Commit Graph

2092 Commits

Author SHA1 Message Date
Roger Dingledine
f3ea883ccd bugfix: stop trying to print a null pointer if an OR conn fails
because we didn't like its cert.


svn:r4111
2005-04-25 15:43:37 +00:00
Roger Dingledine
12ea3a26b2 bugfix: make a debug log stop lying
svn:r4110
2005-04-25 15:43:05 +00:00
Roger Dingledine
92b970d2a6 only complain about wedged cpuworkers after an hour, not 100 secs
svn:r4100
2005-04-23 20:58:44 +00:00
Roger Dingledine
6128cdead0 let NumCpus be singular without whining
svn:r4097
2005-04-23 19:29:09 +00:00
Nick Mathewson
5827e2e216 Fix "JAP-client" hideous ASN1 bug, twice. (Fix1: check more thoroughly for TLS errors when handling certs. Fix2: stop assert(0)ing on uncaught TLS errors.)
svn:r4085
2005-04-23 14:26:02 +00:00
Roger Dingledine
ad4eca60ec add geoff's NoPublish patch
svn:r4084
2005-04-21 10:40:48 +00:00
Nick Mathewson
ce9995a579 Replace an infinite-stack-recursion bug with a generate-infinite-debug-messages bug. Maybe it isnt such a good idea to send LOG_DEBUG messages to the controller after all?
svn:r4082
2005-04-17 22:52:02 +00:00
Nick Mathewson
cf84759b4b zero-out errno before calling event_dispatch, just in case we have misdiagnosed poll(2)/EINVAL bug.
svn:r4081
2005-04-17 22:38:39 +00:00
Roger Dingledine
769541ff7e be willing to load balance over up to 2mB, not 1mB, of advertised capacity
svn:r4080
2005-04-12 22:09:47 +00:00
Roger Dingledine
bf79b5266f those should be notices, not warns
svn:r4079
2005-04-12 22:09:04 +00:00
Nick Mathewson
f1dfa30064 Fix postdescriptor behavior: on non-complete success, only say "done" once.
svn:r4073
2005-04-11 02:52:09 +00:00
Roger Dingledine
e2cdc87485 remove some redundant code
svn:r4062
2005-04-08 06:16:09 +00:00
Nick Mathewson
656e31ba33 If it has not been 100 seconds since we told the cpuworker about an onion, complain! Oh, wait...
svn:r4059
2005-04-08 04:59:34 +00:00
Nick Mathewson
74dc7eedc5 Add kludgy function to force controllers to flush EVENT_ERR_MSG events.
svn:r4057
2005-04-08 03:36:39 +00:00
Nick Mathewson
31c12e265f Move a quote, fix a log.
svn:r4056
2005-04-08 03:26:44 +00:00
Nick Mathewson
c6769a80fb Fixs a signed/unsigned comparison
svn:r4055
2005-04-08 03:26:23 +00:00
Nick Mathewson
6face4f482 Add missing "char"
svn:r4054
2005-04-08 03:07:07 +00:00
Roger Dingledine
97637549bb tie MAX_DIR_SIZE to MAX_BUF_SIZE
resolve bug 80


svn:r4052
2005-04-07 22:13:17 +00:00
Roger Dingledine
dabafddb82 fix format bug
svn:r4051
2005-04-07 22:03:56 +00:00
Roger Dingledine
ec3650f00f only whine about clock skew from the dirserver if he's a trusted
dirserver.


svn:r4048
2005-04-07 21:09:19 +00:00
Nick Mathewson
8cc3307e0d Improve conn_*_to_string; add circuit_state_to_string; make skewed-descriptor messages better.
svn:r4047
2005-04-07 21:07:19 +00:00
Roger Dingledine
dd98412ce1 allow servers to publish descriptors from 12 hours in the future.
hopefully this will allow more slack and let crummier servers help
out too.


svn:r4046
2005-04-07 21:01:00 +00:00
Nick Mathewson
b897cb1ec3 Change conn_(type|state)_to_string from const arrays to switch functions so we (I) will not make Tor crash the next time we (I) add a new state.
svn:r4045
2005-04-07 20:25:22 +00:00
Roger Dingledine
6bdfd97272 add function comment
svn:r4043
2005-04-07 20:07:34 +00:00
Roger Dingledine
ab8a0bb52c We have a bug that I can't find. Sometimes, very rarely, cpuworkers get
stuck in the 'busy' state, even though the cpuworker process thinks of
itself as idle. I don't know why. But here's a workaround to kill any
cpuworker that's been busy for more than 100 seconds.


