUSB2</span>
@@ -356,9 +579,12 @@ beaconing functionality described above.</p>
<h3><a href="index.html">Table of Contents</a></h3>
<ul>
<li><a class="reference internal" href="#">Supported Interfaces</a><ul>
<li><a class="reference internal" href="#udp-interface">UDP Interface</a></li>
<li><a class="reference internal" href="#common-interface-options">Common Interface Options</a></li>
<li><a class="reference internal" href="#auto-interface">Auto Interface</a></li>
<li><a class="reference internal" href="#i2p-interface">I2P Interface</a></li>
<li><a class="reference internal" href="#tcp-server-interface">TCP Server Interface</a></li>
<li><a class="reference internal" href="#tcp-client-interface">TCP Client Interface</a></li>
<li><a class="reference internal" href="#udp-interface">UDP Interface</a></li>
<li><a class="reference internal" href="#rnode-lora-interface">RNode LoRa Interface</a></li>
<li><a class="reference internal" href="#serial-interface">Serial Interface</a></li>
<li><a class="reference internal" href="#kiss-interface">KISS Interface</a></li>
@@ -406,7 +632,7 @@ beaconing functionality described above.</p>
<li class="right" >
<a href="networks.html" title="Building Networks"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Supported Interfaces</a></li>
</ul>
</div>
+34 -8
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Building Networks &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Building Networks &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -31,7 +31,7 @@
<li class="right" >
<a href="using.html" title="Using Reticulum on Your System"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Building Networks</a></li>
</ul>
</div>
@@ -70,6 +70,13 @@ also very useful when just a few devices needs to communicate.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">Low-bandwidth networks, like LoRa and packet radio, can interoperate and
interconnect with much larger and higher bandwidth networks without issue.
Reticulum automatically manages the flow of information to and from various
network segments, and when bandwidth is limited, local traffic is prioritised.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">Reticulum provides sender/initiator anonymity by default. There is no way
to filter traffic or discriminate it based on the source of the traffic.</div>
</div>
@@ -97,24 +104,43 @@ transport node. Letting every node be a transport node will in most cases
degrade the performance and reliability of the network.</div>
</div>
<blockquote>
<div><p>In general terms, if a node is stationary, well-connected and kept running
<div><p><em>In general terms, if a node is stationary, well-connected and kept running
most of the time, it is a good candidate to be a transport node. For optimal
performance, a network should contain the amount of transport nodes that
provides connectivity to the intended area / topography, and not many more
than that.</p>
than that.</em></p>
</div></blockquote>
</li>
<li><div class="line-block">
<div class="line">Reticulum is designed to work reliably in open, trustless environments. This
means you can use it to create open-access networks, where participants can
join and leave in an free and unorganised manner. This property allows an
entirely new, and so far, mostly unexplored class of networked applications,
where networks, and the information flow within them can form and dissolve
organically.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">You can just as easily create closed networks, since Reticulum allows you to
add authentication to any interface. This means you can restrict access on
any interface type, even when using legacy devices, such as modems. You can
also mix authenticated and open interfaces on the same system. See the
<a class="reference internal" href="interfaces.html#interfaces-options"><span class="std std-ref">Common Interface Options</span></a> section of the <a class="reference internal" href="interfaces.html#interfaces-main"><span class="std std-ref">Interfaces</span></a>
chapter of this manual for information on how to set up interface authentication.</div>
</div>
</li>
</ul>
</div></blockquote>
<p>Reticulum allows you to mix very different kinds of networking mediums into a
unified mesh, or to keep everything within one medium. You could build a “virtual
network” running entirely over the Internet, where all nodes communicate over TCP
and UDP “channels”. You could also build such a network using MQTT or ZeroMQ as
the underlying carrier for Reticulum.</p>
and UDP “channels”. You could also build such a network using other already-established
communications channels as the underlying carrier for Reticulum.</p>
<p>However, most real-world networks will probably involve either some form of
wireless or direct hardline communications. To allow Reticulum to communicate
over any type of medium, you must specify it in the configuration file, by default
located at <code class="docutils literal notranslate"><span class="pre">~/.reticulum/config</span></code>.</p>
located at <code class="docutils literal notranslate"><span class="pre">~/.reticulum/config</span></code>. See the <a class="reference internal" href="interfaces.html#interfaces-main"><span class="std std-ref">Supported Interfaces</span></a>
chapter of this manual for interface configuration examples.</p>
<p>Any number of interfaces can be configured, and Reticulum will automatically
decide which are suitable to use in any given situation, depending on where
traffic needs to flow.</p>
@@ -246,7 +272,7 @@ connected outliers are now an integral part of the network.</p>
<li class="right" >
<a href="using.html" title="Using Reticulum on Your System"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Building Networks</a></li>
</ul>
</div>
Binary file not shown.
+31 -11
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>API Reference &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>API Reference &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -16,7 +16,7 @@
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="Examples" href="examples.html" />
<link rel="next" title="Code Examples" href="examples.html" />
<link rel="prev" title="Understanding Reticulum" href="understanding.html" />
</head><body>
<div class="related" role="navigation" aria-label="related navigation">
@@ -26,12 +26,12 @@
<a href="genindex.html" title="General Index"
accesskey="I">index</a></li>
<li class="right" >
<a href="examples.html" title="Examples"
<a href="examples.html" title="Code Examples"
accesskey="N">next</a> |</li>
<li class="right" >
<a href="understanding.html" title="Understanding Reticulum"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">API Reference</a></li>
</ul>
</div>
@@ -76,7 +76,7 @@ other programs to use on demand.</p>
<dt class="sig sig-object py" id="RNS.Reticulum.MTU">
<span class="sig-name descname"><span class="pre">MTU</span></span><em class="property"> <span class="pre">=</span> <span class="pre">500</span></em><a class="headerlink" href="#RNS.Reticulum.MTU" title="Permalink to this definition"></a></dt>
<dd><p>The MTU that Reticulum adheres to, and will expect other peers to
adhere to. By default, the MTU is 500 bytes. In custom RNS network
adhere to. By default, the MTU is 507 bytes. In custom RNS network
implementations, it is possible to change this value, but doing so will
completely break compatibility with all other RNS networks. An identical
MTU is a prerequisite for peers to communicate in the same network.</p>
@@ -84,6 +84,23 @@ MTU is a prerequisite for peers to communicate in the same network.</p>
the default value.</p>
</dd></dl>
<dl class="py attribute">
<dt class="sig sig-object py" id="RNS.Reticulum.ANNOUNCE_CAP">
<span class="sig-name descname"><span class="pre">ANNOUNCE_CAP</span></span><em class="property"> <span class="pre">=</span> <span class="pre">2</span></em><a class="headerlink" href="#RNS.Reticulum.ANNOUNCE_CAP" title="Permalink to this definition"></a></dt>
<dd><p>The maximum percentage of interface bandwidth that, at any given time,
may be used to propagate announces. If an announce was scheduled for
broadcasting on an interface, but doing so would exceed the allowed
bandwidth allocation, the announce will be queued for transmission
when there is bandwidth available.</p>
<p>Reticulum will always prioritise propagating announces with fewer
hops, ensuring that distant, large networks with many peers on fast
links dont overwhelm the capacity of smaller networks on slower
mediums. If an announce remains queued for an extended amount of time,
it will eventually be dropped.</p>
<p>This value will be applied by default to all created interfaces,
but it can be configured individually on a per-interface basis.</p>
</dd></dl>
<dl class="py method">
<dt class="sig sig-object py" id="RNS.Reticulum.should_use_implicit_proof">
<em class="property"><span class="pre">static</span> </em><span class="sig-name descname"><span class="pre">should_use_implicit_proof</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#RNS.Reticulum.should_use_implicit_proof" title="Permalink to this definition"></a></dt>
@@ -671,7 +688,7 @@ destinations, reticulum will use ephemeral keys, and offers <strong>Forward Secr
<dl class="py attribute">
<dt class="sig sig-object py" id="RNS.Packet.PLAIN_MDU">
<span class="sig-name descname"><span class="pre">PLAIN_MDU</span></span><em class="property"> <span class="pre">=</span> <span class="pre">477</span></em><a class="headerlink" href="#RNS.Packet.PLAIN_MDU" title="Permalink to this definition"></a></dt>
<span class="sig-name descname"><span class="pre">PLAIN_MDU</span></span><em class="property"> <span class="pre">=</span> <span class="pre">476</span></em><a class="headerlink" href="#RNS.Packet.PLAIN_MDU" title="Permalink to this definition"></a></dt>
<dd><p>The maximum size of the payload data in a single unencrypted packet</p>
</dd></dl>
@@ -1156,13 +1173,16 @@ Transport system of Reticulum.</p>
<dl class="py method">
<dt class="sig sig-object py" id="RNS.Transport.request_path">
<em class="property"><span class="pre">static</span> </em><span class="sig-name descname"><span class="pre">request_path</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">destination_hash</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#RNS.Transport.request_path" title="Permalink to this definition"></a></dt>
<em class="property"><span class="pre">static</span> </em><span class="sig-name descname"><span class="pre">request_path</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">destination_hash</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">on_interface</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#RNS.Transport.request_path" title="Permalink to this definition"></a></dt>
<dd><p>Requests a path to the destination from the network. If
another reachable peer on the network knows a path, it
will announce it.</p>
<dl class="field-list simple">
<dt class="field-odd">Parameters</dt>
<dd class="field-odd"><p><strong>destination_hash</strong> A destination hash as <em>bytes</em>.</p>
<dd class="field-odd"><ul class="simple">
<li><p><strong>destination_hash</strong> A destination hash as <em>bytes</em>.</p></li>
<li><p><strong>on_interface</strong> If specified, the path request will only be sent on this interface. In normal use, Reticulum handles this automatically, and this parameter should not be used.</p></li>
</ul>
</dd>
</dl>
</dd></dl>
@@ -1204,7 +1224,7 @@ will announce it.</p>
title="previous chapter">Understanding Reticulum</a></p>
<h4>Next topic</h4>
<p class="topless"><a href="examples.html"
title="next chapter">Examples</a></p>
title="next chapter">Code Examples</a></p>
<div role="note" aria-label="source link">
<h3>This Page</h3>
<ul class="this-page-menu">
@@ -1233,12 +1253,12 @@ will announce it.</p>
<a href="genindex.html" title="General Index"
>index</a></li>
<li class="right" >
<a href="examples.html" title="Examples"
<a href="examples.html" title="Code Examples"
>next</a> |</li>
<li class="right" >
<a href="understanding.html" title="Understanding Reticulum"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">API Reference</a></li>
</ul>
</div>
+3 -3
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Search &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Search &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -29,7 +29,7 @@
<li class="right" style="margin-right: 10px">
<a href="genindex.html" title="General Index"
accesskey="I">index</a></li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Search</a></li>
</ul>
</div>
@@ -85,7 +85,7 @@
<li class="right" style="margin-right: 10px">
<a href="genindex.html" title="General Index"
>index</a></li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Search</a></li>
</ul>
</div>
File diff suppressed because one or more lines are too long
+284 -194
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Understanding Reticulum &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Understanding Reticulum &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -17,7 +17,7 @@
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
<link rel="next" title="API Reference" href="reference.html" />
<link rel="prev" title="Building Networks" href="networks.html" />
<link rel="prev" title="Supported Interfaces" href="interfaces.html" />
</head><body>
<div class="related" role="navigation" aria-label="related navigation">
<h3>Navigation</h3>
@@ -29,9 +29,9 @@
<a href="reference.html" title="API Reference"
accesskey="N">next</a> |</li>
<li class="right" >
<a href="networks.html" title="Building Networks"
<a href="interfaces.html" title="Supported Interfaces"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Understanding Reticulum</a></li>
</ul>
</div>
@@ -43,44 +43,50 @@
<div class="section" id="understanding-reticulum">
<span id="understanding-main"></span><h1>Understanding Reticulum<a class="headerlink" href="#understanding-reticulum" title="Permalink to this headline"></a></h1>
<p>This chapter will briefly describe the overall purpose and operating principles of Reticulum, a
networking stack designed for reliable and secure communication over high-latency, low-bandwidth
links. It should give you an overview of how the stack works, and an understanding of how to
<p>This chapter will briefly describe the overall purpose and operating principles of Reticulum.
It should give you an overview of how the stack works, and an understanding of how to
develop networked applications using Reticulum.</p>
<p>This document is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the best place to go for such information is the Python reference implementation of Reticulum, along
with the code examples and API reference. It is however an essential resource to understanding the
general principles of Reticulum, how to apply them when creating your own networks or software.</p>
<p>This chapter is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the only complete repository, and final authority on how Reticulum actually functions, is the Python
reference implementation and API reference. That being said, this chapter is an essential resource in
understanding how Reticulum works from a high-level perspective, along with the general principles of
Reticulum, and how to apply them when creating your own networks or software.</p>
<p>After reading this document, you should be well-equipped to understand how a Reticulum network
operates, what it can achieve, and how you can use it yourself. If you want to help out with the
development, this is also the place to start, since it will provide a pretty clear overview of the
sentiments and the philosophy behind Reticulum.</p>
sentiments and the philosophy behind Reticulum, what problems it seeks to solve, and how it
approaches those solutions.</p>
<div class="section" id="motivation">
<span id="understanding-motivation"></span><h2>Motivation<a class="headerlink" href="#motivation" title="Permalink to this headline"></a></h2>
<p>The primary motivation for designing and implementing Reticulum has been the current lack of
reliable, functional and secure minimal-infrastructure modes of digital communication. It is my
belief that it is highly desirable to create a cheap and reliable way to set up a wide-range digital
communication network that can securely allow exchange of information between people and
belief that it is highly desirable to create a reliable and efficient way to set up long-range digital
communication networks that can securely allow exchange of information between people and
machines, with no central point of authority, control, censorship or barrier to entry.</p>
<p>Almost all of the various networking systems in use today share a common limitation, namely that they
require large amounts of coordination and trust to work, and to join the networks you need approval
<p>Almost all of the various networking systems in use today share a common limitation: They
require large amounts of coordination and centralised trust and power to function. To join such networks, you need approval
of gatekeepers in control. This need for coordination and trust inevitably leads to an environment of
central control, where its very easy for infrastructure operators or governments to control or alter
traffic, and censor or persecute unwanted actors.</p>
<p>Reticulum aims to require as little coordination and trust as possible. In fact, the only
“coordination” required is to know the characteristics of physical medium carrying Reticulum traffic.</p>
<p>Since Reticulum is completely medium agnostic, this could be whatever is best suited to the situation.
In some cases, this might be 1200 baud packet radio links over VHF frequencies, in other cases it might
be a microwave network using off-the-shelf radios. At the time of release of this document, the
recommended setup for development and testing is using LoRa radio modules with an open source firmware
(see the section <a class="reference internal" href="#understanding-referencesystem"><span class="std std-ref">Reference System Setup</span></a>), connected to a small
computer like a Raspberry Pi. As an example, the default reference setup provides a channel capacity
of 5.4 Kbps, and a usable direct node-to-node range of around 15 kilometers (indefinitely extendable
by using multiple hops).</p>
traffic, and censor or persecute unwanted actors. It also makes it completely impossible to freely deploy
and use networks at will, like one would use other common tools that enhance individual agency and freedom.</p>
<p>Reticulum aims to require as little coordination and trust as possible. It aims to make secure,
anonymous and permissionless networking and information exchange a tool that anyone can just pick up and use.</p>
<p>Since Reticulum is completely medium agnostic, it can be used to build networks on whatever is best
suited to the situation, or whatever you have available. In some cases, this might be packet radio
links over VHF frequencies, in other cases it might be a 2.4 GHz
network using off-the-shelf radios, or it might be using common LoRa development boards.</p>
<p>At the time of release of this document, the fastest and easiest setup for development and testing is using
LoRa radio modules with an open source firmware (see the section <a class="reference internal" href="#understanding-referencesystem"><span class="std std-ref">Reference Setup</span></a>),
connected to any kind of computer or mobile device that Reticulum can run on.</p>
<p>The ultimate aim of Reticulum is to allow anyone to be their own network operator, and to make it
cheap and easy to cover vast areas with a myriad of independent, interconnectable and autonomous networks.
