Adam Ierymenko
|
d166b494ee
|
Rule parse fix.
|
10 years ago |
|
Adam Ierymenko
|
c9d7845fea
|
Minor bug fix and some instrumentation stuff for testing.
|
10 years ago |
|
Adam Ierymenko
|
e1310a764a
|
More cleanup and removal of cruft due to obsolete network-specific relays (will be replaced with federation stuff).
|
10 years ago |
|
Adam Ierymenko
|
dee7f75f7e
|
Minor cleanup.
|
10 years ago |
|
Adam Ierymenko
|
774c7e0ea5
|
Put CONFIG_REFRESH back.
|
10 years ago |
|
Adam Ierymenko
|
4d498b3765
|
Handling of multi-part chunked network configs on the inbound side.
|
10 years ago |
|
Adam Ierymenko
|
bcd05fbdfa
|
Chunking of network config replies.
|
10 years ago |
|
Adam Ierymenko
|
2ba9343607
|
Encode and decode of tags and capabilities in NetworkConfig.
|
10 years ago |
|
Adam Ierymenko
|
00fd9c3a15
|
It builds... almost ready to test some rules engine stuff.
|
10 years ago |
|
Adam Ierymenko
|
4d9b74b171
|
.
|
10 years ago |
|
Adam Ierymenko
|
331382cf2f
|
More cleanup and a tiny federation prep item.
|
10 years ago |
|
Adam Ierymenko
|
98152d974a
|
More cleanup and removal of DeferredPackets, will do the latter in a more elegant way.
|
10 years ago |
|
Adam Ierymenko
|
5cf410490e
|
.
|
10 years ago |
|
Adam Ierymenko
|
404a0bbddd
|
...
|
10 years ago |
|
Adam Ierymenko
|
f057bb63cd
|
More work on tags and capabilities.
|
10 years ago |
|
Adam Ierymenko
|
7e6e56e2bc
|
Bunch of work on pushing and replication of tags and capabilities, and protocol cleanup.
|
10 years ago |
|
Adam Ierymenko
|
4929be08f7
|
Cleanup and stub out new object transfer messages.
|
10 years ago |
|
Adam Ierymenko
|
7404eb46c4
|
Integration of Filter into inbound and outbound packet path.
|
10 years ago |
Moritz Warning
|
b3073f44a2
|
move sourceAddress out of scope
sourceAddress is used in the catch block
|
10 years ago |
|
Adam Ierymenko
|
9785fde32a
|
Trusted paths work!
|
10 years ago |
|
Adam Ierymenko
|
9657675755
|
Plumbing through trusted path stuff to OneService.
|
10 years ago |
|
Adam Ierymenko
|
765082fdb6
|
Trusted path support, and version bump to 1.1.9
|
10 years ago |
|
Adam Ierymenko
|
19735e7050
|
Revert backgrounding of controller requests hack. Controller code is not really parallel anyway and we fixed the perf problem.
|
10 years ago |
|
Adam Ierymenko
|
45f315e603
|
Defer NETWORK_CONFIG_REQUEST packets and allow multithreaded processing.
|
10 years ago |
|
Adam Ierymenko
|
674b84d908
|
Plumbing for network setting control, and GitHub issue #330
|
10 years ago |
|
Adam Ierymenko
|
b2d048aa0e
|
Make Dictionary templatable so it can be used where we want a higher capacity.
|
10 years ago |
|
Adam Ierymenko
|
e09c1a1c11
|
Big refactor mostly builds. We now have a uniform backward compatible netconf.
|
10 years ago |
Grant Limberg
|
de1b745988
|
comment out a few TRACE messages that don't compile
|
10 years ago |
|
Adam Ierymenko
|
548730660b
|
Ready to test whole new netconf refactor.
|
10 years ago |
|
Adam Ierymenko
|
8b9519f0af
|
Simplify a bunch of NetworkConfig stuff by eliminating accessors, also makes network controller easier to refactor.
|
10 years ago |
|
Adam Ierymenko
|
90e1262a8b
|
More refactoring to remove old Dictionary dependencies.
|
10 years ago |
|
Adam Ierymenko
|
368efaa2ba
|
Kill some old debug code.
|
10 years ago |
|
Adam Ierymenko
|
2f18a92e20
|
Cleanup in numerous places, reduce network chattiness around MULTICAST_LIKE, and fix a "how was that working" latent bug causing some control traffic to take the scenic route.
|
10 years ago |
|
Adam Ierymenko
|
4c455876f9
|
Revise peer path weighting to always prioritize cluster-optimal paths.
|
10 years ago |
|
Adam Ierymenko
|
cecfa99b7b
|
(1) cluster members send a flag indicating that a PUSH_DIRECT_PATHS is a cluster redirect, (2) 1.1.5 uses this to avoid a bug (this bug does not exist in 1.1.4)
|
10 years ago |
|
Adam Ierymenko
|
ad1e83d8b9
|
Refactor IncomingPacket for new NetworkConfig
|
10 years ago |
|
Adam Ierymenko
|
fe3a84a422
|
Fix problems with previous commit.
|
10 years ago |
|
Adam Ierymenko
|
7b5c1696eb
|
Fix a CIRCUIT_TEST bug in forwarding of tests along hop paths on private networks. Unfortunately this means full circuit testing for private nets will need an upgrade. :(
|
10 years ago |
|
Adam Ierymenko
|
c7c61b4ac0
|
Fix checking of path address validity.
|
10 years ago |
|
Adam Ierymenko
|
43fff1a87e
|
Deprecate reporting of local clock in circuit tests since a small number of users might have security problems with this.
|
10 years ago |
|
Adam Ierymenko
|
2aa7138373
|
Reduce direct ping delay back to 1m and make SelfAwareness aware of local received-on address to eliminate false symmetric classification.
|
10 years ago |
|
Adam Ierymenko
|
4e4fd51117
|
boring doc stuff
|
10 years ago |
|
Adam Ierymenko
|
83ef98a9dc
|
Add a network-associated user ptr in API.
|
10 years ago |
|
Adam Ierymenko
|
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.
|
10 years ago |
|
Adam Ierymenko
|
1023ef23b7
|
Remove somewhat ugly and costly anti-recursion hack -- we will switch to more explicit methods.
|
10 years ago |
|
Adam Ierymenko
|
a56fbc1929
|
Close another potential anti-recursion loophole.
|
10 years ago |
|
Adam Ierymenko
|
d8143a5e18
|
Implement first pass on rapid dead path detection, and increment version to 1.1.3 (dev)
|
10 years ago |
|
Adam Ierymenko
|
436c1fac1d
|
Selectively move over changes from "edge" to "dev" excluding netcon.
|
10 years ago |
|
Adam Ierymenko
|
2160164e8c
|
(1) Get rid of path sorting and just scan them, since sorting may have been a premature optimization that introduced a regression and path instability in a few edge cases, and (2) do not attempt to contact remote paths received via PUSH_DIRECT_PATH if we already have that path and it is already active (dumb, should have done this originally)
|
10 years ago |
|
Adam Ierymenko
|
82aa3f59d6
|
Fix bug in ECHO handling (OK was invalid!), and use ECHO on newer peers for path confirmation. Also get rid of path confirmation circuit breaker since this causes issues with some peers and should be done more intelligently anyway.
|
10 years ago |
