Commit Graph

12853 Commits

Author SHA1 Message Date
Robert Ransom b3133d1cad Exit immediately if we can't monitor our owning controller process
tor_process_monitor_new can't currently return NULL, but if it ever can,
we want that to be an explicitly fatal error, without relying on the fact
that monitor_owning_controller_process's chain of caller will exit if it
fails.
2011-05-20 08:25:42 -07:00
Robert Ransom 0caa37db4d Fix some comments 2011-05-20 08:25:42 -07:00
Robert Ransom 4b266c6e72 Implement __OwningControllerProcess option
Implements part of feature 3049.
2011-05-20 08:25:42 -07:00
Robert Ransom c714a098ea Improve a documentation comment 2011-05-12 02:57:09 -07:00
Robert Ransom cb9df5e53c Fix comment typo 2011-05-12 00:27:19 -07:00
Robert Ransom b7452dcbcb Fix comment typo 2011-05-10 05:15:02 -07:00
Nick Mathewson 09d7af7789 Merge remote-tracking branch 'public/bug3022_v2' into maint-0.2.2 2011-05-09 13:37:56 -04:00
Nick Mathewson 0b3166fffa Merge branch 'clang_fixes' into maint-0.2.2 2011-05-09 13:20:04 -04:00
Sebastian Hahn 9da4e25183 Remove some dead code, found by clang 2011-05-09 13:19:46 -04:00
Sebastian Hahn d7d25558fa Remove a duplicated line, found by clang 2011-05-09 13:19:46 -04:00
Sebastian Hahn 1c668540fe Fix potential null pointer deref during dirvote
Found by using clang's analyzer.
2011-05-09 13:19:46 -04:00
Sebastian Hahn 1827e60976 Fix a potential null deref when rebuilding md cache
Issue discovered using clang's static analyzer
2011-05-09 13:19:45 -04:00
Sebastian Hahn 8ebb3ce6e2 CONN_LOG_PROTECT()'s first argument may not be 0
Make that explicit by adding an assert and removing a null-check. All of
its callers currently depend on the argument being non-null anyway.
Silences a few clang complaints.
2011-05-09 13:19:45 -04:00
Sebastian Hahn 80e57af50f Appease clang - and my tortured mind
This possible div by 0 warning from clang's analyzer was quite fun to
track down. Turns out the current behaviour is safe.
2011-05-09 13:19:45 -04:00
Sebastian Hahn 58a16a4d6f Add an assert to un-confuse clang's analyzer
The analyzer assumed that bootstrap_percent could be less than 0 when we
call control_event_bootstrap_problem(), which would mean we're calling
log_fn() with undefined values. The assert makes it clear this can't
happen.
2011-05-09 13:19:44 -04:00
Sebastian Hahn 532c13693e Fix a docstring 2011-05-09 13:19:44 -04:00
Nick Mathewson 330116f034 Fix up some check-spaces issues 2011-05-05 21:53:46 -04:00
Andrew Lewman be7e8f5bb9 fix the website directions. 2011-05-05 15:28:29 -04:00
Nick Mathewson 9583a534ac Merge remote-tracking branch 'rransom/bug3106' into maint-0.2.2 2011-05-04 23:18:47 -04:00
Robert Ransom 66339f74b4 Handle crypto_pk_get_digest failures semi-sensibly
Fixes bug 3106.
2011-05-04 20:06:24 -07:00
Nick Mathewson ab1460f3ae Merge remote-tracking branch 'sebastian/win_warning' into maint-0.2.2 2011-05-04 20:33:16 -04:00
Sebastian Hahn 9a490bb53b Fix compile warning on windows 2011-05-05 02:22:46 +02:00
Nick Mathewson 0ee524b57f Merge branch 'bug2379' into maint-0.2.2 2011-05-04 17:37:56 -04:00
Nick Mathewson 29f1ffa0fe More notes on how to upload tarballs from erinn 2011-05-04 17:21:33 -04:00
Nick Mathewson b04dca448d Add some missing torrc entries to tor.1.txt
Fixes bug 2379
2011-05-03 22:14:40 -04:00
Nick Mathewson 4126de6888 Fix circuit_list_path_impl(): internal circuits do not have an "exit". Trivial fix for 3079. 2011-05-03 21:53:59 -04:00
Nick Mathewson 68ae5afa5a Change who calls microdesc_cache_rebuild().
Previously we ensured that it would get called periodically by doing
it from inside the code that added microdescriptors.  That won't work
though: it would interfere with our code that tried to read microdescs
from disk initially.  Instead, we should consider rebuilding the cache
periodically, and on startup.
2011-05-03 17:28:28 -04:00
Nick Mathewson cb6c909664 Rebuild the microdesc cache when a sufficient number of bytes are dropped
Previously on 0.2.2, we'd never clean the cache.  Now that we can
clean it, we want to add a condition to rebuild it: that should happen
whenever we have dropped enough microdescriptors that we could save a
lot of space.