Reticulum <strong>is not</strong> <em>one network</em>, it <strong>is a tool</strong> to build <em>thousands of networks</em>.</p>
<p>Networks without kill-switches, surveillance, censorship and control. Networks that can freely interoperate, associate and disassociate
with each other, and require no central oversight. Networks for human beings. <em>Networks for the people</em>.</p>
</div>
<div class="section" id="goals">
<span id="understanding-goals"></span><h2>Goals<a class="headerlink" href="#goals" title="Permalink to this headline"></a></h2>
<p>To be as widely usable and easy to use as possible, the following goals have been used to
<p>To be as widely usable and efficient to deploy as possible, the following goals have been used to
guide the design of Reticulum:</p>
<ul class="simple">
<li><dl class="simple">
@@ -90,22 +96,28 @@ critical to ensuring the availability, security and transparency of the system.<
</dl>
</li>
<li><dl class="simple">
<dt><strong>Hardware layer agnosticism</strong></dt><dd><p>Reticulum shall be fully hardware agnostic, and shall be useable over a wide range
<dt><strong>Hardware layer agnosticism</strong></dt><dd><p>Reticulum must be fully hardware agnostic, and shall be useable over a wide range of
physical networking layers, such as data radios, serial lines, modems, handheld transceivers,
wired ethernet, wifi, or anything else that can carry a digital data stream. Hardware made for
dedicated Reticulum use shall be as cheap as possible and use off-the-shelf components, so
it can be easily replicated.</p>
it can be easily modified and replicated by anyone interested in doing so.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Very low bandwidth requirements</strong></dt><dd><p>Reticulum should be able to function reliably over links with a transmission capacity as low
as <em>1,000 bps</em>.</p>
as <em>500 bits per second</em>.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Encryption by default</strong></dt><dd><p>Reticulum must use encryption by default where possible and applicable.</p>
<dt><strong>Encryption by default</strong></dt><dd><p>Reticulum must use strong encryption by default for all communication.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Initiator Anonymity</strong></dt><dd><p>It must be possible to communicate over a Reticulum network without revealing any identifying
information about oneself.</p>
</dd>
</dl>
</li>
@@ -118,10 +130,10 @@ by connecting a modem to a PMR or CB radio, or by using LoRa or WiFi modules.</p
</dl>
</li>
<li><dl class="simple">
<dt><strong>Supplied software</strong></dt><dd><p>Apart from the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic communication suite using Reticulum must be
implemented and released at the same time as Reticulum itself. This shall serve both as a
functional communication suite, and as an example and learning resource to others wishing
<dt><strong>Supplied software</strong></dt><dd><p>In addition to the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic set of Reticulum-based communication tools must be
implemented and released along with Reticulum itself. These shall serve both as a
functional, basic communication suite, and as an example and learning resource to others wishing
to build applications with Reticulum.</p>
</dd>
</dl>
@@ -129,7 +141,7 @@ to build applications with Reticulum.</p>
<li><dl class="simple">
<dt><strong>Ease of use</strong></dt><dd><p>The reference implementation of Reticulum is written in Python, to make it easy to use
and understand. A programmer with only basic experience should be able to use
Reticulum in their own applications.</p>
Reticulum to write networked applications.</p>
</dd>
</dl>
</li>
@@ -148,23 +160,31 @@ needs to be purchased.</p>
<p>Reticulum is a networking stack suited for high-latency, low-bandwidth links. Reticulum is at its
core a <em>message oriented</em> system. It is suited for both local point-to-point or point-to-multipoint
scenarios where alle nodes are within range of each other, as well as scenarios where packets need
to be transported over multiple hops to reach the recipient.</p>
to be transported over multiple hops in a complex network to reach the recipient.</p>
<p>Reticulum does away with the idea of addresses and ports known from IP, TCP and UDP. Instead
Reticulum uses the singular concept of <em>destinations</em>. Any application using Reticulum as its
networking stack will need to create one or more destinations to receive data, and know the
destinations it needs to send data to.</p>
<p>All destinations in Reticulum are represented internally as 10 bytes, derived from truncating a full
<p>All destinations in Reticulum are represented as a 10 byte hash, derived from truncating a full
SHA-256 hash of identifying characteristics of the destination. To users, the destination addresses
will be displayed as 10 bytes in hexadecimal representation, as in the following example: <code class="docutils literal notranslate"><span class="pre">&lt;80e29bf7cccaf31431b3&gt;</span></code>.</p>
<p>By default Reticulum encrypts all data using public-key cryptography. Any message sent to a
destination is encrypted with that destinations public key. Reticulum can also set up an encrypted
channel to a destination with <em>Perfect Forward Secrecy</em> and <em>Initiator Anonymity</em> using a elliptic
<p>The truncation size of 10 bytes (80 bits) for destinations has been choosen as a reasonable tradeoff between address space
and packet overhead. The address space accomodated by this size can support many billions of
simultaneously active devices on the same network, while keeping packet overhead low, which is
essential on low-bandwidth networks. In the very unlikely case that this address space nears
congestion, a one-line code change can upgrade the Reticulum address space all the way up to 256
bits, ensuring the Reticulum address space could potentially support galactic-scale networks.
This is obviusly complete and ridiculous over-allocation, and as such, the current 80 bits should
be sufficient, even far into the future.</p>
<p>By default Reticulum encrypts all data using elliptic curve cryptography. Any packet sent to a
destination is encrypted with a derived ephemeral key. Reticulum can also set up an encrypted
channel to a destination with <em>Forward Secrecy</em> and <em>Initiator Anonymity</em> using a elliptic
curve cryptography and ephemeral keys derived from a Diffie Hellman exchange on Curve25519. In
Reticulum terminology, this is called a <em>Link</em>.</p>
Reticulum terminology, this is called a <em>Link</em>. The multi-hop transport, coordination, verification
and reliability layers are fully autonomous and also based on elliptic curve cryptography.</p>
<p>Reticulum also offers symmetric key encryption for group-oriented communications, as well as
unencrypted packets for broadcast purposes, or situations where you need the communication to be in
plain text. The multi-hop transport, coordination, verification and reliability layers are fully
autonomous and based on public key cryptography.</p>
plain text.</p>
<p>Reticulum can connect to a variety of interfaces such as radio modems, data radios and serial ports,
and offers the possibility to easily tunnel Reticulum traffic over IP links such as the Internet or
private IP networks.</p>
@@ -174,23 +194,30 @@ private IP networks.</p>
destinations. Reticulum uses three different basic destination types, and one special:</p>
<ul class="simple">
<li><dl class="simple">
<dt><strong>Single</strong></dt><dd><p>The <em>single</em> destination type defines a public-key encrypted destination. Any data sent to this
destination will be encrypted with the destinations public key, and will only be readable by
the creator of the destination.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Group</strong></dt><dd><p>The <em>group</em> destination type defines a symmetrically encrypted destination. Data sent to this
destination will be encrypted with a symmetric key, and will be readable by anyone in
possession of the key. The <em>group</em> destination can be used just as well by only two peers, as it
can by many.</p>
<dt><strong>Single</strong></dt><dd><p>The <em>single</em> destination type is the most common type in Reticulum, and should be used for
most purposes. It is always identified by a unique public key. Any data sent to this
destination will be encrypted using ephemeral keys derived from an ECDH key exchange, and will
only be readable by the creator of the destination, who holds the corresponding private key.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Plain</strong></dt><dd><p>A <em>plain</em> destination type is unencrypted, and suited for traffic that should be broadcast to a
number of users, or should be readable by anyone. Traffic to a <em>plain</em> destination is not encrypted.</p>
number of users, or should be readable by anyone. Traffic to a <em>plain</em> destination is not encrypted.
Generally, <em>plain</em> destinations can be used for broadcast information intended to be public.
Plain destinations are only reachable directly, and packets adressed to plain destinations are
never transported over multiple hops in the network. To be transportable over multiple hops in Reticulum, information
<em>must</em> be encrypted, since Reticulum uses the per-packet encryption to verify routing paths and
keep them alive.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Group</strong></dt><dd><p>The <em>group</em> special destination type, that defines a symmetrically encrypted virtual destination.
Data sent to this destination will be encrypted with a symmetric key, and will be readable by
anyone in possession of the key, but as with the <em>plain</em> destination type, packets to this type
of destination are not currently transported over multiple hops, although a planned upgrade
to Reticulum will allow globally reachable <em>group</em> destinations.</p>
</dd>
</dl>
</li>
@@ -198,7 +225,8 @@ number of users, or should be readable by anyone. Traffic to a <em>plain</em> de
<dt><strong>Link</strong></dt><dd><p>A <em>link</em> is a special destination type, that serves as an abstract channel to a <em>single</em>
destination, directly connected or over multiple hops. The <em>link</em> also offers reliability and
more efficient encryption, forward secrecy, initiator anonymity, and as such can be useful even
when a node is directly reachable.</p>
when a node is directly reachable. It also offers a more capable API and allows easily carrying
out requests and responses, large data transfers and more.</p>
</dd>
</dl>
</li>
@@ -234,7 +262,7 @@ addressable, because their public keys will differ.</p></li>
<p>In actual use of <em>single</em> destination naming, it is advisable not to use any uniquely identifying
features in aspect naming. Aspect names should be general terms describing what kind of destination
is represented. The uniquely identifying aspect is always acheived by the appending the public key,
which expands the destination into a uniquely identifyable one.</p>
which expands the destination into a uniquely identifyable one. Reticulum does this automatically.</p>
<p>Any destination on a Reticulum network can be addressed and reached simply by knowning its
destination hash (and public key, but if the public key is not known, it can be requested from the
network simply by knowing the destination hash). The use of app names and aspects makes it easy to
@@ -254,7 +282,7 @@ indirectly, but must first be established through a <em>single</em> destination.
</dl>
</li>
<li><dl class="simple">
<dt><strong>Plain</strong></dt><dd><p>When plain-text communication is desirable, for example when broadcasting information.</p>
<dt><strong>Plain</strong></dt><dd><p>When plain-text communication is desirable, for example when broadcasting information, or for local discovery purposes.</p>
</dd>
</dl>
</li>
@@ -262,22 +290,23 @@ indirectly, but must first be established through a <em>single</em> destination.
<p>To communicate with a <em>single</em> destination, you need to know its public key. Any method for
obtaining the public key is valid, but Reticulum includes a simple mechanism for making other
nodes aware of your destinations public key, called the <em>announce</em>. It is also possible to request
an unknown public key from the network, as all participating nodes serve as a distributed ledger
an unknown public key from the network, as all transport instances serve as a distributed ledger
of public keys.</p>
<p>Note that public key information can be shared and verified in many other ways than using the
built-in <em>announce</em> functionality, and that it is therefore not required to use the announce/request
<p>Note that public key information can be shared and verified in other ways than using the
built-in <em>announce</em> functionality, and that it is therefore not required to use the <em>announce</em> and <em>path request</em>
functionality to obtain public keys. It is by far the easiest though, and should definitely be used
if there is not a good reason for doing it differently.</p>
if there is not a very good reason for doing it differently.</p>
</div>
</div>
<div class="section" id="public-key-announcements">
<span id="understanding-keyannouncements"></span><h3>Public Key Announcements<a class="headerlink" href="#public-key-announcements" title="Permalink to this headline"></a></h3>
<p>An <em>announce</em> will send a special packet over any configured interfaces, containing all needed
<p>An <em>announce</em> will send a special packet over any relevant interfaces, containing all needed
information about the destination hash and public key, and can also contain some additional,
application specific data. The entire packet is signed by the sender to ensure authenticity. It is not
required to use the announce functionality, but in many cases it will be the simplest way to share
public keys on the network. As an example, an announce in a simple messenger application might
contain the following information:</p>
public keys on the network. The announce mechanism also serves to establish end-to-end connectivity
to the announced destination, as the announce propagates through the network.</p>
<p>As an example, an announce in a simple messenger application might contain the following information:</p>
<ul class="simple">
<li><p>The announcers destination hash</p></li>
<li><p>The announcers public key</p></li>
@@ -289,29 +318,37 @@ contain the following information:</p>
destination to securely communicate with that destination. You might have noticed that there is one
piece of information lacking to reconstruct full knowledge of the announced destination, and that is
the aspect names of the destination. These are intentionally left out to save bandwidth, since they
will be implicit in almost all cases. If a destination name is not entirely implicit, information can be
included in the application specific data part that will allow the receiver to infer the naming.</p>
will be implicit in almost all cases. The receiving application will already know them. If a destination
name is not entirely implicit, information can be included in the application specific data part that
will allow the receiver to infer the naming.</p>
<p>It is important to note that announces will be forwarded throughout the network according to a
certain pattern. This will be detailed in the section
<a class="reference internal" href="#understanding-announce"><span class="std std-ref">The Announce Mechanism in Detail</span></a>.</p>
<p>In Reticulum, destinations are allowed to move around the network at will. This is very different from
protocols such as IP, where an address is always expected to stay within the network segment it was assigned in.
This limitation does not exist in Reticulum, and any destination is <em>completely portable</em> over the entire topography
of the network, and <em>can even be moved to other Reticulum networks</em> than the one it was created in, and
still become reachable. To update its reachability, a destination simply needs to send an announce on any
networks it is part of. After a short while, it will be globally reachable in the network.</p>
<p>Seeing how <em>single</em> destinations are always tied to a private/public key pair leads us to the next topic.</p>
</div>
<div class="section" id="understanding-identities">
<span id="identities"></span><h3>Identities<a class="headerlink" href="#understanding-identities" title="Permalink to this headline"></a></h3>
<p>In Reticulum, an <em>identity</em> does not necessarily represent a personal identity, but is an abstraction that
can represent any kind of <em>verified entity</em>. This could very well be a person, but it could also be the
can represent any kind of <em>verifiable entity</em>. This could very well be a person, but it could also be the
control interface of a machine, a program, robot, computer, sensor or something else entirely. In
general, any kind of agent that can act, or be acted upon, or store or manipulate information, can be
represented as an identity.</p>
<p>As we have seen, a <em>single</em> destination will always have an <em>identity</em> tied to it, but not <em>plain</em> or <em>group</em>
represented as an identity. An <em>identity</em> can be used to create any number of destinations.</p>
<p>A <em>single</em> destination will always have an <em>identity</em> tied to it, but not <em>plain</em> or <em>group</em>
destinations. Destinations and identities share a multilateral connection. You can create a
destination, and if it is not connected to an identity upon creation, it will just create a new one to use
automatically. This may be desirable in some situations, but often you will probably want to create
the identity first, and then link it to created destinations.</p>
<p>Building upon the simple messenger example, we could use an identity to represent the user of the
application. Destinations created will then be linked to this identity to allow communication to
reach the user. In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately.</p>
the identity first, and then use it to create new destinations.</p>
<p>As an example, we could use an identity to represent the user of a messaging application.
Destinations can then be created by this identity to allow communication to reach the user.
In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately, since obtaining access to the identity keys equals
obtaining access and controlling reachability to any destinations created by that identity.</p>
</div>
<div class="section" id="getting-further">
<span id="understanding-gettingfurther"></span><h3>Getting Further<a class="headerlink" href="#getting-further" title="Permalink to this headline"></a></h3>
@@ -324,65 +361,74 @@ hops in the network.</p>
</div>
<div class="section" id="reticulum-transport">
<span id="understanding-transport"></span><h2>Reticulum Transport<a class="headerlink" href="#reticulum-transport" title="Permalink to this headline"></a></h2>
<p>The term routing has been purposefully avoided until now. The current methods of routing used in IP-based
networks are fundamentally incompatible with the physical link types that Reticulum was designed to handle.
