Commit Graph
5732 Commits
Author SHA1 Message Date
Roger Dingledine dad760ddd1 one of the debugging aids was misleading
svn:r3000
2004-11-28 06:33:14 +00:00
Nick Mathewson 32651dd108 connect() on win32 can do more things than we thought?
svn:r2999
2004-11-28 05:58:55 +00:00
Nick Mathewson 0702179d28 Very oops. make windows nonblocking sockets nonblocking
svn:r2998
2004-11-28 05:57:18 +00:00
Nick Mathewson f77ff938b7 remember; tor_socket_errno has side effects!
svn:r2997
2004-11-28 05:48:02 +00:00
Roger Dingledine 40c97ac504 move the rendezous section of the torrc to the middle, since
it isn't server-only, and indeed it may be unwise to run a
hidden service on a server.


svn:r2995
2004-11-27 08:18:22 +00:00
Roger Dingledine e5e45fe980 inform him more sanely
svn:r2993
2004-11-26 04:19:03 +00:00
Roger Dingledine a7f92d7ffe inform the user that you noticed when he sent a USR2
svn:r2992
2004-11-26 04:17:57 +00:00
Roger Dingledine 81dad3e08b that's enough changes to bump to rc4-cvs
svn:r2991
2004-11-26 04:08:07 +00:00
Roger Dingledine f9a0ab62b7 if you want to handle -USR2, you should catch it too.
svn:r2990
2004-11-26 04:03:19 +00:00
Roger Dingledine 7c9a707900 remove emacs droppings, since nick says he doesn't need them anymore
svn:r2989
2004-11-26 04:00:55 +00:00
Roger Dingledine 02a30a348a give up trying to track down the reason we're calling package_inbuf
with a package window of 0. it's safe to just catch the error cases
there and do the right thing.


svn:r2988
2004-11-26 03:43:54 +00:00
Roger Dingledine 179ad27b34 call set_max_file_descriptors() after we configure logs, so
it doesn't always do loglevel info logs on windows.


svn:r2987
2004-11-26 03:05:55 +00:00
Roger Dingledine e24057b2bf when warning that a version is too new, don't use the confusing
text "is_obsolete_version():".


svn:r2986
2004-11-25 05:49:13 +00:00
Roger Dingledine d254aeb09e bump to 0.0.9rc3
svn:r2984
2004-11-25 05:40:06 +00:00
Nick Mathewson f91152ce6a Clarify a bunch of log messages
svn:r2983
2004-11-25 04:20:10 +00:00
Peter Palfrader f2a96ce1fe Flush the logfd after we print "Tor opening log file", so we don't see those messages days later
svn:r2980
2004-11-24 08:42:06 +00:00
Roger Dingledine 914d346095 update conn_state_to_string
svn:r2979
2004-11-24 07:58:15 +00:00
Roger Dingledine a2042a8b1d a little help for dfc to debug his "resolve 800 things at once" test
svn:r2978
2004-11-24 07:51:30 +00:00
Roger Dingledine c89afb7917 avoid infinite loop in tor-resolve if tor hangs up on it
svn:r2977
2004-11-24 07:45:42 +00:00
Roger Dingledine 83e7043eb8 don't process marked-for-close conns further.
(fix assert trigger -- if we're lucky.)


svn:r2975
2004-11-24 06:41:58 +00:00
Roger Dingledine 7d55f3685e make tor-resolve work, uh, more. this time for sure!
svn:r2974
2004-11-24 06:16:36 +00:00
Roger Dingledine cca469ec58 make tor-resolve work again
svn:r2973
2004-11-24 06:01:52 +00:00
Roger Dingledine 0eebfe3558 bugfix: let socks do multiple rounds of negotiation again
(which is needed for socks5)


svn:r2970
2004-11-24 04:35:28 +00:00
Roger Dingledine 090c84627c patch to nick's patch
svn:r2969
2004-11-24 03:58:45 +00:00
Nick Mathewson 74f74a06c6 make nt service version match actual version
svn:r2968
2004-11-24 00:55:39 +00:00
Nick Mathewson e84f964aad bump windows version stuff to 009rc2
svn:r2967
2004-11-24 00:33:30 +00:00
Nick Mathewson 1bec1b3d37 Better notice for hibernate interval spans; when estimate is 0, start out awake.
svn:r2963
2004-11-23 22:34:23 +00:00
Nick Mathewson 5fa55c8317 Bugfix: when no units are given, default to seconds or bytes.
svn:r2960
2004-11-23 10:52:27 +00:00
Roger Dingledine a284b0db8f another slightly less confusing message
nick: we need to figure out if we should say anything different
here when the wake-up time and the expected back-to-sleep time
have both passed.


