Commit Graph

1094 Commits

Author SHA1 Message Date
2e5caa335a Old SF root is dead. Now we are just on Alice and Bob. (world update for 1.1.4) 2016-01-13 10:18:41 -08:00
4e4fd51117 boring doc stuff 2016-01-12 14:04:55 -08:00
3883ac08c7 Docs and cleanup. 2016-01-12 13:17:30 -08:00
740eb6ebc4 Simplify Peer locking to eliminate deadlock with new path recursion check code (and also probably improve performance). 2016-01-12 12:12:25 -08:00
d6f0f1a82a Use network user ptr in lookup for Ethernet frame handling to eliminate map lookup. 2016-01-12 11:34:22 -08:00
83ef98a9dc Add a network-associated user ptr in API. 2016-01-12 11:04:35 -08:00
704205c5f7 Dead code removal. 2016-01-12 09:33:14 -08:00
bbcc3304a0 Check shouldUsePathForZeroTierTraffic in legacy beacon responder. 2016-01-11 15:57:58 -08:00
b3e3d4cacc Instead of using binary packet comparison, add a callback to the API to explicitly check whether paths should be used. Check in with this callback (if present) when learning new paths or sending initial packets. 2016-01-11 10:17:44 -08:00
1023ef23b7 Remove somewhat ugly and costly anti-recursion hack -- we will switch to more explicit methods. 2016-01-11 09:06:10 -08:00
a56fbc1929 Close another potential anti-recursion loophole. 2016-01-06 15:35:27 -08:00
47ce52228b Roots should probably not do this since it would likely be a waste of packets. 2016-01-06 12:54:51 -08:00
9aee72099e AntiRecursion cleanup and some other minor things. 2016-01-06 10:59:39 -08:00
05b2c0743f Tighten up dead path detection. Should now auto-detect dead paths in less than 10 seconds at a very small cost in ECHO requests (or HELLOs for older peers). GitHib issue #272 2016-01-06 10:00:03 -08:00
4d94ae77b4 simplify if 2016-01-05 16:48:35 -08:00
d8143a5e18 Implement first pass on rapid dead path detection, and increment version to 1.1.3 (dev) 2016-01-05 16:41:54 -08:00
cba739fd6b more dead code 2016-01-05 14:46:26 -08:00
fb5237d5b6 Outline dead path detection mechanism. 2016-01-05 14:42:56 -08:00
258f95b2cd dead code removal 2016-01-05 14:19:16 -08:00
436c1fac1d Selectively move over changes from "edge" to "dev" excluding netcon. 2015-12-21 16:15:39 -08:00
ceaef19fb7 Fix for GitHub issue #260 -- fix for (non-exploitable) crash in network preferred relay code 2015-11-30 17:20:12 -08:00
e6808c69f8 Increase maximum number of paths per IP scope to 4
Otherwise, local discovered routes are ignored.
Don't know what the best value would be. Taking 4 for now.
2015-11-30 19:48:38 +01:00
ce5b7f0305 The World as of 2015-11-20 2015-11-20 12:49:23 -08:00
9169b6c999 Fix crash on exit in Windows (does not affect other OSes) and revert debugging tweak in Windows version that prevented service from starting after real install. 2015-11-17 09:56:19 -08:00
456bfea58d Real World -- edge now talks to real net. 2015-11-16 19:27:36 -08:00
75f8ee1b9c get the compiler to stop complaining about two string literals (one a macro) next to each other without an intervening space 2015-11-14 18:19:33 -08:00
0d9f33dc4f Fix: (1) Windows stack overflow due to buffer too large in peer deserialize, (2) clean up some other stuff seen during debugging and reduce the sizes of some buffers due to Windows small stack size, (3) remove a redundant try/catch. 2015-11-13 12:14:28 -08:00
ccae473783 Eliminate Windows warnings. 2015-11-12 16:51:33 -08:00
54f68280bd Minor Windows build fixes. Builds on VS2012 again. 2015-11-12 16:48:42 -08:00
8ac411a9e4 Stashing utun work for OSX -- abandoning for now since utun excludes mDNS among other issues. 2015-11-12 10:47:50 -08:00
0940d673db Always advertise to the cluster when we have a peer even if we have also initiated handoff. This might be the cause of the warmup problem -- will test later. At the very least it should not hurt anything due to pick-latest logic and the fact that cluster members with only suboptimal paths do not respond to WANT_PEER. 2015-11-11 14:36:22 -08:00
b94335f115 Increase WHOIS retry delay slightly to allow for global cluster latency (e.g. 300+ms from New York to Johhannesburg) 2015-11-11 11:07:08 -08:00
451b8aa7b2 Query both root and network controller for multicast last resort GATHER. 2015-11-10 17:37:38 -08:00
2854f14966 Some cluster parameter tweaks, and change the test code in tests/http to dramatically reduce the amount of data transfer to/from the test master. Also add results of first successful 50k test. 2015-11-10 11:59:03 -08:00
4328c6c3bc Fix delete oldest logic. 2015-11-10 09:46:14 -08:00
b171c9a0db Replace slow std::list<> with an O(log(N)) data structure for the cluster relaying send queue. 2015-11-10 09:05:01 -08:00
32ec378e3b Announce that we have peers on the cluster when we first see them to improve startup times, and add a result crunching script to tests/http. 2015-11-09 18:01:23 -08:00
2cc50bdb10 Try bringing back TTL escalation -- may help with Docker (IP-MASQ) type NAT 2015-11-09 15:44:13 -08:00
94f4316a0e Fix for possible high CPU usage on multicast queries. 2015-11-09 14:54:05 -08:00
35c4e28f31 Mark geo-redirected paths as suboptimal and do not report that we have a peer if all we have is one of these. Also a few other small fixes. 2015-11-09 14:25:28 -08:00
47424df417 Fix (1) a deadlock bug and (2) a bug that prevented distributed rendezvous messages from being sent. 2015-11-09 13:39:06 -08:00
73e2c6e511 How did that ever work? 2015-11-09 12:24:49 -08:00
4ad2ff2f71 . 2015-11-09 11:08:52 -08:00
b57c855a8c PROXY_SEND fix. 2015-11-09 10:25:20 -08:00
ed954c5ca2 . 2015-11-09 10:09:16 -08:00
21341ab15c Ready to test. 2015-11-09 09:45:43 -08:00
f4d12603e0 Tweak how Switch passes off to Cluster on relay to fit new paradigm. 2015-11-09 09:19:03 -08:00
ff81415ffe Build fixes... 2015-11-09 09:14:26 -08:00
57b71bfff0 Cluster simplification and refactor work in progress... 2015-11-08 13:57:02 -08:00
6bc8c9d8ef Clustering cleanup, still a work in progress. 2015-11-06 16:12:41 -08:00