These routing methodologies assume trust at the physical layer, and often needs a lot more bandwidth than
Reticulum can assume is available.</p>
<p>Since Reticulum is designed to run over open radio spectrum, no such trust exists, and bandwidth is often
very limited. Existing routing protocols like BGP or OSPF carry too much overhead to be practically
useable over bandwidth-limited, high-latency links.</p>
<p>To overcome such challenges, Reticulums <em>Transport</em> system uses public-key cryptography to
implement the concept of <em>paths</em> that allow discovery of how to get information to a certain
<p>The methods of routing used in traditional networks are fundamentally incompatible with the physical medium
types and circumstances that Reticulum was designed to handle. These mechanisms mostly assume trust at the physical layer,
and often needs a lot more bandwidth than Reticulum can assume is available. Since Reticulum is designed to
survive running over open radio spectrum, no such trust can be assumed, and bandwidth is often very limited.</p>
<p>To overcome such challenges, Reticulums <em>Transport</em> system uses asymmetric elliptic curve cryptography to
implement the concept of <em>paths</em> that allow discovery of how to get information closer to a certain
destination. It is important to note that no single node in a Reticulum network knows the complete
path to a destination. Every Transport node participating in a Reticulum network will only
know what the most direct way to get a packet one hop closer to its destination is.</p>
know the most direct way to get a packet one hop closer to its destination.</p>
<div class="section" id="node-types">
<span id="understanding-nodetypes"></span><h3>Node Types<a class="headerlink" href="#node-types" title="Permalink to this headline"></a></h3>
<p>Currently, Reticulum distinguishes between two types of network nodes. All nodes on a Reticulum network
are <em>Reticulum Instances</em>, and some are alo <em>Transport Nodes</em>. If a system running Reticulum is fixed in
one place, and is intended to be kept available most of the time, it is a good contender to be a <em>Transport Node</em>.</p>
<p>Any Reticulum Instance can become a Transport Node by enabling it in the configuration.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for wider connectivity.</p>
<p>If a node is an <em>Instance</em> it should be given the configuration directive <code class="docutils literal notranslate"><span class="pre">enable_transport</span> <span class="pre">=</span> <span class="pre">No</span></code>, which
is the default setting.</p>
<p>If it is a <em>Transport Node</em>, it should be given the configuration directive <code class="docutils literal notranslate"><span class="pre">enable_transport</span> <span class="pre">=</span> <span class="pre">Yes</span></code>.</p>
</div>
<div class="section" id="the-announce-mechanism-in-detail">
<span id="understanding-announce"></span><h3>The Announce Mechanism in Detail<a class="headerlink" href="#the-announce-mechanism-in-detail" title="Permalink to this headline"></a></h3>
<p>When an <em>announce</em> is transmitted by a node, it will be forwarded by any node receiving it, but
according to some specific rules:</p>
<p>When an <em>announce</em> for a destination is transmitted by from a Reticulum instance, it will be forwarded by
any transport node receiving it, but according to some specific rules:</p>
<ul>
<li><div class="line-block">
<div class="line">If this exact announce has already been received before, ignore it.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">If not, record into a table which node the announce was received from, and how many times in
<div class="line">If not, record into a table which Transport Node the announce was received from, and how many times in
total it has been retransmitted to get here.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">If the announce has been retransmitted <em>m+1</em> times, it will not be forwarded. By default, <em>m</em> is
set to 18.</div>
<div class="line">If the announce has been retransmitted <em>m+1</em> times, it will not be forwarded any more. By default, <em>m</em> is
set to 128.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The announce will be assigned a delay <em>d</em> = c<sup>h</sup> seconds, where <em>c</em> is a decay constant, and <em>h</em> is the amount of times this packet has already been forwarded.</div>
<div class="line">After a randomised delay, the announce will be retransmitted on all interfaces that have bandwidth
available for processing announces. By default, the maximum bandwidth allocation for processing
announces is set at 2%, but can be configured on a per-interface basis.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">The packet will be given a priority <em>p = 1/d</em>.</div>
<div class="line">If any given interface does not have enough bandwidth available for retransmitting the announce,
the announce will be assigned a priority inversely proportional to its hop count, and be inserted
into a queue managed by the interface.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">If at least <em>d</em> seconds has passed since the announce was received, and no other packets with a
priority higher than <em>p</em> are waiting in the queue (see Packet Prioritisation), and the channel is
not utilized by other traffic, the announce will be forwarded.</div>
<div class="line">When the interface has bandwidth available for processing an announce, it will prioritise announces
for destinations that are closest in terms of hops, thus prioritising reachability and connectivity
of local nodes, even on slow networks that connect to wider and faster networks.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">If no other nodes are heard retransmitting the announce with a greater hop count than when
it left this node, transmitting it will be retried <em>r</em> times. By default, <em>r</em> is set to 1. Retries
follow same rules as above, with the exception that it must wait for at least <em>d</em> = c<sup>h+1</sup> +
t + rand(0, rw) seconds. This amount of time is equal to the amount of time it would take the next
node to retransmit the packet, plus a random window. By default, <em>t</em> is set to 10 seconds, and the
random window <em>rw</em> is set to 10 seconds.</div>
<div class="line">After the announce has been re-transmitted, and if no other nodes are heard retransmitting the announce
with a greater hop count than when it left this node, transmitting it will be retried <em>r</em> times. By default,
<em>r</em> is set to 1.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">If a newer announce from the same destination arrives, while an identical one is already in
the queue, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce, but will use <em>d</em> and <em>p</em> of the old
announce.</div>
<div class="line">If a newer announce from the same destination arrives, while an identical one is already waiting
to be transmitted, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce.</div>
</div>
</li>
</ul>
@@ -391,11 +437,15 @@ node will be able to reach the destination the announce originated from, simply
addressed to that destination. Any node with knowledge of the announce will be able to direct the
packet towards the destination by looking up the next node with the shortest amount of hops to the
destination.</p>
<p>According to these rules and default constants, an announce will propagate throughout the network
in a predictable way. In an example network utilising the default constants, and with an average link
distance of <em>Lavg =</em> 15 kilometers, an announce will be able to propagate outwards to a radius of 180
kilometers in 34 minutes, and a <em>maximum announce radius</em> of 270 kilometers in approximately 3
days.</p>
<p>According to these rules, an announce will propagate throughout the network in a predictable way,
and make the announced destination reachable in a short amount of time. Fast networks that have the
capacity to process many announces can reach full convergence very quickly, even when constantly adding
new destinations. Slower segments of such networks might take a bit longer to gain full knowledge about
the wide and fast networks they are connected to, but can still do so over time, while prioritising full
and quickly converging end-to-end connectivity for their local, slower segments.</p>
<p>In general, even extremely complex networks, that utilize the maximum 128 hops will converge to full
end-to-end connectivity in about one minute, given there is enough bandwidth available to process
the required amount of announces.</p>
</div>
<div class="section" id="reaching-the-destination">
<span id="understanding-paths"></span><h3>Reaching the Destination<a class="headerlink" href="#reaching-the-destination" title="Permalink to this headline"></a></h3>
@@ -453,7 +503,7 @@ strictly necessary to use one of the others.</div>
<ul>
<li><div class="line-block">
<div class="line">First, the node that wishes to establish a link will send out a special packet, that
traverses the network and locates the desired destination. Along the way, the nodes that
traverses the network and locates the desired destination. Along the way, the Transport Nodes that
forward the packet will take note of this <em>link request</em>.</div>
</div>
</li>
@@ -470,10 +520,10 @@ remember the <em>link</em> , and it can subsequently be used by referring to a h
</div>
</li>
<li><div class="line-block">
<div class="line">As a part of the <em>link request</em> , a Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes, using elliptic curve cryptography. As such,
this mode of communication is preferred, even for situations when nodes can directly communicate,
when the amount of data to be exchanged numbers in the tens of packets.</div>
<div class="line">As a part of the <em>link request</em>, an Elliptic Curve Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes. As such, this mode of communication is preferred,
even for situations when nodes can directly communicate, when the amount of data to be exchanged numbers
in the tens of packets, or whenever the use of the more advanced API functions is desired.</div>
</div>
</li>
<li><div class="line-block">
@@ -482,6 +532,12 @@ the same <em>proof</em> mechanism discussed before, so the sending node can obta
that the information reached the intended recipient.</div>
</div>
</li>
<li><div class="line-block">
<div class="line">Once the <em>link</em> has been set up, the initiator can remain anonymous, or choose to authenticate towards
the destination using a Reticulum Identity. This authentication is happening inside the encrypted
link, and is only revealed to the verified destination, and no intermediaries.</div>
</div>
</li>
</ul>
<p>In a moment, we will discuss the details of how this methodology is implemented, but lets first
recap what purposes this methodology serves. We first ensure that the node answering our request
@@ -575,102 +631,108 @@ the transfer is needed.</p>
<p>This is the purpose of the Reticulum <a class="reference internal" href="reference.html#api-resource"><span class="std std-ref">Resource</span></a>. A <em>Resource</em> can automatically
handle the reliable transfer of an arbitrary amount of data over an established <a class="reference internal" href="reference.html#api-link"><span class="std std-ref">Link</span></a>.
Resources can auto-compress data, will handle breaking the data into individual packets, sequencing
the transfer and reassembling the data on the other end.</p>
the transfer, integrity verification and reassembling the data on the other end.</p>
<p><a class="reference internal" href="reference.html#api-resource"><span class="std std-ref">Resources</span></a> are programmatically very simple to use, and only requires a few lines
of codes to reliably transfer any amount of data. They can be used to transfer data stored in memory,
or stream data directly from files.</p>
</div>
</div>
<div class="section" id="reference-system-setup">
<span id="understanding-referencesystem"></span><h2>Reference System Setup<a class="headerlink" href="#reference-system-setup" title="Permalink to this headline"></a></h2>
<p>This section will detail the recommended <em>Reference System Setup</em> for Reticulum. It is important to
note that Reticulum is designed to be usable over more or less any medium that allows you to send
and receive data in a digital form, and satisfies some very low minimum requirements. The
communication channel must support at least half-duplex operation, and provide an average
throughput of around 1000 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum software should be able to run on more or less any hardware that can provide a Python 3.x
<div class="section" id="reference-setup">
<span id="understanding-referencesystem"></span><h2>Reference Setup<a class="headerlink" href="#reference-setup" title="Permalink to this headline"></a></h2>
<p>This section will detail a recommended <em>Reference Setup</em> for Reticulum. It is important to
note that Reticulum is designed to be usable on more or less any computing device, and over more
or less any medium that allows you to send and receive data, which satisfies some very low
minimum requirements.</p>
<p>The communication channel must support at least half-duplex operation, and provide an average
throughput of around 500 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum stack should be able to run on more or less any hardware that can provide a Python 3.x
runtime environment.</p>
<p>That being said, the reference setup has been outlined to provide a common platform for anyone
<p>That being said, this reference setup has been outlined to provide a common platform for anyone
who wants to help in the development of Reticulum, and for everyone who wants to know a
recommended setup to get started. A reference system consists of three parts:</p>
recommended setup to get started experimenting. A reference system consists of three parts:</p>
<ul class="simple">
<li><dl class="simple">
<dt><strong>A channel access device</strong></dt><dd><p>Or <em>CAD</em> , in short, provides access to the physical medium whereupon the communication
<dt><strong>An Interface Device</strong></dt><dd><p>Which provides access to the physical medium whereupon the communication
takes place, for example a radio with an integrated modem. A setup with a separate modem
connected to a radio would also be termed a “channel access device.</p>
connected to a radio would also be an interface device.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>A host device</strong></dt><dd><p>Some sort of computing device that can run the necessary software, communicates with the
channel access device, and provides user interaction.</p>
<dt><strong>A Host Device</strong></dt><dd><p>Some sort of computing device that can run the necessary software, communicate with the
interface device, and provide user interaction.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>A software stack</strong></dt><dd><p>The software implementing the Reticulum protocol and applications using it.</p>
<dt><strong>A Software Stack</strong></dt><dd><p>The software implementing the Reticulum protocol and applications using it.</p>
</dd>
</dl>
</li>
</ul>
<p>The reference setup can be considered a relatively stable platform to develop on, and also to start
building networks on. While details of the implementation might change at the current stage of
building networks or applications on. While details of the implementation might change at the current stage of
development, it is the goal to maintain hardware compatibility for as long as entirely possible, and
the current reference setup has been determined to provide a functional platform for many years
into the future. The current Reference System Setup is as follows:</p>
<ul class="simple">
<li><dl class="simple">
<dt><strong>Channel Access Device</strong></dt><dd><p>A data radio consisting of a LoRa radio module, and a microcontroller with open source
<dt><strong>Interface Device</strong></dt><dd><p>A data radio consisting of a LoRa radio module, and a microcontroller with open source
firmware, that can connect to host devices via USB. It operates in either the 430, 868 or 900
MHz frequency bands. More details can be found on the <a class="reference external" href="https://unsigned.io/rnode">RNode Page</a>.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Host device</strong></dt><dd><p>Any computer device running Linux and Python. A Raspberry Pi with a Debian based OS is
<dt><strong>Host Device</strong></dt><dd><p>Any computer device running Linux and Python. A Raspberry Pi with a Debian based OS is
recommended.</p>
</dd>
</dl>
</li>
<li><dl class="simple">
<dt><strong>Software stack</strong></dt><dd><p>The current Reference Implementation Release of Reticulum, running on a Debian based
<dt><strong>Software Stack</strong></dt><dd><p>The most recently released Python Implementation of Reticulum, running on a Debian based
operating system.</p>
</dd>
</dl>
</li>
</ul>
<p>It is very important to note, that the reference channel access device <strong>does not</strong> use the LoRaWAN
standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an MCU with the correct firmware. Full details on how to
<p>To avoid confusion, it is very important to note, that the reference interface device <strong>does not</strong>
use the LoRaWAN standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an controller with the correct firmware. Full details on how to
get or make such a device is available on the <a class="reference external" href="https://unsigned.io/rnode">RNode Page</a>.</p>
<p>With the current reference setup, it should be possible to get on a Reticulum network for around 100$
even if you have none of the hardware already, and need to purchase everything.</p>
<p>This reference setup is of course just a recommendation for getting started easily, and you should
tailor it to your own specific needs, or whatever hardware you have available.</p>
</div>
<div class="section" id="protocol-specifics">
<span id="understanding-protocolspecifics"></span><h2>Protocol Specifics<a class="headerlink" href="#protocol-specifics" title="Permalink to this headline"></a></h2>
<p>This chapter will detail protocol specific information that is essential to the implementation of
Reticulum, but non critical in understanding how the protocol works on a general level. It should be
treated more as a reference than as essential reading.</p>
<div class="section" id="node-types">
<h3>Node Types<a class="headerlink" href="#node-types" title="Permalink to this headline"></a></h3>
<p>Currently Reticulum defines two node types, the <em>Station</em> and the <em>Peer</em>. A node is a <em>station</em> if it fixed
in one place, and if it is intended to be kept online most of the time. Otherwise the node is a <em>peer</em>.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for connectivity.</p>
<p>If a node is a <em>Peer</em> it should be given the configuration directive <code class="docutils literal notranslate"><span class="pre">enable_transport</span> <span class="pre">=</span> <span class="pre">No</span></code>.</p>
<p>If it is a <em>Station</em>, it should be given the configuration directive <code class="docutils literal notranslate"><span class="pre">enable_transport</span> <span class="pre">=</span> <span class="pre">Yes</span></code>.</p>
</div>
<div class="section" id="packet-prioritisation">
<h3>Packet Prioritisation<a class="headerlink" href="#packet-prioritisation" title="Permalink to this headline"></a></h3>
<p>Currently, Reticulum is completely priority-agnostic regarding general traffic. All traffic is handled
on a first-come, first-serve basis. Announce re-transmission are handled according to the re-transmission
times and priorities described earlier in this chapter.</p>
<p>It is possible that a prioritisation engine could be added to Reticulum in the future, but in
the light of Reticulums goal of equal access, doing so would need to be the subject of careful
investigation of the consequences first.</p>
</div>
<div class="section" id="binary-packet-format">
<span id="understanding-packetformat"></span><h3>Binary Packet Format<a class="headerlink" href="#binary-packet-format" title="Permalink to this headline"></a></h3>
<div class="section" id="interface-access-codes">
<h3>Interface Access Codes<a class="headerlink" href="#interface-access-codes" title="Permalink to this headline"></a></h3>
<p>Reticulum can create named virtual networks, and networks that are only accessible by knowing a preshared
passphrase. The configuration of this is detailed in the <a class="reference internal" href="interfaces.html#interfaces-options"><span class="std std-ref">Common Interface Options</span></a>
section. To implement these feature, Reticulum uses the concept of Interface Access Codes, that are calculated
and verified per packet.</p>
<p>An interface with a named virtual network or passphrase authentication enabled will derive a shared Ed25519
signing identity, and for every outbound packet generate a signature of the entire packet. This signature is
then inserted into the packet as an Interface Access Code before transmission. Depending on the speed and
capabilities of the interface, the IFAC can be the full 512-bit Ed25519 signature, or a truncated version.
Configured IFAC length can be inspected for all interfaces with the <code class="docutils literal notranslate"><span class="pre">rnstatus</span></code> utility.</p>
<p>Upon receipt, the interface will check that the signature matches the expected value, and drop the packet if it
does not. This ensures that only packets sent with the correct naming and/or passphrase parameters are allowed to
pass onto the network.</p>
</div>
<div class="section" id="wire-format">
<span id="understanding-packetformat"></span><h3>Wire Format<a class="headerlink" href="#wire-format" title="Permalink to this headline"></a></h3>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>== Reticulum Wire Format ======
A Reticulum packet is composed of the following fields:
@@ -678,9 +740,14 @@ A Reticulum packet is composed of the following fields:
[HEADER 2 bytes] [ADDRESSES 10/20 bytes] [CONTEXT 1 byte] [DATA 0-477 bytes]
* The HEADER field is 2 bytes long.
* Byte 1: [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 1: [IFAC Flag], [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 2: Number of hops
* Interface Access Code field if the IFAC flag was set.
* The length of the Interface Access Code can vary from
1 to 64 bytes according to physical interface
capabilities and configuration.
* The ADDRESSES field contains either 1 or 2 addresses.
* Each address is 10 bytes long.
* The Header Type flag in the HEADER field determines
@@ -693,12 +760,16 @@ A Reticulum packet is composed of the following fields:
* The DATA field is between 0 and 477 bytes.
* It contains the packets data payload.
IFAC Flag
-----------------
open 0 Packet for publically accessible interface
authenticated 1 Interface authentication is included in packet
Header Types
-----------------
type 1 00 Two byte header, one 10 byte address field
type 2 01 Two byte header, two 10 byte address fields
type 3 10 Reserved
type 4 11 Reserved
type 1 0 Two byte header, one 10 byte address field
type 2 1 Two byte header, two 10 byte address fields
Propagation Types
@@ -727,45 +798,63 @@ proof 11
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELDS CONTEXT FIELD DATA FIELD
_______|_______ ________________|________________ ________|______ __|_
| | | | | | | |
01010000 00000100 [ADDR1, 10 bytes] [ADDR2, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 4
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = TRANSPORT
+------------- Header Type = HEADER_2 (two byte header, two address fields)
01010000 00000100 [HASH1, 10 bytes] [HASH2, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 4
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = TRANSPORT
|+------------- Header Type = HEADER_2 (two byte header, two address fields)
+-------------- Access Codes = DISABLED
+- Packet Example -+
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ _______|_______ ________|______ __|_
| | | | | | | |
00000000 00000111 [ADDR1, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 7
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = BROADCAST
+------------- Header Type = HEADER_1 (two byte header, one address field)
00000000 00000111 [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = DISABLED
Size examples of different packet types
---------------------------------------
+- Packet Example -+
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size including all fields.
HEADER FIELD IFAC FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ ______|______ _______|_______ ________|______ __|_
| | | | | | | | | |
10000000 00000111 [IFAC, N bytes] [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = ENABLED
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
Size examples of different packet types
---------------------------------------
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size counting all fields including headers,
but excluding any interface access codes.
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
</pre></div>
</div>
</div>
@@ -795,6 +884,7 @@ proof 11
</ul>
</li>
<li><a class="reference internal" href="#reticulum-transport">Reticulum Transport</a><ul>
<li><a class="reference internal" href="#node-types">Node Types</a></li>
<li><a class="reference internal" href="#the-announce-mechanism-in-detail">The Announce Mechanism in Detail</a></li>
<li><a class="reference internal" href="#reaching-the-destination">Reaching the Destination</a><ul>
<li><a class="reference internal" href="#link-establishment-in-detail">Link Establishment in Detail</a></li>
@@ -803,11 +893,11 @@ proof 11
<li><a class="reference internal" href="#resources">Resources</a></li>
</ul>
</li>
<li><a class="reference internal" href="#reference-system-setup">Reference System Setup</a></li>
<li><a class="reference internal" href="#reference-setup">Reference Setup</a></li>
<li><a class="reference internal" href="#protocol-specifics">Protocol Specifics</a><ul>
<li><a class="reference internal" href="#node-types">Node Types</a></li>
<li><a class="reference internal" href="#packet-prioritisation">Packet Prioritisation</a></li>
<li><a class="reference internal" href="#binary-packet-format">Binary Packet Format</a></li>
<li><a class="reference internal" href="#interface-access-codes">Interface Access Codes</a></li>
<li><a class="reference internal" href="#wire-format">Wire Format</a></li>
</ul>
</li>
</ul>
@@ -815,8 +905,8 @@ proof 11
</ul>
<h4>Previous topic</h4>
<p class="topless"><a href="networks.html"
title="previous chapter">Building Networks</a></p>
<p class="topless"><a href="interfaces.html"
title="previous chapter">Supported Interfaces</a></p>
<h4>Next topic</h4>
<p class="topless"><a href="reference.html"
title="next chapter">API Reference</a></p>
@@ -851,9 +941,9 @@ proof 11
<a href="reference.html" title="API Reference"
>next</a> |</li>
<li class="right" >
<a href="networks.html" title="Building Networks"
<a href="interfaces.html" title="Supported Interfaces"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Understanding Reticulum</a></li>
</ul>
</div>
+130 -24
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>Using Reticulum on Your System &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>Using Reticulum on Your System &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -31,7 +31,7 @@
<li class="right" >
<a href="gettingstartedfast.html" title="Getting Started Fast"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Using Reticulum on Your System</a></li>
</ul>
</div>
@@ -45,15 +45,17 @@
<span id="using-main"></span><h1>Using Reticulum on Your System<a class="headerlink" href="#using-reticulum-on-your-system" title="Permalink to this headline"></a></h1>
<p>Reticulum is not installed as a driver or kernel module, as one might expect
of a networking stack. Instead, Reticulum is distributed as a Python module.
