Commit Graph

980 Commits

Author SHA1 Message Date
Nick Mathewson a0ae80788c Replace 4 more sscanf()s with tor_sscanf()
For some inexplicable reason, Coverity departs from its usual
standards of avoiding false positives here, and warns about all
sscanf usage, even when the formatting strings are totally safe.

Addresses CID # 447, 446.
2011-07-01 11:26:30 -04:00
Robert Ransom ca6efcf507 Make handle_control_setevents table-driven 2011-06-25 17:45:33 -07:00
Robert Ransom c780bc4d0b Merge branch 'bug3465-022' into bug3465-023
* bug3465-022:
  Add BUILDTIMEOUT_SET to the result of GETINFO events/names
  Correct a comment
  Fix minor comment issues
2011-06-25 15:04:07 -07:00
Robert Ransom 93d52f6739 Add BUILDTIMEOUT_SET to the result of GETINFO events/names 2011-06-25 15:02:11 -07:00
Nick Mathewson 64a5a4177f Merge branch 'bug3367' 2011-06-21 11:32:52 -04:00
Nick Mathewson 4186b1362b Merge remote-tracking branch 'public/bug3264' 2011-06-21 10:28:25 -04:00
Nick Mathewson e617a34d58 Add, use a bufferevent-safe connection_flush()
A couple of places in control.c were using connection_handle_write()
to flush important stuff (the response to a SIGNAL command, an
ERR-level status event) before Tor went down.  But
connection_handle_write() isn't meaningful for bufferevents, so we'd
crash.

This patch adds a new connection_flush() that works for all connection
backends, and makes control.c use that instead.

Fix for bug 3367; bugfix on 0.2.3.1-alpha.
2011-06-21 10:22:54 -04:00
Jacob Appelbaum 8b0d9452ca Fix tor-fw-helper-natpmp.c API usage
libnatpmp-20110618 changed the API that tor-fw-helper used and for a time
tor-fw-helper could not build against the newest libnatpmp. This patch brings
support for libnatpmp to tor-fw-helper.
2011-06-20 20:01:46 -04:00
Nick Mathewson 410e440a8d Log SSL state changes at LOG_DEBUG, LD_HANDSHAKE.
This can be slightly useful for debugging blocking events.