svn:r2956
2004-11-23 10:05:56 +00:00
Roger Dingledine b6cb547667 less confusing message upon hibernation
svn:r2955
2004-11-23 09:21:03 +00:00
Roger Dingledine 921c40abe7 fail more gracefully next time
svn:r2954
2004-11-23 09:01:37 +00:00
Roger Dingledine 57a392d8d8 initialize rephist before we init_from_config, since init_from_config
can make us init_keys if we're going to hibernate, which makes us build
a router_desc.


svn:r2953
2004-11-23 09:00:55 +00:00
Roger Dingledine d7fb38cfe3 update unit tests to reflect changed interface
svn:r2952
2004-11-23 08:36:26 +00:00
Roger Dingledine 7ed738b34c fix some typos; move to 0.0.9rc1
svn:r2951
2004-11-23 07:37:25 +00:00
Nick Mathewson f5319a010d Dr. Seuss on iterating circular lists: "It's fun to have fun, but you
have to know how."


svn:r2948
2004-11-23 06:08:08 +00:00
Nick Mathewson 0a58bbe0dd Fix critical bug in circuit_list_path: cpath is a circular list! (Also reimplement circuit_log_cpath using circuit_list_cpath).
svn:r2946
2004-11-23 00:11:36 +00:00
Nick Mathewson cd70264377 Clean up some logging and interfaces
svn:r2945
2004-11-23 00:08:26 +00:00
Nick Mathewson 22dba27d8d Normalize a few more kinds of whitespace. We now dislike:
- func (args)
  - if (x){
This doesn't normalize if(x), for(x); while(x), and friends.


svn:r2943
2004-11-22 23:28:26 +00:00
Nick Mathewson fe32c2d141 Fix sign-related warnings
svn:r2942
2004-11-22 22:50:17 +00:00
Nick Mathewson d4754b334c Compile cleanly on windows; prevent some insane bandwidth cases (e.g., "BandwidthBurst 1000 TB" from occuring.
svn:r2941
2004-11-22 22:24:10 +00:00
Nick Mathewson 983a335f3b Normalize space
svn:r2939
2004-11-22 22:13:34 +00:00
Nick Mathewson f17c08beed Apply NT service patch from Osamu Fujino. This still has some problems (indicated in a comment adapted from the patch submission email), but it is better than before and better than nothing.
svn:r2938
2004-11-22 22:10:37 +00:00
Nick Mathewson 956b463dfe Implement weekly/monthly/daily accounting
svn:r2936
2004-11-22 21:56:51 +00:00
Nick Mathewson 55df2620c6 Add function to format local time in ISO format
svn:r2935
2004-11-22 21:38:44 +00:00
Nick Mathewson 805c9e3673 Absolutely never call close_temp_logs while validating log options.
svn:r2934
2004-11-22 20:08:07 +00:00
Roger Dingledine 4f2c2c9948 The crowning bugfix.
The problem was that with high load, circuit package window was
reaching 0. Whenever we got a circuit-level sendme, we were
reading a lot on each socket, but only writing out a bit. So we
would eventually reach eof. This would be noticed and acted on
even when there are still bytes sitting in the inbuf.


svn:r2932
2004-11-21 11:30:33 +00:00
Roger Dingledine 27b8797245 clean up some more.
i think i'm getting closer.


svn:r2931
2004-11-21 11:20:28 +00:00
Roger Dingledine 310a8da0d2 break reached_eof() out of process_inbuf()
svn:r2930
2004-11-21 10:14:57 +00:00
Roger Dingledine 671d84dc2f some more debugging aids
svn:r2929
2004-11-21 09:39:01 +00:00
Roger Dingledine 6a516dfdd3 be more greedy about filling up all relay cells.
this may have some bugs in it still.
and it may end up not being what we want to do.


svn:r2928
2004-11-21 07:43:12 +00:00