This means that no special privileges are required to install or use it.
This means that no special privileges are required to install or use it. It
is also very light-weight, and easy to transfer to and install on new systems.
Any program or application that uses Reticulum will automatically load and
initialise Reticulum when it starts.</p>
<p>In many cases, this approach is sufficient. When any program needs to use
Reticulum, it is loaded, initialised, interfaces are brought up, and the
program can now communicate over Reticulum. If another program starts up
and also wants access to the same Reticulum network, the instance is simply
shared. This works for any number of programs running concurrently, and is
very easy to use, but depending on your use case, there are other options.</p>
program can now communicate over any Reticulum networks available. If another
program starts up and also wants access to the same Reticulum network, the
instance is simply shared. This works for any number of programs running
concurrently, and is very easy to use, but depending on your use case, there
are other options.</p>
<div class="section" id="included-utility-programs">
<h2>Included Utility Programs<a class="headerlink" href="#included-utility-programs" title="Permalink to this headline"></a></h2>
<p>If you often use Reticulum from several different programs, or simply want
@@ -87,6 +89,7 @@ optional arguments:
--version show program&#39;s version number and exit
</pre></div>
</div>
<p>You can easily add <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> as an always-on service by <a class="reference internal" href="#using-systemd"><span class="std std-ref">configuring a service</span></a>.</p>
</div>
<div class="section" id="the-rnstatus-utility">
<h3>The rnstatus Utility<a class="headerlink" href="#the-rnstatus-utility" title="Permalink to this headline"></a></h3>
@@ -97,32 +100,48 @@ rnstatus
# Example output
Shared Instance[37428]
Status: Up
Connected applications: 1
RX: 1.13 KB
TX: 1.07 KB
Status : Up
Serving : 1 program
Rate : 1.00 Gbps
Traffic : 83.13 KB
86.10 KB↓
UDPInterface[Default UDP Interface/0.0.0.0:4242]
Status: Up
RX: 1.01 KB
TX: 1.01 KB
AutoInterface[Local]
Status : Up
Mode : Full
Rate : 10.00 Mbps
Peers : 1 reachable
Traffic : 63.23 KB↑
80.17 KB↓
TCPInterface[RNS Testnet Frankfurt/frankfurt.rns.unsigned.io:4965]
Status: Up
RX: 1.37 KB
TX: 9.02 KB
Status : Up
Mode : Full
Rate : 10.00 Mbps
Traffic : 187.27 KB↑
74.17 KB↓
RNodeInterface[RNode UHF]
Status : Up
Mode : Access Point
Rate : 1.30 kbps
Access : 64-bit IFAC by &lt;…e702c42ba8&gt;
Traffic : 8.49 KB↑
9.23 KB↓
Reticulum Transport Instance &lt;5245a8efe1788c6a70e1&gt; running
</pre></div>
</div>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>usage: rnsd [-h] [--config CONFIG] [-v] [-q] [--version]
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>usage: rnstatus [-h] [--config CONFIG] [--version] [-a] [-v]
Reticulum Network Stack Daemon
Reticulum Network Stack Status
optional arguments:
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
-v, --verbose
-q, --quiet
--version show program&#39;s version number and exit
-a, --all show all interfaces
-v, --verbose
</pre></div>
</div>
</div>
@@ -137,7 +156,8 @@ rnpath eca6f4e4dc26ae329e61
Path found, destination &lt;eca6f4e4dc26ae329e61&gt; is 4 hops away via &lt;56b115c30cd386cad69c&gt; on TCPInterface[Testnet/frankfurt.rns.unsigned.io:4965]
</pre></div>
</div>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>usage: rnpath.py [-h] [--config CONFIG] [--version] [-v] [destination]
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>usage: rnpath [-h] [--config CONFIG] [--version] [-t] [-d] [-w seconds] [-v]
[destination]
Reticulum Path Discovery Utility
@@ -148,6 +168,9 @@ optional arguments:
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
--version show program&#39;s version number and exit
-t, --table show all known paths
-d, --drop remove the path to a destination
-w seconds timeout before giving up
-v, --verbose
</pre></div>
</div>
@@ -184,6 +207,84 @@ optional arguments:
</div>
</div>
</div>
<div class="section" id="improving-system-configuration">
<h2>Improving System Configuration<a class="headerlink" href="#improving-system-configuration" title="Permalink to this headline"></a></h2>
<p>If you are setting up a system for permanent use with Reticulum, there is a
few system configuration changes that can make this easier to administrate.
These changes will be detailed here.</p>
<div class="section" id="fixed-serial-port-names">
<h3>Fixed Serial Port Names<a class="headerlink" href="#fixed-serial-port-names" title="Permalink to this headline"></a></h3>
<p>On a Reticulum instance with several serial port based interfaces, it can be
beneficial to use the fixed device names for the serial ports, instead
of the dynamically allocated shorthands such as <code class="docutils literal notranslate"><span class="pre">/dev/ttyUSB0</span></code>. Under most
Debian-based distributions, including Ubuntu and Raspberry Pi OS, these nodes
can be found under <code class="docutils literal notranslate"><span class="pre">/dev/serial/by-id</span></code>.</p>
<p>You can use such a device path directly in place of the numbered shorthands.
Here is an example of a packet radio TNC configured as such:</p>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>[[Packet Radio KISS Interface]]
type = KISSInterface
interface_enabled = True
outgoing = true
port = /dev/serial/by-id/usb-FTDI_FT230X_Basic_UART_43891CKM-if00-port0
speed = 115200
databits = 8
parity = none
stopbits = 1
preamble = 150
txtail = 10
persistence = 200
slottime = 20
</pre></div>
</div>
<p>Using this methodology avoids potential naming mix-ups where physical devices
might be plugged and unplugged in different orders, or when device name
assignment varies from one boot to another.</p>
</div>
<div class="section" id="reticulum-as-a-system-service">
<span id="using-systemd"></span><h3>Reticulum as a System Service<a class="headerlink" href="#reticulum-as-a-system-service" title="Permalink to this headline"></a></h3>
<p>Instead of starting Reticulum manually, you can install <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> as a system
service and have it start automatically at boot.</p>
<p>If you installed Reticulum with <code class="docutils literal notranslate"><span class="pre">pip</span></code>, the <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> program will most likely
be located in a user-local installation path only, which means <code class="docutils literal notranslate"><span class="pre">systemd</span></code> will not
be able to execute it. In this case, you can simply symlink the <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> program
into a directory that is in systemds path:</p>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>sudo ln -s $(which rnsd) /usr/local/bin/
</pre></div>
</div>
<p>You can then create the service file <code class="docutils literal notranslate"><span class="pre">/etc/systemd/system/rnsd.service</span></code> with the
following content:</p>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>[Unit]
Description=Reticulum Network Stack Daemon
After=multi-user.target
[Service]
# If you run Reticulum on WiFi devices,
# or other devices that need some extra
# time to initialise, you might want to
# add a short delay before Reticulum is
# started by systemd:
# ExecStartPre=/bin/sleep 10
Type=simple
Restart=always
RestartSec=3
User=USERNAMEHERE
ExecStart=rnsd --service
[Install]
WantedBy=multi-user.target
</pre></div>
</div>
<p>Be sure to replace <code class="docutils literal notranslate"><span class="pre">USERNAMEHERE</span></code> with the user you want to run <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> as.</p>
<p>To manually start <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> run:</p>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>sudo systemctl start rnsd
</pre></div>
</div>
<p>If you want to automatically start <code class="docutils literal notranslate"><span class="pre">rnsd</span></code> at boot, run:</p>
<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>sudo systemctl enable rnsd
</pre></div>
</div>
</div>
</div>
</div>
@@ -203,6 +304,11 @@ optional arguments:
<li><a class="reference internal" href="#the-rnprobe-utility">The rnprobe Utility</a></li>
</ul>
</li>
<li><a class="reference internal" href="#improving-system-configuration">Improving System Configuration</a><ul>
<li><a class="reference internal" href="#fixed-serial-port-names">Fixed Serial Port Names</a></li>
<li><a class="reference internal" href="#reticulum-as-a-system-service">Reticulum as a System Service</a></li>
</ul>
</li>
</ul>
</li>
</ul>
@@ -246,7 +352,7 @@ optional arguments:
<li class="right" >
<a href="gettingstartedfast.html" title="Getting Started Fast"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">Using Reticulum on Your System</a></li>
</ul>
</div>
+49 -28
View File
@@ -5,7 +5,7 @@
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<title>What is Reticulum? &#8212; Reticulum Network Stack 0.2.6 beta documentation</title>
<title>What is Reticulum? &#8212; Reticulum Network Stack 0.3.5 beta documentation</title>
<link rel="stylesheet" type="text/css" href="_static/pygments.css" />
<link rel="stylesheet" type="text/css" href="_static/classic.css" />
@@ -31,7 +31,7 @@
<li class="right" >
<a href="index.html" title="Reticulum Network Stack Manual"
accesskey="P">previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">What is Reticulum?</a></li>
</ul>
</div>
@@ -43,25 +43,22 @@
<div class="section" id="what-is-reticulum">
<h1>What is Reticulum?<a class="headerlink" href="#what-is-reticulum" title="Permalink to this headline"></a></h1>
<p>Reticulum is a cryptography-based networking stack for wide-area networks built on readily available hardware, and can operate even with very high latency and extremely low bandwidth.</p>
<p>Reticulum allows you to build very wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.</p>
<p>Reticulum is a cryptography-based networking stack for building wide-area networks with readily available hardware, that can continue to operate even with extremely low bandwidth and very high latency.</p>
<p>Reticulum allows you to build wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.</p>
<p>Reticulum is a complete networking stack, and does not need IP or higher layers, although it is easy to utilise IP (with TCP or UDP) as the underlying carrier for Reticulum. It is therefore trivial to tunnel Reticulum over the Internet or private IP networks. Reticulum is built directly on cryptographic principles, allowing resilience and stable functionality in open and trustless networks.</p>
<p>No kernel modules or drivers are required. Reticulum runs completely in userland, and can run on practically any system that runs Python 3. Reticulum runs well even on small single-board computers like the Pi Zero.</p>
<div class="section" id="current-status">
<h2>Current Status<a class="headerlink" href="#current-status" title="Permalink to this headline"></a></h2>
<p>Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered relatively stable at the moment, but could change if warranted.</p>
</div>
<div class="section" id="caveat-emptor">
<h2>Caveat Emptor<a class="headerlink" href="#caveat-emptor" title="Permalink to this headline"></a></h2>
<p>Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.</p>
<p>Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered stable at the moment, but could change if absolutely warranted.</p>
</div>
<div class="section" id="what-does-reticulum-offer">
<h2>What does Reticulum Offer?<a class="headerlink" href="#what-does-reticulum-offer" title="Permalink to this headline"></a></h2>
<ul class="simple">
<li><p>Coordination-less globally unique adressing and identification</p></li>
<li><p>Fully self-configuring multi-hop routing</p></li>
<li><p>Asymmetric X25519 encryption and Ed25519 signatures as a basis for all communication</p></li>
<li><p>Forward Secrecy with ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519</p></li>
<li><p>Complete initiator anonymity, communicate without revealing your identity</p></li>
<li><p>Asymmetric encryption based on X25519, and Ed25519 signatures as a basis for all communication</p></li>
<li><p>Forward Secrecy by using ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519</p></li>
<li><p>Reticulum uses the <a class="reference external" href="https://github.com/fernet/spec/blob/master/Spec.md">Fernet</a> specification for on-the-wire / over-the-air encryption</p>
<ul>
<li><p>All keys are ephemeral and derived from an ECDH key exchange on Curve25519</p></li>
@@ -73,6 +70,12 @@
<li><p>Unforgeable packet delivery confirmations</p></li>
<li><p>A variety of supported interface types</p></li>
<li><p>An intuitive and developer-friendly API</p></li>
<li><p>Efficient link establishment</p>
<ul>
<li><p>Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes</p></li>
<li><p>Low cost of keeping links open at only 0.62 bits per second</p></li>
</ul>
</li>
<li><p>Reliable and efficient transfer of arbritrary amounts of data</p>
<ul>
<li><p>Reticulum can handle a few bytes of data or files of many gigabytes</p></li>
@@ -80,25 +83,21 @@
<li><p>The API is very easy to use, and provides transfer progress</p></li>
</ul>
</li>
<li><p>Efficient link establishment</p>
<ul>
<li><p>Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes</p></li>
<li><p>Low cost of keeping links open at only 0.62 bits per second</p></li>
</ul>
</li>
<li><p>Authentication and virtual network segmentation on all supported interface types</p></li>
<li><p>Flexible scalability allowing extremely low-bandwidth networks to co-exist and interoperate with large, high-bandwidth networks</p></li>
</ul>
</div>
<div class="section" id="where-can-reticulum-be-used">
<h2>Where can Reticulum be Used?<a class="headerlink" href="#where-can-reticulum-be-used" title="Permalink to this headline"></a></h2>
<p>Over practically any medium that can support at least a half-duplex channel
with 1.000 bits per second throughput, and an MTU of 500 bytes. Data radios,
with 500 bits per second throughput, and an MTU of 500 bytes. Data radios,
modems, LoRa radios, serial lines, AX.25 TNCs, amateur radio digital modes,
ad-hoc WiFi, free-space optical links and similar systems are all examples
of the types of interfaces Reticulum was designed for.</p>
<p>An open-source LoRa-based interface called <a class="reference external" href="https://unsigned.io/rnode">RNode</a>
has been designed specifically for use with Reticulum. It is possible to build
yourself, or it can be purchased as a complete transceiver that just needs a
USB connection to the host.</p>
has been designed as an example transceiver that is very suitable for
Reticulum. It is possible to build it yourself, to transform a common LoRa
development board into one, or it can be purchased as a complete transceiver.</p>
<p>Reticulum can also be encapsulated over existing IP networks, so theres
nothing stopping you from using it over wired ethernet or your local WiFi
network, where itll work just as well. In fact, one of the strengths of
@@ -106,23 +105,45 @@ Reticulum is how easily it allows you to connect different mediums into a
self-configuring, resilient and encrypted mesh.</p>
<p>As an example, its possible to set up a Raspberry Pi connected to both a
LoRa radio, a packet radio TNC and a WiFi network. Once the interfaces are
configured, Reticulum will take care of the rest, and any device on the WiFi
added, Reticulum will take care of the rest, and any device on the WiFi
network can communicate with nodes on the LoRa and packet radio sides of the
network, and vice versa.</p>
</div>
<div class="section" id="interface-types-and-devices">
<h2>Interface Types and Devices<a class="headerlink" href="#interface-types-and-devices" title="Permalink to this headline"></a></h2>
<p>Reticulum implements a range of generalised interface types that covers most of the communications hardware that Reticulum can run over. If your hardware is not supported, its relatively simple to implement an interface class. Currently, the following interfaces are supported:</p>
<p>Reticulum implements a range of generalised interface types that covers the communications hardware that Reticulum can run over. If your hardware is not supported, its relatively simple to implement an interface class. Currently, Reticulum can use the following devices and communication mediums:</p>
<ul class="simple">
<li><p>Any ethernet device</p></li>
<li><p>LoRa using <a class="reference external" href="https://unsigned.io/rnode">RNode</a></p></li>
<li><p>Packet Radio TNCs, such as <a class="reference external" href="https://unsigned.io/openmodem">OpenModem</a></p></li>
<li><p>Any ethernet device</p>
<ul>
<li><p>WiFi devices</p></li>
<li><p>Wired ethernet devices</p></li>
<li><p>Fibre-optic transceivers</p></li>
<li><p>Data radios with ethernet ports</p></li>
</ul>
</li>
<li><p>LoRa using <a class="reference external" href="https://unsigned.io/rnode">RNode</a></p>
<ul>
<li><p>Can be installed on <a class="reference external" href="https://github.com/markqvist/rnodeconfigutil#supported-devices">many popular LoRa boards</a></p></li>
<li><p>Can be purchased as a <a class="reference external" href="https://unsigned.io/rnode">ready to use transceiver</a></p></li>
</ul>
</li>
<li><p>Packet Radio TNCs, such as <a class="reference external" href="https://unsigned.io/openmodem">OpenModem</a></p>
<ul>
<li><p>Any packet radio TNC in KISS mode</p></li>
<li><p>Ideal for VHF and UHF radio</p></li>
</ul>
</li>
<li><p>Any device with a serial port</p></li>
<li><p>The I2P network</p></li>
<li><p>TCP over IP networks</p></li>
<li><p>UDP over IP networks</p></li>
</ul>
<p>For a full list and more details, see the <a class="reference internal" href="interfaces.html#interfaces-main"><span class="std std-ref">Supported Interfaces</span></a> chapter.</p>
</div>
<div class="section" id="caveat-emptor">
<h2>Caveat Emptor<a class="headerlink" href="#caveat-emptor" title="Permalink to this headline"></a></h2>
<p>Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.</p>
</div>
</div>
@@ -136,10 +157,10 @@ network, and vice versa.</p>
<ul>
<li><a class="reference internal" href="#">What is Reticulum?</a><ul>
<li><a class="reference internal" href="#current-status">Current Status</a></li>
<li><a class="reference internal" href="#caveat-emptor">Caveat Emptor</a></li>
<li><a class="reference internal" href="#what-does-reticulum-offer">What does Reticulum Offer?</a></li>
<li><a class="reference internal" href="#where-can-reticulum-be-used">Where can Reticulum be Used?</a></li>
<li><a class="reference internal" href="#interface-types-and-devices">Interface Types and Devices</a></li>
<li><a class="reference internal" href="#caveat-emptor">Caveat Emptor</a></li>
</ul>
</li>
</ul>
@@ -183,7 +204,7 @@ network, and vice versa.</p>
<li class="right" >
<a href="index.html" title="Reticulum Network Stack Manual"
>previous</a> |</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.2.6 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-0"><a href="index.html">Reticulum Network Stack 0.3.5 beta documentation</a> &#187;</li>
<li class="nav-item nav-item-this"><a href="">What is Reticulum?</a></li>
</ul>
</div>
+2 -2
View File
@@ -22,7 +22,7 @@ copyright = '2021, Mark Qvist'
author = 'Mark Qvist'
# The full version, including alpha/beta/rc tags
release = '0.2.6 beta'
release = '0.3.5 beta'
# -- General configuration ---------------------------------------------------
@@ -65,4 +65,4 @@ html_static_path = ['_static']
# return False
# def setup(app):
# app.connect('autodoc-skip-member', check_skip_member)
# app.connect('autodoc-skip-member', check_skip_member)
+228 -9
View File
@@ -6,17 +6,37 @@ The best way to get started with the Reticulum Network Stack depends on what
you want to do. This guide will outline sensible starting paths for different
scenarios.