Addresses ticket 3116; based on loud_ssl_states branch.
2011-06-20 17:45:12 -04:00
intrigeri 8b265543eb Add port 6523 (Gobby) to LongLivedPorts. 2011-06-20 12:08:46 -04:00
Nick Mathewson a046966baf Merge remote-tracking branch 'origin/maint-0.2.2'
Conflicts:
	src/or/config.c
	src/or/or.h
2011-06-19 21:00:15 -04:00
Nick Mathewson 8e5dfe42a0 Tweak bug2355_revert at suggestion from Roger 2011-06-19 20:58:46 -04:00
Robert Ransom eeedc5c13c Put hashes of all tor source files into or_sha1.i 2011-06-18 00:54:04 -07:00
Nick Mathewson 1811bef666 Add changes file for bug2355 revert 2011-06-17 16:54:31 -04:00
Nick Mathewson 42d6f34590 Merge remote-tracking branch 'origin/maint-0.2.2' 2011-06-17 15:25:18 -04:00
Nick Mathewson 85d4c290d7 Merge remote-tracking branch 'origin/maint-0.2.1' into maint-0.2.2 2011-06-17 15:24:23 -04:00
Robert Ransom 010b8dd4f6 Abandon rendezvous circuits on SIGNAL NEWNYM 2011-06-17 15:22:23 -04:00
Nick Mathewson 31b9b1a5bb Merge remote-tracking branch 'origin/maint-0.2.2' 2011-06-17 15:04:29 -04:00
Nick Mathewson 209229f100 Merge branch 'bug3407' into maint-0.2.2 2011-06-17 14:59:39 -04:00
Robert Ransom 227896e447 Make send_control_event_impl's behaviour sane 2011-06-17 13:57:25 -04:00
Robert Ransom 68cd0a9abe Make connection_printf_to_buf's behaviour sane 2011-06-17 13:57:25 -04:00
Nick Mathewson a857f61e27 Merge remote-tracking branch 'rransom-tor/bug3332-v2' 2011-06-15 11:33:40 -04:00
Robert Ransom 44eafa9697 Assert that HS operations are not performed using single-hop circuits
(with fixes by Nick Mathewson to unbreak the build)
2011-06-14 21:32:49 -07:00
Nick Mathewson 875a551409 Merge remote-tracking branch 'origin/maint-0.2.2' 2011-06-14 13:46:46 -04:00
Nick Mathewson a25c0a5bf8 Merge remote-tracking branch 'public/bug3369' into maint-0.2.2 2011-06-14 13:46:10 -04:00
Nick Mathewson e70e8367c9 Add changes file for bug3393 2011-06-14 13:45:12 -04:00
Nick Mathewson 19febed9e5 Merge remote-tracking branch 'rransom-tor/bug3349' 2011-06-14 13:39:26 -04:00
Nick Mathewson 8839b86085 Merge remote-tracking branch 'origin/maint-0.2.2' 2011-06-14 12:25:33 -04:00
Nick Mathewson e470f6a886 changelog entry for bug3369 2011-06-13 13:42:00 -04:00
Sebastian Hahn 910dadd6ea Ensure the release tarball contains all manpages
Thanks to Nick for the idea for the fix.
2011-06-12 13:58:53 +02:00
Nick Mathewson ecc9a364c2 fix typo in changes/coverity_master spotted by rransom 2011-06-10 16:40:47 -04:00
Nick Mathewson b2e7c356db Prefer tt_assert in unit tests, not tor_assert 2011-06-08 20:33:53 -04:00
Nick Mathewson 2fd5603057 tweak coverity_master changes file 2011-06-08 20:29:10 -04:00
Sebastian Hahn 8283767365 Fix a memleak in nodelist_assert_ok() 2011-06-08 21:38:42 +02:00
Sebastian Hahn f303274490 Fix a rare memleak during stats writing
If rep_hist_buffer_stats_write() was called unitinitalized, we'd leak
memory.
2011-06-08 21:35:26 +02:00
Sebastian Hahn 680646e0de Don't use signed 1-bit bitfields
This was harmless, we never compared it to anything but itself or 0.
But Coverity complained, and it had a point.
2011-06-08 21:30:41 +02:00
Sebastian Hahn e6fff7235e Remove a few dead assignments during router parsing 2011-06-08 21:30:40 +02:00
Sebastian Hahn 3033f8459b Remove a dead variable in the heartbeat code 2011-06-08 21:11:42 +02:00
Sebastian Hahn ff75e8b02d Check some more return values in unit tests 2011-06-08 21:08:56 +02:00
Sebastian Hahn db7dd3ee7a remove some dead code, found by coverity 2011-06-08 21:07:54 +02:00
Sebastian Hahn 3bf808c341 Fix a minor coverity-found bug in tests 2011-06-08 20:00:47 +02:00
Nick Mathewson d696bd1bfc Fix crash bug (3361) when setting off-flavor networkstatus
When we set a networkstatus in the non-preferred flavor, we'd check
the time in the current_consensus.  But that might have been NULL,
which could produce a crash as seen in bug 3361.
2011-06-07 14:46:28 -04:00
Nick Mathewson 8cd5a3c186 Merge remote-tracking branch 'origin/maint-0.2.2' 2011-06-06 16:20:22 -04:00
Nick Mathewson 9e3fc5f616 Merge branch 'bug3306_nm_squashed' into maint-0.2.2 2011-06-06 16:18:24 -04:00
Nick Mathewson 42e4e156d9 Detect insanely large circuit build state; don't give its length to rand_int 2011-06-06 16:18:06 -04:00
Nick Mathewson 5afab5ca19 Check maximum properly in crypto_rand_int()
George Kadianakis notes that if you give crypto_rand_int() a value
above INT_MAX, it can return a negative number, which is not what
the documentation would imply.

The simple solution is to assert that the input is in [1,INT_MAX+1].
If in the future we need a random-value function that can return
values up to UINT_MAX, we can add one.

Fixes bug 3306; bugfix on 0.2.2pre14.
2011-06-06 16:18:06 -04:00
Roger Dingledine 96da10aace Merge branch 'maint-0.2.2' 2011-06-05 11:14:28 -04:00
Roger Dingledine 83e3975c8b Merge branch 'maint-0.2.1' into maint-0.2.2 2011-06-05 11:13:50 -04:00
Roger Dingledine bf4b819aae move to the june 1 2011 maxmind geoip db 2011-06-05 11:12:28 -04:00
Robert Ransom 8a69cc0468 Report SIGNAL event to controllers when acting on delayed NEWNYM
Fixes bug 3349.
2011-06-05 05:43:27 -07:00