No changes file, since 0.2.3 doesn't need one and 0.2.2 already has some
changes files for the backport of the microdesc_clean_cahce() function.
2011-05-03 17:03:49 -04:00
Nick Mathewson 56fbd728c2 Backport microdesc_cache_clean to 0.2.2
Otherwise we have no way to keep authorities' microdesc caches in 0.2.2
from growing without bound.
2011-05-03 16:45:15 -04:00
Nick Mathewson 970715dd8f Fix a check for when to rebuild the microdesc cache. (Backport from 0.2.3. 2011-05-03 16:29:39 -04:00
Nick Mathewson 698fa0fc67 Add missing code to set cache->journal_len when reading microdesc journal
This could be one reason that authorities' journals would grow without
bound; related to bug 2230. Bugfix on 0.2.2.6-alpha.  Fix by
"cypherpunks".
2011-05-03 16:22:31 -04:00
Nick Mathewson 033c27ac5e Add a couple of notes to doc/HACKING based on 0.2.2.25-alpha process 2011-05-02 16:55:50 -04:00
Nick Mathewson dbd73b9689 Clean up a warning a bit 2011-04-29 11:14:53 -04:00
Roger Dingledine 93b0183d3c add another heuristic for changes stanzas 2011-04-29 01:35:21 -04:00
Nick Mathewson 676190e895 Update hacking file with terse notes on formatting changelog 2011-04-28 23:44:48 -04:00
Nick Mathewson f0d9e2d650 Merge remote-tracking branch 'arma/bug3012' into maint-0.2.2 2011-04-28 21:15:14 -04:00
Roger Dingledine 66de6f7eb8 relays checkpoint their state file twice a day 2011-04-28 21:06:25 -04:00
Nick Mathewson cd42ae7185 Only authorities should automatically download v2 networkstatus documents
Clients and relays haven't used them since early 0.2.0.x.  The only
remaining use by authorities learning about new relays ahead of scedule;
see proposal 147 for what we intend to do about that.

We're leaving in an option (FetchV2Networkstatus) to manually fetch v2
networkstatuses, because apparently dnsel and maybe bwauth want them.

This fixes bug 3022.
2011-04-28 21:06:07 -04:00
Nick Mathewson 4010427b51 Merge remote-tracking branch 'arma/bug3039' into maint-0.2.2 2011-04-28 20:53:44 -04:00
Roger Dingledine df3cf881d1 stop putting wacky values into state->lastwritten 2011-04-28 20:40:15 -04:00
Nick Mathewson 5693fedb60 Clarify comment to say which version fixed 2722 2011-04-28 20:38:15 -04:00
Nick Mathewson 6b9aadd557 Merge remote-tracking branches 'rransom/bug2722' and 'rransom/bug2722b' into maint-0.2.2 2011-04-28 20:36:38 -04:00
Nick Mathewson c122897925 Merge remote-tracking branch 'sebastian/bug3020' into maint-0.2.2 2011-04-28 20:03:57 -04:00
Roger Dingledine 710227a77f fix a function comment 2011-04-28 19:19:04 -04:00
Sebastian Hahn 525d2700dd Correctly check elapsed time in last hibernation period
Fix bug 3020.
2011-04-29 01:18:32 +02:00
Nick Mathewson 30003e4318 Merge branch 'bug3000-022-part2-squashed' into maint-0.2.2 2011-04-28 18:13:17 -04:00
Nick Mathewson 6dfc0d5301 Avoid false positives from proxy_mode()
Previously it would erroneously return true if ListenAddr was set for
a client port, even if that port itself was 0.  This would give false
positives, which were not previously harmful... but which were about
to become.
2011-04-28 18:11:58 -04:00
Robert Ransom b8708b5bd3 Fix bug 1930 2011-04-28 18:11:58 -04:00
Robert Ransom ddd1b7be2d Ignore SIGNAL NEWNYM on relay-only Tor instances 2011-04-28 18:10:17 -04:00
Robert Ransom df5c7fedbd Don't allow v0 HS auths to act as clients
A v0 HS authority stores v0 HS descriptors in the same descriptor
cache that its HS client functionality uses.  Thus, if the HS
authority operator clears its client HS descriptor cache, ALL v0
HS descriptors will be lost.  That would be bad.
2011-04-28 18:10:16 -04:00