Try Using a Reticulum-based Program
=============================================
If you simply want to try using a program built with Reticulum, you can take
a look at `Nomad Network <https://github.com/markqvist/nomadnet>`_, which
provides a basic encrypted communications suite built completely on Reticulum.
If you simply want to try using a program built with Reticulum, a few different
programs exist that allow basic communication and a range of other useful functions
over even extremely low-bandwidth Reticulum networks.
These programs will let you get a feel for how Reticulum works. They have been designed
to run well over networks based on LoRa or packet radio, but can also be used completely
over local WiFi, wired ethernet, the Internet, or any combination.
As such, it is easy to get started experimenting, without having to set up any radio
transceivers or infrastructure just to try it out. Launching the programs on separate
devices connected to the same WiFi network is enough to get started, and physical
radio interfaces can then be added later.
Nomad Network
^^^^^^^^^^^^^
The terminal-based program `Nomad Network <https://github.com/markqvist/nomadnet>`_
provides a complete encrypted communications suite built with Reticulum. It features
encrypted messaging (both direct and delayed-delivery for offline users), file sharing,
and has a built-in text-browser and page server with support for dynamically rendered pages,
user authentication and more.
.. image:: screenshots/nomadnet_3.png
:target: _images/nomadnet_3.png
`Nomad Network <https://github.com/markqvist/nomadnet>`_ is a user-facing client
in the development for the messaging and information-sharing protocol
for the messaging and information-sharing protocol
`LXMF <https://github.com/markqvist/lxmf>`_, another project built with Reticulum.
You can install Nomad Network via pip:
@@ -29,28 +49,136 @@ You can install Nomad Network via pip:
# ... and run
nomadnet
**Please Note**: If this is the very first time you use pip to install a program
on your system, you might need to reboot your system for your program to become
available. If you get a "command not found" error or similar when running the
program, reboot your system and try again.
Sideband
^^^^^^^^
If you would rather use a program with a graphical user interface, you can take
a look at `Sideband <https://unsigned.io/sideband>`_, which is available for Android,
Linux and macOS.
.. image:: screenshots/sideband_1.png
:width: 400px
:align: center
:target: _images/sideband_1.png
Sideband is currently in the early stages of development, but already provides basic
communication features, and interoperates with Nomad Network, or any other LXMF client.
Using the Included Utilities
=============================================
Reticulum comes with a range of included utilities that make it easier to
manage your network, check connectivity and make Reticulum available to other
programs on your system.
You can use ``rnsd`` to run Reticulum as a background or foreground service,
and the ``rnstatus``, ``rnpath`` and ``rnprobe`` utilities to view and query
network status and connectivity.
To learn more about these utility programs, have a look at the
:ref:`Using Reticulum on Your System<using-main>` chapter of this manual.
Creating a Network With Reticulum
=============================================
To create a network, you will need to specify one or more *interfaces* for
Reticulum to use. This is done in the Reticulum configuration file, which by
default is located at ``~/.reticulum/config``.
default is located at ``~/.reticulum/config``. You can edit this file by hand,
or use the interactive ``rnsconfig`` utility.
When Reticulum is started for the first time, it will create a default
configuration file, with one active interface. This default interface uses
your existing ethernet network (if there is one), and only allows you to
communicate with other Reticulum peers within your local broadcast domain.
your existing ethernet and WiFi networks (if any), and only allows you to
communicate with other Reticulum peers within your local broadcast domains.
To communicate further, you will have to add one or more interfaces. The default
configuration includes a number of examples, ranging from using TCP over the
internet, to LoRa and Packet Radio interfaces.
With Reticulum, you only need to configure what interfaces you want to communicate
over. There is no need to configure address spaces, subnets, routing tables,
or other things you might be used to from other network types.
Once Reticulums knows which interfaces it should use, it will automatically
discover topography and configure transport of data to any destinations it
knows about.
In situations where you already have an established WiFi or ethernet network, and
many devices that want to utilise the same external Reticulum network (for example over
LoRa), it will often be sufficient to let one system act as a Reticulum gateway, by
adding any external interfaces to this systems configuration, and enabling transport. Any
other device on your local WiFi will then be able to connect to this wider Reticulum
network just using the default interface configuration.
Possibly, the examples in the config file are enough to get you started. If
you want more information, you can read the :ref:`Building Networks<networks-main>`
and :ref:`Interfaces<interfaces-main>` chapters of this manual.
Connecting Reticulum Instances Over the Internet
================================================
Reticulum currently offers two interfaces suitable for connecting instances over the Internet: :ref:`TCP<interfaces-tcps>`
and :ref:`I2P<interfaces-i2p>`. Each interface offers a different set of features, and Reticulum
users should carefully choose the interface which best suites their needs.
The ``TCPServerInterface`` allows users to host an instance accessible over TCP/IP. This
method is generally faster, lower latency, and more energy efficient than using ``I2PInterface``,
however it also leaks more data about the server host.
TCP connections reveal the IP address of both your instance and the server to anyone who can
inspect the connection. Someone could use this information to determine your location or identity. Adversaries
inspecting your packets may be able to record packet metadata like time of transmission and packet size.
Even though Reticulum encrypts traffic, TCP does not, so an adversary may be able to use
packet inspection to learn that a system is running Reticulum, and what other IP adresses connect to it.
Hosting a publicly reachable instance over TCP also requires a publicly reachable IP address,
which most Internet connections don't offer anymore.
The ``I2PInterface`` routes messages through the `Invisible Internet Protocol
(I2P) <https://geti2p.net/en/>`_. To properly use this interface, users must also run an I2P daemon in
parallel to ``rnsd``. For always-on I2P nodes it is recommended to use `i2pd <https://i2pd.website/>`_.
By default, I2P will encrypt and mix all traffic sent over the Internet, and
hide both the sender and receiver Reticulum instance IP addresses. Running an I2P node
will also relay other I2P user's encrypted packets, which will use extra
bandwidth and compute power, but also makes timing attacks and other forms of
deep-packet-inspection much more difficult.
I2P also allows users to host globally available Reticulum instances from non-public IPs and behind firewalls.
In general it is recommended to use an I2P node if you want to host a publically accessible
instance, while preserving anonymity. If you care more about performance, and a slightly
easier setup, use TCP.
Connect to the Public Testnet
===========================================
An experimental public testnet has been made accessible over both I2P and TCP. You can join it
by adding one of the following interfaces to your ``.reticulum/config`` file:
.. code::
# For connecting over TCP/IP:
[[RNS Testnet Frankfurt]]
type = TCPClientInterface
interface_enabled = yes
outgoing = True
target_host = frankfurt.rns.unsigned.io
target_port = 4965
# For connecting over I2P:
[[RNS Testnet I2P Node A]]
type = I2PInterface
interface_enabled = yes
peers = ykzlw5ujbaqc2xkec4cpvgyxj257wcrmmgkuxqmqcur7cq3w3lha.b32.i2p
Many other Reticulum instances are connecting to this testnet, and you can also join it
via other entry points if you know them. There is absolutely no control over the network
topography, usage or what types of instances connect. It will also occasionally be used
to test various failure scenarios, and there are no availability or service guarantees.
Develop a Program with Reticulum
===========================================
@@ -65,6 +193,13 @@ The above command will install Reticulum and dependencies, and you will be
ready to import and use RNS in your own programs. The next step will most
likely be to look at some :ref:`Example Programs<examples-main>`.
For extended functionality, you can install optional dependencies:
.. code::
pip3 install pyserial netifaces
Further information can be found in the :ref:`API Reference<api-main>`.
@@ -108,4 +243,88 @@ don't use pip, but try this recipe:
python3 Examples/Filetransfer.py -h
When you have experimented with the basic examples, it's time to go read the
:ref:`Understanding Reticulum<understanding-main>` chapter.
:ref:`Understanding Reticulum<understanding-main>` chapter.
Reticulum on ARM64
==============================================
On some architectures, including ARM64, not all dependencies have precompiled
binaries. On such systems, you will need to install ``python3-dev`` before
installing Reticulum or programs that depend on Reticulum.
.. code::
# Install Python and development packages
sudo apt update
sudo apt install python3 python3-pip python3-dev
# Install Reticulum
python3 -m pip install rns
Reticulum on Android
==============================================
Reticulum can be used on Android in different ways. The easiest way to get
started is using an app like `Sideband <https://unsigned.io/sideband>`_.
For more control and features, you can use Reticulum and related programs via
the `Termux app <https://termux.com/>`_, at the time of writing available on
`F-droid <https://f-droid.org>`_.
Termux is a terminal emulator and Linux environment for Android based devices,
which includes the ability to use many different programs and libraries,
including Reticulum.
Since the Python cryptography.io module does not offer pre-built wheels for
Android, the standard one-line install of Reticulum does not work on Android,
and a few extra commands are required.
From within Termux, execute the following:
.. code::
# First, make sure indexes and packages are up to date.
pkg update
pkg upgrade
# Then install dependencies for the cryptography library.
pkg install python build-essential openssl libffi rust
# Make sure pip is up to date, and install the wheel module.
pip3 install wheel pip --upgrade
# To allow the installer to build the cryptography module,
# we need to let it know what platform we are compiling for:
export CARGO_BUILD_TARGET="aarch64-linux-android"
# Start the install process for the cryptography module.
# Depending on your device, this can take several minutes,
# since the module must be compiled locally on your device.
pip3 install cryptography
# If the above installation succeeds, you can now install
# Reticulum and any related software
pip3 install rns
It is also possible to include Reticulum in apps compiled and distributed as
Android APKs. A detailed tutorial and example source code will be included
here at a later point.
Adding Radio Interfaces
==============================================
Once you have Reticulum installed and working, you can add radio interfaces with
any compatible hardware you have available. For information on how to configure
this, see the :ref:`Interfaces<interfaces-main>` section of this manual.
A range of common LoRa development boards and transceiver modules can be used
as interfaces with Reticulum. You can refer to the following external resources
for more information:
* `How To Make Your Own RNodes <https://unsigned.io/how-to-make-your-own-rnodes/>`_
* `Installing RNode Firmware on Compatible LoRa Devices <https://unsigned.io/installing-rnode-firmware-on-t-beam-and-lora32-devices/>`_
* `Private, Secure and Uncensorable Messaging Over a LoRa Mesh <https://unsigned.io/private-messaging-over-lora/>`_
* `RNode Firmware <https://github.com/markqvist/RNode_Firmware/>`_
If you have communications hardware that you think would be suitable for use with Reticulum,
you are welcome to head over to the `GitHub discussion pages <https://github.com/markqvist/Reticulum/discussions>`_
and propose adding an interface for the hardware.
+311 -73
View File
@@ -14,6 +14,308 @@ for Reticulum to use.
The following sections describe the interfaces currently available in Reticulum,
and gives example configurations for the respective interface types.
For a high-level overview of how networks can be formed over different interface
types, have a look at the :ref:`Building Networks<networks-main>` chapter of this
manual.
.. _interfaces-options:
Common Interface Options
========================
A number of general configuration options are available on most interfaces.
These can be used to control various aspects of interface behaviour.
* | The ``enabled`` option tells Reticulum whether or not
to bring up the interface. Defaults to ``False``. For any
interface to be brought up, the ``enabled`` option
must be set to ``True`` or ``Yes``.
* | The ``mode`` option allows selecting the high-level behaviour
of the interface from a number of options.
- The default value is ``full``. In this mode, all discovery,
meshing and transport functionality is available.
- In the ``access_point`` (or shorthand ``ap``) mode, the
interface will operate as a network access point. In this
mode, announces will not be automatically broadcasted on
the interface, and paths to destinations on the interface
will have a much shorter expiry time. This mode is useful
for creating interfaces that are mostly quiet, unless when
someone is actually using them. An example of this could
be a radio interface serving a wide area, where users are
expected to connect momentarily, use the network, and then
disappear again.
* | The ``outgoing`` option sets whether an interface is allowed
to transmit. Defaults to ``True``. If set to ``False`` or ``No``
the interface will only receive data, and never transmit.
* | The ``network_name`` option sets the virtual network name for
the interface. This allows multiple separate network segments
to exist on the same physical channel or medium.
* | The ``passphrase`` option sets an authentication passphrase on
the interface. This option can be used in conjunction with the
``network_name`` option, or be used alone.
* | The ``ifac_size`` option allows customising the length of the
Interface Authentication Codes carried by each packet on named
and/or authenticated network segments. It is set by default to
a size suitable for the interface in question, but can be set
to a custom size between 8 and 512 bits by using this option.
In normal usage, this option should not be changed from the
default.
* | The ``announce_cap`` option lets you configure the maximum
bandwidth to allocate, at any given time, to propagating
announces and other network upkeep traffic. It is configured at
2% by default, and should normally not need to be changed. Can
be set to any value between ``1`` and ``100``.
* | The ``bitrate`` option configures the interface bitrate.
Reticulum will use interface speeds reported by hardware, or
try to guess a suitable rate when the hardware doesn't report
any. In most cases, the automatically found rate should be
sufficient, but it can be configured by using the ``bitrate``
option, to set the interface speed in *bits per second*.
.. _interfaces-auto:
Auto Interface
==============
The Auto Interface enables communication with other discoverable Reticulum
nodes over autoconfigured IPv6 and UDP. It does not need any functional IP
infrastructure like routers or DHCP servers, but will require at least some
sort of switching medium between peers (a wired switch, a hub, a WiFi access
point or similar), and that link-local IPv6 is enabled in your operating
system, which should be enabled by default in almost all OSes.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[Default Interface]]
type = AutoInterface
interface_enabled = True
# You can create multiple isolated Reticulum
# networks on the same physical LAN by
# specifying different Group IDs.
group_id = reticulum
# You can also select specifically which
# kernel networking devices to use.
devices = wlan0,eth1
# Or let AutoInterface use all suitable
# devices except for a list of ignored ones.
ignored_devices = tun0,eth0
If you are connected to the Internet with IPv6, and your provider will route
IPv6 multicast, you can potentially configure the Auto Interface to globally
autodiscover other Reticulum nodes within your selected Group ID. You can specify
the discovery scope by setting it to one of ``link``, ``admin``, ``site``,
``organisation`` or ``global``.
.. code::
[[Default Interface]]
type = AutoInterface
interface_enabled = True
# Configure global discovery
group_id = custom_network_name
discovery_scope = global
# Other configuration options
discovery_port = 48555
data_port = 49555
.. _interfaces-i2p:
I2P Interface
=============
The I2P interface lets you connect Reticulum instances over the
`Invisible Internet Protocol <https://i2pd.website>`_. This can be
especially useful in cases where you want to host a globally reachable
Reticulum instance, but do not have access to any public IP addresses,
have a frequently changing IP address, or have firewalls blocking
inbound traffic.
Using the I2P interface, you will get a globally reachable, portable
and persistent I2P address that your Reticulum instance can be reached
at.
To use the I2P interface, you must have an I2P router running
on your system. The easiest way to acheive this is to download and
install the `latest release <https://github.com/PurpleI2P/i2pd/releases/latest>`_
of the ``ì2pd`` package. For more details about I2P, see the
`geti2p.net website <https://geti2p.net/en/about/intro>`_.`
When an I2P router is running on your system, you can simply add
an I2P interface to reticulum:
.. code::
[[I2P]]
type = I2PInterface
interface_enabled = yes
connectable = yes
On the first start, Reticulum will generate a new I2P address for the
interface and start listening for inbound traffic on it. This can take
a while the first time, especially if your I2P router was also just
started, and is not yet well-connected to the I2P network. When ready,
you should see I2P base32 address printed to your log file. You can
also inspect the status of the interface using the ``rnstatus`` utility.
To connect to other Reticulum instances over I2P, just add a comma-separated
list of I2P base32 addresses to the ``peers`` option of the interface:
.. code::
[[I2P]]
type = I2PInterface
interface_enabled = yes
connectable = yes
peers = 5urvjicpzi7q3ybztsef4i5ow2aq4soktfj7zedz53s47r54jnqq.b32.i2p
It can take anywhere from a few seconds to a few minutes to establish
I2P connections to the desired peers, so Reticulum handles the process
in the background, and will output relevant events to the log.
**Please Note!** While the I2P interface is the simplest way to use
Reticulum over I2P, it is also possible to tunnel the TCP server and
client interfaces over I2P manually. This can be useful in situations
where more control is needed, but requires manual tunnel setup through
the I2P daemon configuration.
It is important to note that the two methods are *interchangably compatible*.
You can use the I2PInterface to connect to a TCPServerInterface that
was manually tunneled over I2P, for example. This offers a high degree
of flexibility in network setup, while retaining ease of use in simpler
use-cases.
.. _interfaces-tcps:
TCP Server Interface
====================
The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[TCP Server Interface]]
type = TCPServerInterface
interface_enabled = True
# This configuration will listen on all IP
# interfaces on port 4242
listen_ip = 0.0.0.0
listen_port = 4242
# Alternatively you can bind to a specific IP
# listen_ip = 10.0.0.88
# listen_port = 4242
# Or a specific network device
# device = eth0
# port = 4242
**Please Note!** The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:
.. code::
[[TCP Server on I2P]]
type = TCPServerInterface
interface_enabled = yes
listen_ip = 127.0.0.1
listen_port = 5001
i2p_tunneled = yes
.. _interfaces-tcpc:
TCP Client Interface
====================
To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.
