summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Expand)Author
2004-04-19use the real socketpair if we've got itRoger Dingledine
2004-04-19the socketpair bug is no longer a bugRoger Dingledine
2004-04-19expire wedged dir conns after 5min without write, not simply after 5minRoger Dingledine
2004-04-19give wedged dir conns a last chance to flushRoger Dingledine
2004-04-18tmp patch to also expire old clean rend_ready circs for nowRoger Dingledine
2004-04-18bugfix: you mean we're supposed to reuse fresh descriptors andRoger Dingledine
2004-04-18when you get an intro ack, don't also count it as an intro nackRoger Dingledine
2004-04-18if an intro circ waiting for an ack dies before getting one, thenRoger Dingledine
2004-04-18reverse the logic, captainRoger Dingledine
2004-04-18expire dir connections that live for more than 5 minutesRoger Dingledine
2004-04-17bugfix: give dir_conn a state before building the bridge, so it hasRoger Dingledine
2004-04-17log n_circ_ids when attaching rend streamsRoger Dingledine
2004-04-17a ! goes a long way toward more helpful debug messagesRoger Dingledine
2004-04-17include sys/limits.h if it's foundRoger Dingledine
2004-04-17util.c uses LONG_MAX, needs another include on fbsd 5.xRoger Dingledine
2004-04-17clean up some warningsRoger Dingledine
2004-04-17more useful logs to track weasel's bugRoger Dingledine
2004-04-17more debugging to hunt for a bugRoger Dingledine
2004-04-17bugfix: when you sleep your hidden-service laptop, as soonRoger Dingledine
2004-04-16fix another rare bug: when we had tried all the introtor-0.0.6incompat-mergedRoger Dingledine
2004-04-16fix a pair of duplicate mark-for-closesRoger Dingledine
2004-04-16when alice uses a port that the hidden service doesn't accept,Roger Dingledine
2004-04-16more useful logging for rend services and clientsRoger Dingledine
2004-04-16wait a minute, weasel said there was a memory-stomping bug hereRoger Dingledine
2004-04-16i was wrong about one of my invariantsRoger Dingledine
2004-04-16put in an extra assert, might find a bug these wayRoger Dingledine
2004-04-15compress end-of-second summary into one lineRoger Dingledine
2004-04-15quiet a -l info that should be -l debugRoger Dingledine
2004-04-15Must detach before attaching, or else we infinite-loopNick Mathewson
2004-04-15Strings need room for a NUL.Nick Mathewson
2004-04-15Add a missing return -1Nick Mathewson
2004-04-15make unattached rend streams expire after a while (60s)Roger Dingledine
2004-04-15narrow in closerRoger Dingledine
2004-04-14prune abandoned circs betterRoger Dingledine
2004-04-14Retry non-final-hop rendezvous failuresNick Mathewson
2004-04-14Never ever leave ri.platform unset.Nick Mathewson
2004-04-14make rendezvous points work againRoger Dingledine
2004-04-14handle more purposes in new_route_lenRoger Dingledine
2004-04-14call circuit_build_failed() even when the failure is immediateRoger Dingledine
2004-04-14Do not segfault on missing intro points.Nick Mathewson
2004-04-13remove unused variableRoger Dingledine
2004-04-13let introcircs receive acks. rend streams now deal correctly again.Roger Dingledine
2004-04-13Scan list of servers in directory after reloading fingerprint list; remove an...Nick Mathewson
2004-04-13only consider uploading descriptors every 5 secNick Mathewson
2004-04-13or, not andNick Mathewson
2004-04-13Handle END_STREAM_REASON_{EXITPOLICY|RESOLVEFAILED}Nick Mathewson
2004-04-13Mark service descriptors as dirty when we have lost an intro pointNick Mathewson
2004-04-13Separate "start-establishing-introduction" from "upload descriptors"; only up...Nick Mathewson
2004-04-13clean up circuit rebuilding some; fix some bugs we hadn't hit yetRoger Dingledine
2004-04-13Handle the rendezvous-circ elements of retrying introductions.Nick Mathewson