summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2009-03-09Log cached-at-exit exit policies to try to fix bug 672.tor-0.2.1.13-alphaNick Mathewson
svn:r18827
2009-03-09doxygen tweakRoger Dingledine
svn:r18818
2009-03-09bump to 0.2.1.13-alphaRoger Dingledine
svn:r18814
2009-03-093! 3 not 2.Roger Dingledine
svn:r18813
2009-03-09We were already rejecting relay begin cells with destination portRoger Dingledine
of 0. Now also reject extend cells with destination port or address of 0. Suggested by lark. svn:r18812
2009-03-09declare that i'm going to tag 0.2.1.13-alpha tonightRoger Dingledine
svn:r18811
2009-03-08Build correctly from outside main source tree. Patch from Michael Gold.Nick Mathewson
svn:r18809
2009-03-06Remove contrib/tor.wxs from AC_CONFIG_FILES. The file is gone.Karsten Loesing
svn:r18791
2009-03-06Move wxs files to vidalia trunk for integration with existing bundle ↵Martin Peck
translations. svn:r18789
2009-03-05Update to the "February 26 2009" ip-to-country fileRoger Dingledine
svn:r18774
2009-03-03Actually use tor_sscanf() to parse untrusted input.Nick Mathewson
svn:r18761
2009-03-03Add a simple locale-independent no-surprises sscanf replacement.Nick Mathewson
tor_sscanf() only handles %u and %s for now, which will make it adequate to replace sscanf() for date/time/IP parsing. We want this to prevent attackers from constructing weirdly formed descriptors, cells, addresses, HTTP responses, etc, that validate under some locales but not others. svn:r18760
2009-03-02Add and use set/get_uint64 on onion tags. [bug 604; backportable]Nick Mathewson
It seems that 64-bit Sparc Solaris demands 64-bit-aligned access to uint64_t, but does not 64-bit-align the stack-allocated char array we use for cpuworker tags. So this patch adds a set/get_uint64 pair, and uses them to access the conn_id field in the tag. svn:r18743
2009-02-27minor updates and questions.Andrew Lewman
svn:r18705
2009-02-23Fix one case of bug 929.Nick Mathewson
svn:r18683
2009-02-23Add some debugging code to try to catch the likely cause of bug 929Nick Mathewson
svn:r18682
2009-02-23Updated the description text, changed packager emailAndrew Lewman
svn:r18675
2009-02-21Patch from lark: if we get two extend cells for the same circuit id, drop ↵Nick Mathewson
the second. Previously, we had leaked an extend_info if the target connection was not open when the second arrived. svn:r18668
2009-02-21Patch from lark: drop BEGIN cells from a rendevous circuit if they do not ↵Nick Mathewson
originate from the end of the circuit. svn:r18667
2009-02-20send the newconsensus event if the controller has asked for newconsensusRoger Dingledine
events, not if he's asked for ns events svn:r18656
2009-02-17typoRoger Dingledine
svn:r18610
2009-02-17add proposal 159, 'exit scanning', from or-devRoger Dingledine
svn:r18609
2009-02-16Possible fix for broken country settings in ExcludeExitNodes.Nick Mathewson
It turns out that we weren't updating the _ExcludeExitNodesUnion set's country numbers when we reloaded (or first loaded!) the IP-to-country file. Spotted by Lark. Bugfix on 0.2.1.6-alpha. svn:r18575
2009-02-16Remove Makefile rule for building performance.tex (now in /projects/performance)Steven Murdoch
svn:r18568
2009-02-16docdoc and changelog for r18556Roger Dingledine
svn:r18566
2009-02-16and two more things to moveRoger Dingledine
svn:r18564
2009-02-16move performance.tex overRoger Dingledine
boy, what a convoluted thing svn is svn:r18563
2009-02-16and list the new event here tooRoger Dingledine
svn:r18557
2009-02-16new controller event NEWCONSENSUS that lists the networkstatusRoger Dingledine
lines for every recommended relay. still needs docdoc and changelog entry. svn:r18556
2009-02-16reindex proposalsRoger Dingledine
svn:r18555
2009-02-15nobody's implementing proposals 134 or 140, so they getRoger Dingledine
bumped to 0.2.2.x svn:r18546
2009-02-15fill in some 'created' timeframesRoger Dingledine
svn:r18545
2009-02-13If the controller claimed responsibility for a stream, but thatRoger Dingledine
stream never finished making its connection, it would live forever in circuit_wait state. Now we close it after SocksTimeout seconds. Bugfix on 0.1.2.7-alpha; reported by Mike Perry. svn:r18516
2009-02-13Update some copyrights, change net.freehaven to org.torproject.Andrew Lewman
svn:r18514
2009-02-12Update to the "January 23 2009" ip-to-country fileRoger Dingledine
svn:r18503
2009-02-12Add SSL test description for kicks. Also spell check is aMike Perry
wonderfukl thign. svn:r18502
2009-02-12Add exit scanning proposal outline from discussions with arma.Mike Perry
svn:r18501
2009-02-11and forward-port the 0.2.0.33 and 0.2.0.34 changelogsRoger Dingledine
svn:r18497
2009-02-11Revert an erroneous part of the non-fix to bug 326, and add comments to ↵Nick Mathewson
explain why it was erroneous. svn:r18494
2009-02-11I worry that the CLEAR() macro in eventdns.c is hiding bugs. That is sad, ↵Nick Mathewson
because it was meant to make them more detectable. Change it to change stuff to garbage rather than to 0. If no bugs turn up, we can remove it in 0.2.2.x svn:r18493
2009-02-11Port some mostly cosmetic eventdns changes from libeventNick Mathewson
svn:r18492
2009-02-10Clients and non-caches do not need to cache unrecognized authority certificates.Nick Mathewson
svn:r18480
2009-02-10Enhance tor-checkkey tool so it can generate key hashes too.Nick Mathewson
svn:r18478
2009-02-10As an exit node, scrub the IP address to which we are exiting in the logs. ↵Karsten Loesing
Bugfix on 0.2.1.8-alpha. svn:r18477
2009-02-10Update WiX MSI package GUIDs for version 0.2.1.12.Martin Peck
svn:r18466
2009-02-10it turns out that removing external deliverables from the list justRoger Dingledine
because they're done will really confuse arma. svn:r18463
2009-02-10Don't extend introduction circuits indefinitely.Nick Mathewson
Doing so could run you out of relay_early cells and give you a senselessly long circuit. Patch from Karsten; may fix bug 878. svn:r18459
2009-02-09New upstream versionPeter Palfrader
svn:r18457
2009-02-09mark a few todo items as deferableRoger Dingledine
svn:r18455
2009-02-09Retry circuits if the exit node is optional and nonexistant.Nick Mathewson
Previously, when we had the chosen_exit set but marked optional, and we failed because we couldn't find an onion key for it, we'd just give up on the circuit. But what we really want to do is try again, without the forced exit node. Spotted by rovv. Another case of bug 752. I think this might be unreachable in our current code, but proposal 158 could change that. svn:r18451