.. code::
# Here's an example of a TCP Client interface. The
# target_host can either be an IP address or a hostname.
[[TCP Client Interface]]
type = TCPClientInterface
interface_enabled = True
target_host = 127.0.0.1
target_port = 4242
It is also possible to use this interface type to connect via other programs
or hardware devices that expose a KISS interface on a TCP port, for example
software-based soundmodems. To do this, use the ``kiss_framing`` option:
.. code::
# Here's an example of a TCP Client interface that connects
# to a software TNC soundmodem on a KISS over TCP port.
[[TCP KISS Interface]]
type = TCPClientInterface
interface_enabled = True
kiss_framing = True
target_host = 127.0.0.1
target_port = 8001
**Caution!** Only use the KISS framing option when connecting to external devices
and programs like soundmodems and similar over TCP. When using the
``TCPClientInterface`` in conjunction with the ``TCPServerInterface`` you should
never enable ``kiss_framing``, since this will disable internal reliability and
recovery mechanisms that greatly improves performance over unreliable and
intermittent TCP links.
**Please Note!** The TCP interfaces support tunneling over I2P, but to do so reliably,
you must use the i2p_tunneled option:
.. code::
[[TCP Client over I2P]]
type = TCPClientInterface
interface_enabled = yes
target_host = 127.0.0.1
target_port = 5001
i2p_tunneled = yes
.. _interfaces-udp:
UDP Interface
@@ -24,6 +326,12 @@ private and the internet. It can also allow broadcast communication
over IP networks, so it can provide an easy way to enable connectivity
with all other peers on a local area network.
*Please Note!* Using broadcast UDP traffic has performance implications,
especially on WiFi. If your goal is simply to enable easy communication
with all peers in your local ethernet broadcast domain, the
:ref:`Auto Interface<interfaces-auto>` performs better, and is just as
easy to use.
The below example is enabled by default on new Reticulum installations,
as it provides an easy way to get started and to test Reticulum on a
pre-existing LAN.
@@ -36,7 +344,7 @@ pre-existing LAN.
[[Default UDP Interface]]
type = UDPInterface
interface_enabled = True
outgoing = True
listen_ip = 0.0.0.0
listen_port = 4242
forward_ip = 255.255.255.255
@@ -44,9 +352,7 @@ pre-existing LAN.
# The above configuration will allow communication
# within the local broadcast domains of all local
# IP interfaces. This is enabled by default as an
# easy way to get started, but you might want to
# consider altering it to something more specific.
# IP interfaces.
# Instead of specifying listen_ip, listen_port,
# forward_ip and forward_port, you can also bind
@@ -74,64 +380,6 @@ pre-existing LAN.
# forward_ip = 10.55.0.16
# forward_port = 4242
.. _interfaces-tcps:
TCP Server Interface
====================
The TCP Server interface is suitable for allowing other peers to connect over
the Internet or private IP networks. When a TCP server interface has been
configured, other Reticulum peers can connect to it with a TCP Client interface.
.. code::
# This example demonstrates a TCP server interface.
# It will listen for incoming connections on the
# specified IP address and port number.
[[TCP Server Interface]]
type = TCPServerInterface
interface_enabled = True
outgoing = True
# This configuration will listen on all IP
# interfaces on port 4242
listen_ip = 0.0.0.0
listen_port = 4242
# Alternatively you can bind to a specific IP
# listen_ip = 10.0.0.88
# listen_port = 4242
# Or a specific network device
# device = eth0
# port = 4242
.. _interfaces-tcpc:
TCP Client Interface
====================
To connect to a TCP server interface, you would naturally use the TCP client
interface. Many TCP Client interfaces from different peers can connect to the
same TCP Server interface at the same time.
.. code::
# Here's an example of a TCP Client interface. The
# target_host can either be an IP address or a hostname.
[[TCP Client Interface]]
type = TCPClientInterface
interface_enabled = True
outgoing = True
target_host = 127.0.0.1
target_port = 4242
.. _interfaces-rnode:
@@ -152,11 +400,6 @@ can be used, and offers full control over LoRa parameters.
# Enable interface if you want use it!
interface_enabled = True
# Allow transmit on interface. Setting
# this to false will create a listen-
# only interface.
outgoing = true
# Serial port for the device
port = /dev/ttyUSB0
@@ -207,7 +450,6 @@ directly over a wire-pair, or for using devices such as data radios and lasers.
[[Serial Interface]]
type = SerialInterface
interface_enabled = True
outgoing = True
# Serial port for the device
port = /dev/ttyUSB0
@@ -234,7 +476,6 @@ for station identification purposes.
[[Packet Radio KISS Interface]]
type = KISSInterface
interface_enabled = True
outgoing = true
# Serial port for the device
port = /dev/ttyUSB1
@@ -305,9 +546,6 @@ beaconing functionality described above.
# Enable interface if you want use it!
interface_enabled = True
# Allow transmit on interface.
outgoing = True
# Serial port for the device
port = /dev/ttyUSB2
@@ -339,4 +577,4 @@ beaconing functionality described above.
# Whether to use KISS flow-control.
# This is useful for modems with a
# small internal packet buffer.
flow_control = false
flow_control = false
+25 -5
View File
@@ -27,6 +27,11 @@ with Reticulum:
While the adress space can support billions of endpoints, Reticulum is
also very useful when just a few devices needs to communicate.
* | Low-bandwidth networks, like LoRa and packet radio, can interoperate and
interconnect with much larger and higher bandwidth networks without issue.
Reticulum automatically manages the flow of information to and from various
network segments, and when bandwidth is limited, local traffic is prioritised.
* | Reticulum provides sender/initiator anonymity by default. There is no way
to filter traffic or discriminate it based on the source of the traffic.
@@ -47,23 +52,38 @@ with Reticulum:
transport node. Letting every node be a transport node will in most cases
degrade the performance and reliability of the network.
In general terms, if a node is stationary, well-connected and kept running
*In general terms, if a node is stationary, well-connected and kept running
most of the time, it is a good candidate to be a transport node. For optimal
performance, a network should contain the amount of transport nodes that
provides connectivity to the intended area / topography, and not many more
than that.
than that.*
* | Reticulum is designed to work reliably in open, trustless environments. This
means you can use it to create open-access networks, where participants can
join and leave in an free and unorganised manner. This property allows an
entirely new, and so far, mostly unexplored class of networked applications,
where networks, and the information flow within them can form and dissolve
organically.
* | You can just as easily create closed networks, since Reticulum allows you to
add authentication to any interface. This means you can restrict access on
any interface type, even when using legacy devices, such as modems. You can
also mix authenticated and open interfaces on the same system. See the
:ref:`Common Interface Options<interfaces-options>` section of the :ref:`Interfaces<interfaces-main>`
chapter of this manual for information on how to set up interface authentication.
Reticulum allows you to mix very different kinds of networking mediums into a
unified mesh, or to keep everything within one medium. You could build a "virtual
network" running entirely over the Internet, where all nodes communicate over TCP
and UDP "channels". You could also build such a network using MQTT or ZeroMQ as
the underlying carrier for Reticulum.
and UDP "channels". You could also build such a network using other already-established
communications channels as the underlying carrier for Reticulum.
However, most real-world networks will probably involve either some form of
wireless or direct hardline communications. To allow Reticulum to communicate
over any type of medium, you must specify it in the configuration file, by default
located at ``~/.reticulum/config``.
located at ``~/.reticulum/config``. See the :ref:`Supported Interfaces<interfaces-main>`
chapter of this manual for interface configuration examples.
Any number of interfaces can be configured, and Reticulum will automatically
decide which are suitable to use in any given situation, depending on where
Binary file not shown.

After

Width:  |  Height:  |  Size: 79 KiB

+285 -186
View File
@@ -3,20 +3,21 @@
***********************
Understanding Reticulum
***********************
This chapter will briefly describe the overall purpose and operating principles of Reticulum, a
networking stack designed for reliable and secure communication over high-latency, low-bandwidth
links. It should give you an overview of how the stack works, and an understanding of how to
This chapter will briefly describe the overall purpose and operating principles of Reticulum.
It should give you an overview of how the stack works, and an understanding of how to
develop networked applications using Reticulum.
This document is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the best place to go for such information is the Python reference implementation of Reticulum, along
with the code examples and API reference. It is however an essential resource to understanding the
general principles of Reticulum, how to apply them when creating your own networks or software.
This chapter is not an exhaustive source of information on Reticulum, at least not yet. Currently,
the only complete repository, and final authority on how Reticulum actually functions, is the Python
reference implementation and API reference. That being said, this chapter is an essential resource in
understanding how Reticulum works from a high-level perspective, along with the general principles of
Reticulum, and how to apply them when creating your own networks or software.
After reading this document, you should be well-equipped to understand how a Reticulum network
operates, what it can achieve, and how you can use it yourself. If you want to help out with the
development, this is also the place to start, since it will provide a pretty clear overview of the
sentiments and the philosophy behind Reticulum.
sentiments and the philosophy behind Reticulum, what problems it seeks to solve, and how it
approaches those solutions.
.. _understanding-motivation:
@@ -25,34 +26,42 @@ Motivation
The primary motivation for designing and implementing Reticulum has been the current lack of
reliable, functional and secure minimal-infrastructure modes of digital communication. It is my
belief that it is highly desirable to create a cheap and reliable way to set up a wide-range digital
communication network that can securely allow exchange of information between people and
belief that it is highly desirable to create a reliable and efficient way to set up long-range digital
communication networks that can securely allow exchange of information between people and
machines, with no central point of authority, control, censorship or barrier to entry.
Almost all of the various networking systems in use today share a common limitation, namely that they
require large amounts of coordination and trust to work, and to join the networks you need approval
Almost all of the various networking systems in use today share a common limitation: They
require large amounts of coordination and centralised trust and power to function. To join such networks, you need approval
of gatekeepers in control. This need for coordination and trust inevitably leads to an environment of
central control, where it's very easy for infrastructure operators or governments to control or alter
traffic, and censor or persecute unwanted actors.
traffic, and censor or persecute unwanted actors. It also makes it completely impossible to freely deploy
and use networks at will, like one would use other common tools that enhance individual agency and freedom.
Reticulum aims to require as little coordination and trust as possible. In fact, the only
“coordination” required is to know the characteristics of physical medium carrying Reticulum traffic.
Reticulum aims to require as little coordination and trust as possible. It aims to make secure,
anonymous and permissionless networking and information exchange a tool that anyone can just pick up and use.
Since Reticulum is completely medium agnostic, this could be whatever is best suited to the situation.
In some cases, this might be 1200 baud packet radio links over VHF frequencies, in other cases it might
be a microwave network using off-the-shelf radios. At the time of release of this document, the
recommended setup for development and testing is using LoRa radio modules with an open source firmware
(see the section :ref:`Reference System Setup<understanding-referencesystem>`), connected to a small
computer like a Raspberry Pi. As an example, the default reference setup provides a channel capacity
of 5.4 Kbps, and a usable direct node-to-node range of around 15 kilometers (indefinitely extendable
by using multiple hops).
Since Reticulum is completely medium agnostic, it can be used to build networks on whatever is best
suited to the situation, or whatever you have available. In some cases, this might be packet radio
links over VHF frequencies, in other cases it might be a 2.4 GHz
network using off-the-shelf radios, or it might be using common LoRa development boards.
At the time of release of this document, the fastest and easiest setup for development and testing is using
LoRa radio modules with an open source firmware (see the section :ref:`Reference Setup<understanding-referencesystem>`),
connected to any kind of computer or mobile device that Reticulum can run on.
The ultimate aim of Reticulum is to allow anyone to be their own network operator, and to make it
cheap and easy to cover vast areas with a myriad of independent, interconnectable and autonomous networks.
Reticulum **is not** *one network*, it **is a tool** to build *thousands of networks*.
Networks without kill-switches, surveillance, censorship and control. Networks that can freely interoperate, associate and disassociate
with each other, and require no central oversight. Networks for human beings. *Networks for the people*.
.. _understanding-goals:
Goals
=====
To be as widely usable and easy to use as possible, the following goals have been used to
To be as widely usable and efficient to deploy as possible, the following goals have been used to
guide the design of Reticulum:
@@ -60,31 +69,34 @@ guide the design of Reticulum:
Reticulum must be implemented with, and be able to run using only open source software. This is
critical to ensuring the availability, security and transparency of the system.
* **Hardware layer agnosticism**
Reticulum shall be fully hardware agnostic, and shall be useable over a wide range
Reticulum must be fully hardware agnostic, and shall be useable over a wide range of
physical networking layers, such as data radios, serial lines, modems, handheld transceivers,
wired ethernet, wifi, or anything else that can carry a digital data stream. Hardware made for
dedicated Reticulum use shall be as cheap as possible and use off-the-shelf components, so
it can be easily replicated.
it can be easily modified and replicated by anyone interested in doing so.
* **Very low bandwidth requirements**
Reticulum should be able to function reliably over links with a transmission capacity as low
as *1,000 bps*.
as *500 bits per second*.
* **Encryption by default**
Reticulum must use encryption by default where possible and applicable.
Reticulum must use strong encryption by default for all communication.
* **Initiator Anonymity**
It must be possible to communicate over a Reticulum network without revealing any identifying
information about oneself.
* **Unlicensed use**
Reticulum shall be functional over physical communication mediums that do not require any
form of license to use. Reticulum must be designed in a way, so it is usable over ISM radio
frequency bands, and can provide functional long distance links in such conditions, for example
by connecting a modem to a PMR or CB radio, or by using LoRa or WiFi modules.
* **Supplied software**
Apart from the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic communication suite using Reticulum must be
implemented and released at the same time as Reticulum itself. This shall serve both as a
functional communication suite, and as an example and learning resource to others wishing
In addition to the core networking stack and API, that allows a developer to build
applications with Reticulum, a basic set of Reticulum-based communication tools must be
implemented and released along with Reticulum itself. These shall serve both as a
functional, basic communication suite, and as an example and learning resource to others wishing
to build applications with Reticulum.
* **Ease of use**
The reference implementation of Reticulum is written in Python, to make it easy to use
and understand. A programmer with only basic experience should be able to use
Reticulum in their own applications.
Reticulum to write networked applications.
* **Low cost**
It shall be as cheap as possible to deploy a communication system based on Reticulum. This
should be achieved by using cheap off-the-shelf hardware that potential users might already
@@ -99,27 +111,36 @@ Introduction & Basic Functionality
Reticulum is a networking stack suited for high-latency, low-bandwidth links. Reticulum is at its
core a *message oriented* system. It is suited for both local point-to-point or point-to-multipoint
scenarios where alle nodes are within range of each other, as well as scenarios where packets need
to be transported over multiple hops to reach the recipient.
to be transported over multiple hops in a complex network to reach the recipient.
Reticulum does away with the idea of addresses and ports known from IP, TCP and UDP. Instead
Reticulum uses the singular concept of *destinations*. Any application using Reticulum as its
networking stack will need to create one or more destinations to receive data, and know the
destinations it needs to send data to.
All destinations in Reticulum are represented internally as 10 bytes, derived from truncating a full
All destinations in Reticulum are represented as a 10 byte hash, derived from truncating a full
SHA-256 hash of identifying characteristics of the destination. To users, the destination addresses
will be displayed as 10 bytes in hexadecimal representation, as in the following example: ``<80e29bf7cccaf31431b3>``.
By default Reticulum encrypts all data using public-key cryptography. Any message sent to a
destination is encrypted with that destinations public key. Reticulum can also set up an encrypted
channel to a destination with *Perfect Forward Secrecy* and *Initiator Anonymity* using a elliptic
The truncation size of 10 bytes (80 bits) for destinations has been choosen as a reasonable tradeoff between address space
and packet overhead. The address space accomodated by this size can support many billions of
simultaneously active devices on the same network, while keeping packet overhead low, which is
essential on low-bandwidth networks. In the very unlikely case that this address space nears
congestion, a one-line code change can upgrade the Reticulum address space all the way up to 256
bits, ensuring the Reticulum address space could potentially support galactic-scale networks.
This is obviusly complete and ridiculous over-allocation, and as such, the current 80 bits should
be sufficient, even far into the future.
By default Reticulum encrypts all data using elliptic curve cryptography. Any packet sent to a
destination is encrypted with a derived ephemeral key. Reticulum can also set up an encrypted
channel to a destination with *Forward Secrecy* and *Initiator Anonymity* using a elliptic
curve cryptography and ephemeral keys derived from a Diffie Hellman exchange on Curve25519. In
Reticulum terminology, this is called a *Link*.
Reticulum terminology, this is called a *Link*. The multi-hop transport, coordination, verification
and reliability layers are fully autonomous and also based on elliptic curve cryptography.
Reticulum also offers symmetric key encryption for group-oriented communications, as well as
unencrypted packets for broadcast purposes, or situations where you need the communication to be in
plain text. The multi-hop transport, coordination, verification and reliability layers are fully
autonomous and based on public key cryptography.
plain text.
Reticulum can connect to a variety of interfaces such as radio modems, data radios and serial ports,
and offers the possibility to easily tunnel Reticulum traffic over IP links such as the Internet or
@@ -135,22 +156,30 @@ destinations. Reticulum uses three different basic destination types, and one sp
* **Single**
The *single* destination type defines a public-key encrypted destination. Any data sent to this
destination will be encrypted with the destinations public key, and will only be readable by
the creator of the destination.
* **Group**
The *group* destination type defines a symmetrically encrypted destination. Data sent to this
destination will be encrypted with a symmetric key, and will be readable by anyone in
possession of the key. The *group* destination can be used just as well by only two peers, as it
can by many.
The *single* destination type is the most common type in Reticulum, and should be used for
most purposes. It is always identified by a unique public key. Any data sent to this
destination will be encrypted using ephemeral keys derived from an ECDH key exchange, and will
only be readable by the creator of the destination, who holds the corresponding private key.
