Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | bugfix: we were closing socks-request connections immediately, rather | Roger Dingledine | 2004-03-29 |
| | | | | | | | than flushing them, if they were timing out before we marked them svn:r1366 | ||
* | clean up a few more bugs | Roger Dingledine | 2004-03-29 |
| | | | | svn:r1365 | ||
* | fix typo | Roger Dingledine | 2004-03-29 |
| | | | | svn:r1364 | ||
* | make normal exit nodes less likely to crash from the conn-munging bug | Roger Dingledine | 2004-03-29 |
| | | | | | | | but leave dirservers just as likely to crash svn:r1363 | ||
* | Re-enable unit tests for directory generation. | Nick Mathewson | 2004-03-29 |
| | | | | svn:r1362 | ||
* | Remove descriptors that are older than 24 hours from the directory. Use ↵ | Nick Mathewson | 2004-03-29 |
| | | | | | | strlcat instead of strncat to generate directories. svn:r1361 | ||
* | Possible fix for c99 aliasing warnings | Nick Mathewson | 2004-03-29 |
| | | | | svn:r1360 | ||
* | make it less noisy for normal ORs when a server uses a false nickname | Roger Dingledine | 2004-03-28 |
| | | | | svn:r1359 | ||
* | make dns resolves a lot faster | Roger Dingledine | 2004-03-28 |
| | | | | | | | (but only enable this once we've found the bug) svn:r1358 | ||
* | oh, i forgot to enable the find-munged-conn function | Roger Dingledine | 2004-03-28 |
| | | | | svn:r1357 | ||
* | hold socks reply open until flushed (if possible) | Roger Dingledine | 2004-03-28 |
| | | | | svn:r1356 | ||
* | there is yet another dns-pending-conn-clobbering bug | Roger Dingledine | 2004-03-28 |
| | | | | | | | somewhere svn:r1354 | ||
* | work around a tsocks bug | Roger Dingledine | 2004-03-27 |
| | | | | | | | | when an AP connection dies early, be sure to do a socks reject for it. if you just hang up, tsocks goes into an infinite loop. svn:r1352 | ||
* | fix an assert trigger where an OP would fail to handshake, and we'd | Roger Dingledine | 2004-03-27 |
| | | | | | | | expect it to have a nickname. svn:r1351 | ||
* | only build circuits after we've fetched the directory | Roger Dingledine | 2004-03-27 |
| | | | | | | | | | | | | this resolves a subtle bug where tor clients were preferentially using the directory servers, since when they start building circuits they know only about the directory servers on the other hand, it now takes longer after startup before there's a working circuit. so it goes. svn:r1350 | ||
* | make it less noisy to consider exit policies | Roger Dingledine | 2004-03-27 |
| | | | | svn:r1349 | ||
* | oops, 0.0.4 is more broken than 0.0.3 | Roger Dingledine | 2004-03-26 |
| | | | | | | | maybe this will fix it svn:r1347 | ||
* | clean up some 'warn' messages while i'm at it | Roger Dingledine | 2004-03-26 |
| | | | | svn:r1342 | ||
* | fix two assert triggers (darn it, I hate releasing software) | Roger Dingledine | 2004-03-26 |
| | | | | | | | | when connecting to a dirserver or OR and the network is down, we would crash. svn:r1340 | ||
* | a missing comma could wreak all sorts of havoc. | Roger Dingledine | 2004-03-26 |
| | | | | svn:r1335 | ||
* | Remove unused open circuits when there are too many, not when they are too old. | Nick Mathewson | 2004-03-21 |
| | | | | svn:r1332 | ||
* | Dont expire circuits we didnt originate. This may be redundant somehow, but ↵ | Nick Mathewson | 2004-03-21 |
| | | | | | | better safe than sorry. svn:r1331 | ||
* | Add an autoconf test for whether unaligned int access is permitted. | Nick Mathewson | 2004-03-21 |
| | | | | | | | | It works on macos, and works correctly when I twiddle the test code to generate a spurious segfault. Will it work anywhere else? svn:r1329 | ||
* | Replace all connection_removes outsice of main.c with connection_mark_for_close | Nick Mathewson | 2004-03-21 |