svn:r4042
2005-04-07 20:02:00 +00:00
Nick Mathewson
c92a2620d4 Try RB_TREE instead of SPLAY_TREE, but with a single-entry caching optimization.
svn:r4041
2005-04-07 05:09:19 +00:00
Nick Mathewson
a035b8e3c3 Make compare_orconn_circid_entries marginally faster and inlined, if that helps
svn:r4040
2005-04-07 04:50:25 +00:00
Nick Mathewson
7709547c08 When we connect and finish TLS negotiation with address:port, it is obvious that any other key or nickname we might know about (as an authdirserver) does not actually exist at address:port.
svn:r4036
2005-04-06 21:09:47 +00:00
Roger Dingledine
03720b39fd stop wasting time doing a case insensitive comparison for every dns name
every time we do any lookup. canonicalize the names to lowercase and be
done with it.


svn:r4035
2005-04-06 20:25:21 +00:00
Roger Dingledine
46089c49c4 blow away most calls to assert_all_pending_dns_resolves_ok, since
they're eating our cpu.


svn:r4034
2005-04-06 20:05:34 +00:00
Roger Dingledine
652dde72f1 bugfix: chdir to your datadirectory at the *end* of the
daemonize process, not the beginning. this was a problem
because the first time you run tor, if your datadir isn't
there, and you have runasdaemon set to 1, it will try to
chdir to it before it tries to create it. oops.


svn:r4033
2005-04-06 19:07:38 +00:00
Roger Dingledine
8596947ee2 quiet another instance where we whine to an innocent OR operator's
logs when some other server switches keys unexpectedly.


svn:r4032
2005-04-06 18:55:14 +00:00
Roger Dingledine
620419b421 oh, and accept() into the right place, too.
svn:r4029
2005-04-06 16:16:31 +00:00
Roger Dingledine
70f3b3ef86 Forward-port the checking of
sin_addr.s_addr == 0 || sin->sin_port == 0.
This just happened on moria2, so I guess it happens rarely
on Linux as well as OS X.
We can't afford to accept OR conns from 0.0.0.0:0, since we
send created cells back to the first addr:port that matches,
and we'd better not send them to the wrong place.
So, let's drop them all for now, and see if we can find a pattern
later.


svn:r4028
2005-04-06 15:42:35 +00:00
Roger Dingledine
9cbaf4603d fix an assert trigger: when we have the rare case of accepting
a conn on 0.0.0.0:0, then when we look through the connection array,
we'll find any of the workers. this is no good.


svn:r4027
2005-04-06 15:19:32 +00:00
Nick Mathewson
88dc243bb5 Rename circuit_get_by_stream to circuit_get_by_edge_conn, and actually start using it. Watch out, kids!
svn:r4026
2005-04-06 06:43:21 +00:00
Roger Dingledine
716d9bc99a when using preferred entry or exit nodes, ignore whether
the circuit wants uptime or capacity. they asked for the
nodes, they get the nodes.


svn:r4025
2005-04-06 06:17:35 +00:00
Nick Mathewson
4453e3794a Add a pointer from edge connections to their corresponding circuit (ulp!); add some debugging sanity-checking for cirid_orconn_map stuff
svn:r4024
2005-04-06 06:13:49 +00:00
Nick Mathewson
2a0b1025e5 Fix bug in last commit when no circid can be found
svn:r4023
2005-04-06 05:45:07 +00:00
Nick Mathewson
b34d9adb59 Fix format warning
svn:r4022
2005-04-06 05:44:05 +00:00
Roger Dingledine
94cecc712d note some features we intend to add.
svn:r4021
2005-04-06 05:35:06 +00:00
Nick Mathewson
b7cdcf3462 Hopefully, this will make ORs much faster, and not break them: keep a big splay tree of (circid,orconn)->circuit mappings to make circuit_get_by_circid_conn much faster.
svn:r4020
2005-04-06 05:33:32 +00:00
Roger Dingledine
712d05c19a a bit more data on gene's dns threading bug
svn:r4019
2005-04-06 05:32:24 +00:00
Roger Dingledine
5f0a84cbe2 update conn-type-state-strings to handle https states for ORs
svn:r4018
2005-04-06 04:55:17 +00:00
Nick Mathewson
3014b03b67 Add missing comma. Oops
svn:r4017
2005-04-06 03:20:06 +00:00
Nick Mathewson
88fe4d6292 Handle changed router status correctly when reloading fingerprint file
svn:r4016
2005-04-06 01:09:26 +00:00
Nick Mathewson
1b5e2182fa Log msg spelling fix
svn:r4015
2005-04-06 00:53:13 +00:00
Roger Dingledine
f860b490be When we're connecting to an OR and he's got a different nickname/key
than we were expecting, only complain loudly if we're an OP or an
authdirserver. Complaining loudly to OR admins just confuses people.


svn:r4014
2005-04-06 00:50:40 +00:00
Nick Mathewson
ac268a8340 Fix Tor for no-threading compilation: move connection_unregister from _connection_free to connection_free.
svn:r4013
2005-04-06 00:46:57 +00:00
Nick Mathewson
374edd356e Change replacement semantics for directory entries: fix bug 121.
svn:r4012
2005-04-06 00:01:35 +00:00