* **Plain**
A *plain* destination type is unencrypted, and suited for traffic that should be broadcast to a
number of users, or should be readable by anyone. Traffic to a *plain* destination is not encrypted.
Generally, *plain* destinations can be used for broadcast information intended to be public.
Plain destinations are only reachable directly, and packets adressed to plain destinations are
never transported over multiple hops in the network. To be transportable over multiple hops in Reticulum, information
*must* be encrypted, since Reticulum uses the per-packet encryption to verify routing paths and
keep them alive.
* **Group**
The *group* special destination type, that defines a symmetrically encrypted virtual destination.
Data sent to this destination will be encrypted with a symmetric key, and will be readable by
anyone in possession of the key, but as with the *plain* destination type, packets to this type
of destination are not currently transported over multiple hops, although a planned upgrade
to Reticulum will allow globally reachable *group* destinations.
* **Link**
A *link* is a special destination type, that serves as an abstract channel to a *single*
destination, directly connected or over multiple hops. The *link* also offers reliability and
more efficient encryption, forward secrecy, initiator anonymity, and as such can be useful even
when a node is directly reachable.
when a node is directly reachable. It also offers a more capable API and allows easily carrying
out requests and responses, large data transfers and more.
.. _understanding-destinationnaming:
@@ -192,7 +221,7 @@ packet.
In actual use of *single* destination naming, it is advisable not to use any uniquely identifying
features in aspect naming. Aspect names should be general terms describing what kind of destination
is represented. The uniquely identifying aspect is always acheived by the appending the public key,
which expands the destination into a uniquely identifyable one.
which expands the destination into a uniquely identifyable one. Reticulum does this automatically.
Any destination on a Reticulum network can be addressed and reached simply by knowning its
destination hash (and public key, but if the public key is not known, it can be requested from the
@@ -208,30 +237,32 @@ To recap, the different destination types should be used in the following situat
When private communication between two or more endpoints is needed. Supports multiple hops
indirectly, but must first be established through a *single* destination.
* **Plain**
When plain-text communication is desirable, for example when broadcasting information.
When plain-text communication is desirable, for example when broadcasting information, or for local discovery purposes.
To communicate with a *single* destination, you need to know its public key. Any method for
obtaining the public key is valid, but Reticulum includes a simple mechanism for making other
nodes aware of your destinations public key, called the *announce*. It is also possible to request
an unknown public key from the network, as all participating nodes serve as a distributed ledger
an unknown public key from the network, as all transport instances serve as a distributed ledger
of public keys.
Note that public key information can be shared and verified in many other ways than using the
built-in *announce* functionality, and that it is therefore not required to use the announce/request
Note that public key information can be shared and verified in other ways than using the
built-in *announce* functionality, and that it is therefore not required to use the *announce* and *path request*
functionality to obtain public keys. It is by far the easiest though, and should definitely be used
if there is not a good reason for doing it differently.
if there is not a very good reason for doing it differently.
.. _understanding-keyannouncements:
Public Key Announcements
------------------------
An *announce* will send a special packet over any configured interfaces, containing all needed
An *announce* will send a special packet over any relevant interfaces, containing all needed
information about the destination hash and public key, and can also contain some additional,
application specific data. The entire packet is signed by the sender to ensure authenticity. It is not
required to use the announce functionality, but in many cases it will be the simplest way to share
public keys on the network. As an example, an announce in a simple messenger application might
contain the following information:
public keys on the network. The announce mechanism also serves to establish end-to-end connectivity
to the announced destination, as the announce propagates through the network.
As an example, an announce in a simple messenger application might contain the following information:
* The announcers destination hash
@@ -244,13 +275,21 @@ With this information, any Reticulum node that receives it will be able to recon
destination to securely communicate with that destination. You might have noticed that there is one
piece of information lacking to reconstruct full knowledge of the announced destination, and that is
the aspect names of the destination. These are intentionally left out to save bandwidth, since they
will be implicit in almost all cases. If a destination name is not entirely implicit, information can be
included in the application specific data part that will allow the receiver to infer the naming.
will be implicit in almost all cases. The receiving application will already know them. If a destination
name is not entirely implicit, information can be included in the application specific data part that
will allow the receiver to infer the naming.
It is important to note that announces will be forwarded throughout the network according to a
certain pattern. This will be detailed in the section
:ref:`The Announce Mechanism in Detail<understanding-announce>`.
In Reticulum, destinations are allowed to move around the network at will. This is very different from
protocols such as IP, where an address is always expected to stay within the network segment it was assigned in.
This limitation does not exist in Reticulum, and any destination is *completely portable* over the entire topography
of the network, and *can even be moved to other Reticulum networks* than the one it was created in, and
still become reachable. To update it's reachability, a destination simply needs to send an announce on any
networks it is part of. After a short while, it will be globally reachable in the network.
Seeing how *single* destinations are always tied to a private/public key pair leads us to the next topic.
.. _understanding-identities:
@@ -259,21 +298,22 @@ Identities
----------
In Reticulum, an *identity* does not necessarily represent a personal identity, but is an abstraction that
can represent any kind of *verified entity*. This could very well be a person, but it could also be the
can represent any kind of *verifiable entity*. This could very well be a person, but it could also be the
control interface of a machine, a program, robot, computer, sensor or something else entirely. In
general, any kind of agent that can act, or be acted upon, or store or manipulate information, can be
represented as an identity.
represented as an identity. An *identity* can be used to create any number of destinations.
As we have seen, a *single* destination will always have an *identity* tied to it, but not *plain* or *group*
A *single* destination will always have an *identity* tied to it, but not *plain* or *group*
destinations. Destinations and identities share a multilateral connection. You can create a
destination, and if it is not connected to an identity upon creation, it will just create a new one to use
automatically. This may be desirable in some situations, but often you will probably want to create
the identity first, and then link it to created destinations.
the identity first, and then use it to create new destinations.
Building upon the simple messenger example, we could use an identity to represent the user of the
application. Destinations created will then be linked to this identity to allow communication to
reach the user. In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately.
As an example, we could use an identity to represent the user of a messaging application.
Destinations can then be created by this identity to allow communication to reach the user.
In all cases it is of great importance to store the private keys associated with any
Reticulum Identity securely and privately, since obtaining access to the identity keys equals
obtaining access and controlling reachability to any destinations created by that identity.
.. _understanding-gettingfurther:
@@ -292,57 +332,73 @@ In the following sections, two concepts that allow this will be introduced, *pat
Reticulum Transport
===================
The term routing has been purposefully avoided until now. The current methods of routing used in IP-based
networks are fundamentally incompatible with the physical link types that Reticulum was designed to handle.
These routing methodologies assume trust at the physical layer, and often needs a lot more bandwidth than
Reticulum can assume is available.
The methods of routing used in traditional networks are fundamentally incompatible with the physical medium
types and circumstances that Reticulum was designed to handle. These mechanisms mostly assume trust at the physical layer,
and often needs a lot more bandwidth than Reticulum can assume is available. Since Reticulum is designed to
survive running over open radio spectrum, no such trust can be assumed, and bandwidth is often very limited.
Since Reticulum is designed to run over open radio spectrum, no such trust exists, and bandwidth is often
very limited. Existing routing protocols like BGP or OSPF carry too much overhead to be practically
useable over bandwidth-limited, high-latency links.
To overcome such challenges, Reticulums *Transport* system uses public-key cryptography to
implement the concept of *paths* that allow discovery of how to get information to a certain
To overcome such challenges, Reticulums *Transport* system uses asymmetric elliptic curve cryptography to
implement the concept of *paths* that allow discovery of how to get information closer to a certain
destination. It is important to note that no single node in a Reticulum network knows the complete
path to a destination. Every Transport node participating in a Reticulum network will only
know what the most direct way to get a packet one hop closer to it's destination is.
know the most direct way to get a packet one hop closer to it's destination.
.. _understanding-nodetypes:
Node Types
----------
Currently, Reticulum distinguishes between two types of network nodes. All nodes on a Reticulum network
are *Reticulum Instances*, and some are alo *Transport Nodes*. If a system running Reticulum is fixed in
one place, and is intended to be kept available most of the time, it is a good contender to be a *Transport Node*.
Any Reticulum Instance can become a Transport Node by enabling it in the configuration.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for wider connectivity.
If a node is an *Instance* it should be given the configuration directive ``enable_transport = No``, which
is the default setting.
If it is a *Transport Node*, it should be given the configuration directive ``enable_transport = Yes``.
.. _understanding-announce:
The Announce Mechanism in Detail
--------------------------------
When an *announce* is transmitted by a node, it will be forwarded by any node receiving it, but
according to some specific rules:
When an *announce* for a destination is transmitted by from a Reticulum instance, it will be forwarded by
any transport node receiving it, but according to some specific rules:
* | If this exact announce has already been received before, ignore it.
* | If not, record into a table which node the announce was received from, and how many times in
* | If not, record into a table which Transport Node the announce was received from, and how many times in
total it has been retransmitted to get here.
* | If the announce has been retransmitted *m+1* times, it will not be forwarded. By default, *m* is
set to 18.
* | If the announce has been retransmitted *m+1* times, it will not be forwarded any more. By default, *m* is
set to 128.
* | The announce will be assigned a delay *d* = c\ :sup:`h` seconds, where *c* is a decay constant, and *h* is the amount of times this packet has already been forwarded.
* | After a randomised delay, the announce will be retransmitted on all interfaces that have bandwidth
available for processing announces. By default, the maximum bandwidth allocation for processing
announces is set at 2%, but can be configured on a per-interface basis.
* | The packet will be given a priority *p = 1/d*.
* | If any given interface does not have enough bandwidth available for retransmitting the announce,
the announce will be assigned a priority inversely proportional to it's hop count, and be inserted
into a queue managed by the interface.
* | If at least *d* seconds has passed since the announce was received, and no other packets with a
priority higher than *p* are waiting in the queue (see Packet Prioritisation), and the channel is
not utilized by other traffic, the announce will be forwarded.
* | When the interface has bandwidth available for processing an announce, it will prioritise announces
for destinations that are closest in terms of hops, thus prioritising reachability and connectivity
of local nodes, even on slow networks that connect to wider and faster networks.
* | If no other nodes are heard retransmitting the announce with a greater hop count than when
it left this node, transmitting it will be retried *r* times. By default, *r* is set to 1. Retries
follow same rules as above, with the exception that it must wait for at least *d* = c\ :sup:`h+1` +
t + rand(0, rw) seconds. This amount of time is equal to the amount of time it would take the next
node to retransmit the packet, plus a random window. By default, *t* is set to 10 seconds, and the
random window *rw* is set to 10 seconds.
* | After the announce has been re-transmitted, and if no other nodes are heard retransmitting the announce
with a greater hop count than when it left this node, transmitting it will be retried *r* times. By default,
*r* is set to 1.
* | If a newer announce from the same destination arrives, while an identical one is already in
the queue, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce, but will use *d* and *p* of the old
announce.
* | If a newer announce from the same destination arrives, while an identical one is already waiting
to be transmitted, the newest announce is discarded. If the newest announce contains different
application specific data, it will replace the old announce.
Once an announce has reached a node in the network, any other node in direct contact with that
node will be able to reach the destination the announce originated from, simply by sending a packet
@@ -350,11 +406,16 @@ addressed to that destination. Any node with knowledge of the announce will be a
packet towards the destination by looking up the next node with the shortest amount of hops to the
destination.
According to these rules and default constants, an announce will propagate throughout the network
in a predictable way. In an example network utilising the default constants, and with an average link
distance of *Lavg =* 15 kilometers, an announce will be able to propagate outwards to a radius of 180
kilometers in 34 minutes, and a *maximum announce radius* of 270 kilometers in approximately 3
days.
According to these rules, an announce will propagate throughout the network in a predictable way,
and make the announced destination reachable in a short amount of time. Fast networks that have the
capacity to process many announces can reach full convergence very quickly, even when constantly adding
new destinations. Slower segments of such networks might take a bit longer to gain full knowledge about
the wide and fast networks they are connected to, but can still do so over time, while prioritising full
and quickly converging end-to-end connectivity for their local, slower segments.
In general, even extremely complex networks, that utilize the maximum 128 hops will converge to full
end-to-end connectivity in about one minute, given there is enough bandwidth available to process
the required amount of announces.
.. _understanding-paths:
@@ -401,7 +462,7 @@ For exchanges of small amounts of information, Reticulum offers the *Packet* API
For exchanges of larger amounts of data, or when longer sessions of bidirectional communication is desired, Reticulum offers the *Link* API. To establish a *link*, the following process is employed:
* | First, the node that wishes to establish a link will send out a special packet, that
traverses the network and locates the desired destination. Along the way, the nodes that
traverses the network and locates the desired destination. Along the way, the Transport Nodes that
forward the packet will take note of this *link request*.
* | Second, if the destination accepts the *link request* , it will send back a packet that proves the
@@ -412,15 +473,19 @@ For exchanges of larger amounts of data, or when longer sessions of bidirectiona
* | When the validity of the *link* has been accepted by forwarding nodes, these nodes will
remember the *link* , and it can subsequently be used by referring to a hash representing it.
* | As a part of the *link request* , a Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes, using elliptic curve cryptography. As such,
this mode of communication is preferred, even for situations when nodes can directly communicate,
when the amount of data to be exchanged numbers in the tens of packets.
* | As a part of the *link request*, an Elliptic Curve Diffie-Hellman key exchange takes place, that sets up an
efficiently encrypted tunnel between the two nodes. As such, this mode of communication is preferred,
even for situations when nodes can directly communicate, when the amount of data to be exchanged numbers
in the tens of packets, or whenever the use of the more advanced API functions is desired.
* | When a *link* has been set up, it automatically provides message receipt functionality, through
the same *proof* mechanism discussed before, so the sending node can obtain verified confirmation
that the information reached the intended recipient.
* | Once the *link* has been set up, the initiator can remain anonymous, or choose to authenticate towards
the destination using a Reticulum Identity. This authentication is happening inside the encrypted
link, and is only revealed to the verified destination, and no intermediaries.
In a moment, we will discuss the details of how this methodology is implemented, but lets first
recap what purposes this methodology serves. We first ensure that the node answering our request
is actually the one we want to communicate with, and not a malicious actor pretending to be so.
@@ -507,7 +572,7 @@ the transfer is needed.
This is the purpose of the Reticulum :ref:`Resource<api-resource>`. A *Resource* can automatically
handle the reliable transfer of an arbitrary amount of data over an established :ref:`Link<api-link>`.
Resources can auto-compress data, will handle breaking the data into individual packets, sequencing
the transfer and reassembling the data on the other end.
the transfer, integrity verification and reassembling the data on the other end.
:ref:`Resources<api-resource>` are programmatically very simple to use, and only requires a few lines
of codes to reliably transfer any amount of data. They can be used to transfer data stored in memory,
@@ -515,57 +580,62 @@ or stream data directly from files.
.. _understanding-referencesystem:
Reference System Setup
Reference Setup
======================
This section will detail the recommended *Reference System Setup* for Reticulum. It is important to
note that Reticulum is designed to be usable over more or less any medium that allows you to send
and receive data in a digital form, and satisfies some very low minimum requirements. The
communication channel must support at least half-duplex operation, and provide an average
throughput of around 1000 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum software should be able to run on more or less any hardware that can provide a Python 3.x
This section will detail a recommended *Reference Setup* for Reticulum. It is important to
note that Reticulum is designed to be usable on more or less any computing device, and over more
or less any medium that allows you to send and receive data, which satisfies some very low
minimum requirements.
The communication channel must support at least half-duplex operation, and provide an average
throughput of around 500 bits per second, and supports a physical layer MTU of 500 bytes. The
Reticulum stack should be able to run on more or less any hardware that can provide a Python 3.x
runtime environment.
That being said, the reference setup has been outlined to provide a common platform for anyone
That being said, this reference setup has been outlined to provide a common platform for anyone
who wants to help in the development of Reticulum, and for everyone who wants to know a
recommended setup to get started. A reference system consists of three parts:
recommended setup to get started experimenting. A reference system consists of three parts:
* **A channel access device**
Or *CAD* , in short, provides access to the physical medium whereupon the communication
* **An Interface Device**
Which provides access to the physical medium whereupon the communication
takes place, for example a radio with an integrated modem. A setup with a separate modem
connected to a radio would also be termed a “channel access device.
* **A host device**
Some sort of computing device that can run the necessary software, communicates with the
channel access device, and provides user interaction.
* **A software stack**
connected to a radio would also be an interface device.
* **A Host Device**
Some sort of computing device that can run the necessary software, communicate with the
interface device, and provide user interaction.
* **A Software Stack**
The software implementing the Reticulum protocol and applications using it.
The reference setup can be considered a relatively stable platform to develop on, and also to start
building networks on. While details of the implementation might change at the current stage of
building networks or applications on. While details of the implementation might change at the current stage of
development, it is the goal to maintain hardware compatibility for as long as entirely possible, and
the current reference setup has been determined to provide a functional platform for many years
into the future. The current Reference System Setup is as follows:
* **Channel Access Device**
* **Interface Device**
A data radio consisting of a LoRa radio module, and a microcontroller with open source
firmware, that can connect to host devices via USB. It operates in either the 430, 868 or 900
MHz frequency bands. More details can be found on the `RNode Page <https://unsigned.io/rnode>`_.
* **Host device**
* **Host Device**
Any computer device running Linux and Python. A Raspberry Pi with a Debian based OS is
recommended.
* **Software stack**
The current Reference Implementation Release of Reticulum, running on a Debian based
* **Software Stack**
The most recently released Python Implementation of Reticulum, running on a Debian based
operating system.
It is very important to note, that the reference channel access device **does not** use the LoRaWAN
standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an MCU with the correct firmware. Full details on how to
To avoid confusion, it is very important to note, that the reference interface device **does not**
use the LoRaWAN standard, but uses a custom MAC layer on top of the plain LoRa modulation! As such, you will
need a plain LoRa radio module connected to an controller with the correct firmware. Full details on how to
get or make such a device is available on the `RNode Page <https://unsigned.io/rnode>`_.