| | | | | svn:r1328 | ||
* | Add a RunTesting option to try to learn link state by creating test ↵ | Nick Mathewson | 2004-03-21 |
| | | | | | | circuits, even when SocksPort is off. svn:r1327 | ||
* | Add macros and functions to wrap memcpy/alignment logic. | Nick Mathewson | 2004-03-21 |
| | | | | svn:r1326 | ||
* | catch signals in a more portable way, so solaris can catch them too | Roger Dingledine | 2004-03-20 |
| | | | | svn:r1324 | ||
* | make uname() more likely to work on solaris | Roger Dingledine | 2004-03-20 |
| | | | | | | | (linux says 0 is success, solaris says 1 is success) svn:r1322 | ||
* | make OR not crash when non-OR connection dies | Roger Dingledine | 2004-03-20 |
| | | | | svn:r1319 | ||
* | warn and fail if server chose a nickname with illegal characters | Roger Dingledine | 2004-03-20 |
| | | | | svn:r1318 | ||
* | make servers not crash when they rep-hist-note circuits that don't start at them | Roger Dingledine | 2004-03-20 |
| | | | | svn:r1317 | ||
* | deal with hardware word alignment | Roger Dingledine | 2004-03-20 |
| | | | | | | | | | this was causing the seg faults on sparc processors i wonder if i got them all. svn:r1314 | ||
* | Integrate the new "rephist" [rep(utation) hist(ory)] module to trace | Nick Mathewson | 2004-03-20 |
| | | | | | | | | | | | | | | successful/failed connections, successful/failed extends, and connection uptimes. It's still not done: more tests are needed, and not everything calls connection/circuit_mark_for_close properly. This skews the results. Also, there needs to be a 'testing' mode for non-OP ORs, where they periodically build circuits just to test whether extends work. svn:r1313 | ||
* | Add initial history-tracking backend code for reputation module. Not called ↵ | Nick Mathewson | 2004-03-20 |
| | | | | | | or tested yet. svn:r1310 | ||
* | Use strmap code for client DNS. | Nick Mathewson | 2004-03-20 |
| | | | | svn:r1309 | ||
* | Add some wrappers around SPLAY_* for the common map-from-string-to-X case. | Nick Mathewson | 2004-03-19 |
| | | | | | | | | It will probably be less blindingly fast than using SPLAY_* directly, but only slightly so. svn:r1306 | ||
* | getpid returns a pid_t. we all know that's an int. but solaris doesn't know. | Roger Dingledine | 2004-03-19 |
| | | | | svn:r1305 | ||
* | isspace and friends take an int. solaris cares. | Roger Dingledine | 2004-03-19 |
| | | | | svn:r1303 | ||
* | ship strlcat and strlcpy with the tarball, but don't compile them | Roger Dingledine | 2004-03-19 |
| | | | | svn:r1297 | ||
* | define INADDR_NONE so we compile on solaris | Roger Dingledine | 2004-03-18 |
| | | | | svn:r1295 | ||
* | clean up some presentation and comments | Roger Dingledine | 2004-03-18 |
| | | | | svn:r1294 | ||
* | Include strlcpy and strlcat where not available, so our string ops can be ↵ | Nick Mathewson | 2004-03-17 |
| | | | | | | less error-prone. svn:r1289 | ||
* | make directory format tests quiet again | Roger Dingledine | 2004-03-15 |
| | | | | svn:r1287 | ||
* | make it warn about internal IPs not only if we had to guess, | Roger Dingledine | 2004-03-15 |
| | | | | | | | but also if they used a hostname rather than an IP svn:r1285 | ||
* | only complain about private IPs if we had to guess | Roger Dingledine | 2004-03-15 |
| | | | | svn:r1279 | ||
* | and another one | Roger Dingledine | 2004-03-15 |
| | | | | svn:r1278 | ||
* | fix a byte order bug | Roger Dingledine | 2004-03-15 |
| | | | | svn:r1277 | ||
* | bugfix: address that strcat vulnerability in circuit.c | Roger Dingledine | 2004-03-14 |
| | | | | svn:r1273 | ||
* | prevent servers from publishing an internal IP as their address | Roger Dingledine | 2004-03-14 |
| | | | | svn:r1272 | ||
* | teach us to recognize internal IPs | Roger Dingledine | 2004-03-14 |
| | | | | svn:r1271 |