With the current reference setup, it should be possible to get on a Reticulum network for around 100$
even if you have none of the hardware already, and need to purchase everything.
This reference setup is of course just a recommendation for getting started easily, and you should
tailor it to your own specific needs, or whatever hardware you have available.
.. _understanding-protocolspecifics:
Protocol Specifics
@@ -576,19 +646,6 @@ Reticulum, but non critical in understanding how the protocol works on a general
treated more as a reference than as essential reading.
Node Types
----------
Currently Reticulum defines two node types, the *Station* and the *Peer*. A node is a *station* if it fixed
in one place, and if it is intended to be kept online most of the time. Otherwise the node is a *peer*.
This distinction is made by the user configuring the node, and is used to determine what nodes on the
network will help forward traffic, and what nodes rely on other nodes for connectivity.
If a node is a *Peer* it should be given the configuration directive ``enable_transport = No``.
If it is a *Station*, it should be given the configuration directive ``enable_transport = Yes``.
Packet Prioritisation
---------------------
@@ -596,15 +653,30 @@ Currently, Reticulum is completely priority-agnostic regarding general traffic.
on a first-come, first-serve basis. Announce re-transmission are handled according to the re-transmission
times and priorities described earlier in this chapter.
It is possible that a prioritisation engine could be added to Reticulum in the future, but in
the light of Reticulums goal of equal access, doing so would need to be the subject of careful
investigation of the consequences first.
Interface Access Codes
----------------------
Reticulum can create named virtual networks, and networks that are only accessible by knowing a preshared
passphrase. The configuration of this is detailed in the :ref:`Common Interface Options<interfaces-options>`
section. To implement these feature, Reticulum uses the concept of Interface Access Codes, that are calculated
and verified per packet.
An interface with a named virtual network or passphrase authentication enabled will derive a shared Ed25519
signing identity, and for every outbound packet generate a signature of the entire packet. This signature is
then inserted into the packet as an Interface Access Code before transmission. Depending on the speed and
capabilities of the interface, the IFAC can be the full 512-bit Ed25519 signature, or a truncated version.
Configured IFAC length can be inspected for all interfaces with the ``rnstatus`` utility.
Upon receipt, the interface will check that the signature matches the expected value, and drop the packet if it
does not. This ensures that only packets sent with the correct naming and/or passphrase parameters are allowed to
pass onto the network.
.. _understanding-packetformat:
Binary Packet Format
--------------------
Wire Format
-----------
.. code-block:: text
@@ -615,9 +687,14 @@ Binary Packet Format
[HEADER 2 bytes] [ADDRESSES 10/20 bytes] [CONTEXT 1 byte] [DATA 0-477 bytes]
* The HEADER field is 2 bytes long.
* Byte 1: [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 1: [IFAC Flag], [Header Type], [Propagation Type], [Destination Type] and [Packet Type]
* Byte 2: Number of hops
* Interface Access Code field if the IFAC flag was set.
* The length of the Interface Access Code can vary from
1 to 64 bytes according to physical interface
capabilities and configuration.
* The ADDRESSES field contains either 1 or 2 addresses.
* Each address is 10 bytes long.
* The Header Type flag in the HEADER field determines
@@ -630,12 +707,16 @@ Binary Packet Format
* The DATA field is between 0 and 477 bytes.
* It contains the packets data payload.
IFAC Flag
-----------------
open 0 Packet for publically accessible interface
authenticated 1 Interface authentication is included in packet
Header Types
-----------------
type 1 00 Two byte header, one 10 byte address field
type 2 01 Two byte header, two 10 byte address fields
type 3 10 Reserved
type 4 11 Reserved
type 1 0 Two byte header, one 10 byte address field
type 2 1 Two byte header, two 10 byte address fields
Propagation Types
@@ -664,42 +745,60 @@ Binary Packet Format
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELDS CONTEXT FIELD DATA FIELD
_______|_______ ________________|________________ ________|______ __|_
| | | | | | | |
01010000 00000100 [ADDR1, 10 bytes] [ADDR2, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 4
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = TRANSPORT
+------------- Header Type = HEADER_2 (two byte header, two address fields)
01010000 00000100 [HASH1, 10 bytes] [HASH2, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 4
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = TRANSPORT
|+------------- Header Type = HEADER_2 (two byte header, two address fields)
+-------------- Access Codes = DISABLED
+- Packet Example -+
+- Packet Example -+
HEADER FIELD ADDRESSES FIELD CONTEXT FIELD DATA FIELD
HEADER FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ _______|_______ ________|______ __|_
| | | | | | | |
00000000 00000111 [ADDR1, 10 bytes] [CONTEXT, 1 byte] [DATA]
| | | | |
| | | | +-- Hops = 7
| | | +------- Packet Type = DATA
| | +--------- Destination Type = SINGLE
| +----------- Propagation Type = BROADCAST
+------------- Header Type = HEADER_1 (two byte header, one address field)
00000000 00000111 [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = DISABLED
Size examples of different packet types
---------------------------------------
+- Packet Example -+
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size including all fields.
HEADER FIELD IFAC FIELD DESTINATION FIELD CONTEXT FIELD DATA FIELD
_______|_______ ______|______ _______|_______ ________|______ __|_
| | | | | | | | | |
10000000 00000111 [IFAC, N bytes] [HASH1, 10 bytes] [CONTEXT, 1 byte] [DATA]
|| | | | |
|| | | | +-- Hops = 0
|| | | +------- Packet Type = DATA
|| | +--------- Destination Type = SINGLE
|| +----------- Propagation Type = BROADCAST
|+------------- Header Type = HEADER_1 (two byte header, one address field)
+-------------- Access Codes = ENABLED
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
Size examples of different packet types
---------------------------------------
The following table lists example sizes of various
packet types. The size listed are the complete on-
wire size counting all fields including headers,
but excluding any interface access codes.
- Path Request : 33 bytes
- Announce : 151 bytes
- Link Request : 77 bytes
- Link Proof : 77 bytes
- Link RTT packet : 83 bytes
- Link keepalive : 14 bytes
+142 -22
View File
@@ -6,16 +6,18 @@ Using Reticulum on Your System
Reticulum is not installed as a driver or kernel module, as one might expect
of a networking stack. Instead, Reticulum is distributed as a Python module.
This means that no special privileges are required to install or use it.
This means that no special privileges are required to install or use it. It
is also very light-weight, and easy to transfer to and install on new systems.
Any program or application that uses Reticulum will automatically load and
initialise Reticulum when it starts.
In many cases, this approach is sufficient. When any program needs to use
Reticulum, it is loaded, initialised, interfaces are brought up, and the
program can now communicate over Reticulum. If another program starts up
and also wants access to the same Reticulum network, the instance is simply
shared. This works for any number of programs running concurrently, and is
very easy to use, but depending on your use case, there are other options.
program can now communicate over any Reticulum networks available. If another
program starts up and also wants access to the same Reticulum network, the
instance is simply shared. This works for any number of programs running
concurrently, and is very easy to use, but depending on your use case, there
are other options.
Included Utility Programs
-------------------------
@@ -57,6 +59,7 @@ the same system.
-q, --quiet
--version show program's version number and exit
You can easily add ``rnsd`` as an always-on service by :ref:`configuring a service<using-systemd>`.
The rnstatus Utility
====================
@@ -71,33 +74,49 @@ interfaces, similar to the ``ifconfig`` program.
# Example output
Shared Instance[37428]
Status: Up
Connected applications: 1
RX: 1.13 KB
TX: 1.07 KB
Status : Up
Serving : 1 program
Rate : 1.00 Gbps
Traffic : 83.13 KB
86.10 KB↓
UDPInterface[Default UDP Interface/0.0.0.0:4242]
Status: Up
RX: 1.01 KB
TX: 1.01 KB
AutoInterface[Local]
Status : Up
Mode : Full
Rate : 10.00 Mbps
Peers : 1 reachable
Traffic : 63.23 KB↑
80.17 KB↓
TCPInterface[RNS Testnet Frankfurt/frankfurt.rns.unsigned.io:4965]
Status: Up
RX: 1.37 KB
TX: 9.02 KB
Status : Up
Mode : Full
Rate : 10.00 Mbps
Traffic : 187.27 KB↑
74.17 KB↓
RNodeInterface[RNode UHF]
Status : Up
Mode : Access Point
Rate : 1.30 kbps
Access : 64-bit IFAC by <…e702c42ba8>
Traffic : 8.49 KB↑
9.23 KB↓
Reticulum Transport Instance <5245a8efe1788c6a70e1> running
.. code:: text
usage: rnsd [-h] [--config CONFIG] [-v] [-q] [--version]
usage: rnstatus [-h] [--config CONFIG] [--version] [-a] [-v]
Reticulum Network Stack Daemon
Reticulum Network Stack Status
optional arguments:
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
-v, --verbose
-q, --quiet
--version show program's version number and exit
-a, --all show all interfaces
-v, --verbose
The rnpath Utility
@@ -116,7 +135,8 @@ destinations on the Reticulum network.
.. code:: text
usage: rnpath.py [-h] [--config CONFIG] [--version] [-v] [destination]
usage: rnpath [-h] [--config CONFIG] [--version] [-t] [-d] [-w seconds] [-v]
[destination]
Reticulum Path Discovery Utility
@@ -127,6 +147,9 @@ destinations on the Reticulum network.
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
--version show program's version number and exit
-t, --table show all known paths
-d, --drop remove the path to a destination
-w seconds timeout before giving up
-v, --verbose
@@ -162,4 +185,101 @@ destinations will not have this option enabled, and will not be probable.
-h, --help show this help message and exit
--config CONFIG path to alternative Reticulum config directory
--version show program's version number and exit
-v, --verbose
-v, --verbose
Improving System Configuration
------------------------------
If you are setting up a system for permanent use with Reticulum, there is a
few system configuration changes that can make this easier to administrate.
These changes will be detailed here.
Fixed Serial Port Names
=======================
On a Reticulum instance with several serial port based interfaces, it can be
beneficial to use the fixed device names for the serial ports, instead
of the dynamically allocated shorthands such as ``/dev/ttyUSB0``. Under most
Debian-based distributions, including Ubuntu and Raspberry Pi OS, these nodes
can be found under ``/dev/serial/by-id``.
You can use such a device path directly in place of the numbered shorthands.
Here is an example of a packet radio TNC configured as such:
.. code:: text
[[Packet Radio KISS Interface]]
type = KISSInterface
interface_enabled = True
outgoing = true
port = /dev/serial/by-id/usb-FTDI_FT230X_Basic_UART_43891CKM-if00-port0
speed = 115200
databits = 8
parity = none
stopbits = 1
preamble = 150
txtail = 10
persistence = 200
slottime = 20
Using this methodology avoids potential naming mix-ups where physical devices
might be plugged and unplugged in different orders, or when device name
assignment varies from one boot to another.
.. _using-systemd:
Reticulum as a System Service
=============================
Instead of starting Reticulum manually, you can install ``rnsd`` as a system
service and have it start automatically at boot.
If you installed Reticulum with ``pip``, the ``rnsd`` program will most likely
be located in a user-local installation path only, which means ``systemd`` will not
be able to execute it. In this case, you can simply symlink the ``rnsd`` program
into a directory that is in systemd's path:
.. code:: text
sudo ln -s $(which rnsd) /usr/local/bin/
You can then create the service file ``/etc/systemd/system/rnsd.service`` with the
following content:
.. code:: text
[Unit]
Description=Reticulum Network Stack Daemon
After=multi-user.target
[Service]
# If you run Reticulum on WiFi devices,
# or other devices that need some extra
# time to initialise, you might want to
# add a short delay before Reticulum is
# started by systemd:
# ExecStartPre=/bin/sleep 10
Type=simple
Restart=always
RestartSec=3
User=USERNAMEHERE
ExecStart=rnsd --service
[Install]
WantedBy=multi-user.target
Be sure to replace ``USERNAMEHERE`` with the user you want to run ``rnsd`` as.
To manually start ``rnsd`` run:
.. code:: text
sudo systemctl start rnsd
If you want to automatically start ``rnsd`` at boot, run:
.. code:: text
sudo systemctl enable rnsd
+45 -21
View File
@@ -2,9 +2,9 @@
What is Reticulum?
******************
Reticulum is a cryptography-based networking stack for wide-area networks built on readily available hardware, and can operate even with very high latency and extremely low bandwidth.
Reticulum is a cryptography-based networking stack for building wide-area networks with readily available hardware, that can continue to operate even with extremely low bandwidth and very high latency.
Reticulum allows you to build very wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.
Reticulum allows you to build wide-area networks with off-the-shelf tools, and offers end-to-end encryption, autoconfiguring cryptographically backed multi-hop transport, efficient addressing, unforgeable packet acknowledgements and more.
Reticulum is a complete networking stack, and does not need IP or higher layers, although it is easy to utilise IP (with TCP or UDP) as the underlying carrier for Reticulum. It is therefore trivial to tunnel Reticulum over the Internet or private IP networks. Reticulum is built directly on cryptographic principles, allowing resilience and stable functionality in open and trustless networks.
@@ -13,12 +13,7 @@ No kernel modules or drivers are required. Reticulum runs completely in userland
Current Status
==============
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered relatively stable at the moment, but could change if warranted.
Caveat Emptor
==============
Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.
Reticulum should currently be considered beta software. All core protocol features are implemented and functioning, but additions will probably occur as real-world use is explored. There will be bugs. The API and wire-format can be considered stable at the moment, but could change if absolutely warranted.
What does Reticulum Offer?
@@ -27,9 +22,11 @@ What does Reticulum Offer?
* Fully self-configuring multi-hop routing
* Asymmetric X25519 encryption and Ed25519 signatures as a basis for all communication
* Complete initiator anonymity, communicate without revealing your identity
* Forward Secrecy with ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
* Asymmetric encryption based on X25519, and Ed25519 signatures as a basis for all communication
* Forward Secrecy by using ephemereal Elliptic Curve Diffie-Hellman keys on Curve25519
* Reticulum uses the `Fernet <https://github.com/fernet/spec/blob/master/Spec.md>`_ specification for on-the-wire / over-the-air encryption
@@ -47,6 +44,12 @@ What does Reticulum Offer?
* An intuitive and developer-friendly API
* Efficient link establishment
* Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes
* Low cost of keeping links open at only 0.62 bits per second
* Reliable and efficient transfer of arbritrary amounts of data
* Reticulum can handle a few bytes of data or files of many gigabytes
@@ -55,25 +58,23 @@ What does Reticulum Offer?
* The API is very easy to use, and provides transfer progress
* Efficient link establishment
* Authentication and virtual network segmentation on all supported interface types
* Total bandwidth cost of setting up a link is only 3 packets, totalling 237 bytes
* Low cost of keeping links open at only 0.62 bits per second
* Flexible scalability allowing extremely low-bandwidth networks to co-exist and interoperate with large, high-bandwidth networks
Where can Reticulum be Used?
============================
Over practically any medium that can support at least a half-duplex channel
with 1.000 bits per second throughput, and an MTU of 500 bytes. Data radios,
with 500 bits per second throughput, and an MTU of 500 bytes. Data radios,
modems, LoRa radios, serial lines, AX.25 TNCs, amateur radio digital modes,
ad-hoc WiFi, free-space optical links and similar systems are all examples
of the types of interfaces Reticulum was designed for.
An open-source LoRa-based interface called `RNode <https://unsigned.io/rnode>`_
has been designed specifically for use with Reticulum. It is possible to build
yourself, or it can be purchased as a complete transceiver that just needs a
USB connection to the host.
has been designed as an example transceiver that is very suitable for
Reticulum. It is possible to build it yourself, to transform a common LoRa
development board into one, or it can be purchased as a complete transceiver.
Reticulum can also be encapsulated over existing IP networks, so there's
nothing stopping you from using it over wired ethernet or your local WiFi
@@ -83,24 +84,47 @@ self-configuring, resilient and encrypted mesh.
As an example, it's possible to set up a Raspberry Pi connected to both a
LoRa radio, a packet radio TNC and a WiFi network. Once the interfaces are
configured, Reticulum will take care of the rest, and any device on the WiFi
added, Reticulum will take care of the rest, and any device on the WiFi
network can communicate with nodes on the LoRa and packet radio sides of the
network, and vice versa.
Interface Types and Devices
===========================
Reticulum implements a range of generalised interface types that covers most of the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. Currently, the following interfaces are supported:
Reticulum implements a range of generalised interface types that covers the communications hardware that Reticulum can run over. If your hardware is not supported, it's relatively simple to implement an interface class. Currently, Reticulum can use the following devices and communication mediums:
* Any ethernet device
* WiFi devices
* Wired ethernet devices
* Fibre-optic transceivers
* Data radios with ethernet ports
* LoRa using `RNode <https://unsigned.io/rnode>`_
* Can be installed on `many popular LoRa boards <https://github.com/markqvist/rnodeconfigutil#supported-devices>`_
* Can be purchased as a `ready to use transceiver <https://unsigned.io/rnode>`_
* Packet Radio TNCs, such as `OpenModem <https://unsigned.io/openmodem>`_
* Any packet radio TNC in KISS mode
* Ideal for VHF and UHF radio
* Any device with a serial port
* The I2P network
* TCP over IP networks
* UDP over IP networks
For a full list and more details, see the :ref:`Supported Interfaces<interfaces-main>` chapter.
For a full list and more details, see the :ref:`Supported Interfaces<interfaces-main>` chapter.
Caveat Emptor
==============
Reticulum is an experimental networking stack, and should be considered as such. While it has been built with cryptography best-practices very foremost in mind, it has not been externally security audited, and there could very well be privacy-breaking bugs. To be considered secure, Reticulum needs a thourough security review by independt cryptographers and security researchers. If you want to help out, or help sponsor an audit, please do get in touch.
+1 -1
View File
@@ -29,6 +29,6 @@ setuptools.setup(
]
},
install_requires=['cryptography>=3.4.7', 'pyserial', 'netifaces>=0.10.4'],
install_requires=['cryptography>=3.4.7', 'pyserial>=3.5', 'netifaces'],
python_requires='>=3.6',
)