summaryrefslogtreecommitdiff
path: root/opentracker.c
AgeCommit message (Collapse)Author
2007-03-06udp is working in theory, return_peers_for_torrent accepts a switch to request an udp answer. Fixed prototypes.erdgeist
2007-03-05Can now handle multiple ips/ports, can BIND to udp ports but not yet handle them, now checks if event=completed comes with left=0 and drops it, if it doesnterdgeist
2007-02-03zero copy sockets do only work on page aligned buffers for page sized blocks. So remove the code that tried to take advantage of false assuming, it would ↵erdgeist
work on any buffer
2007-02-02fixed the last commit. now it really counts the scrapes and fullscrapesdenis
2007-02-01add the scrapes to the valid requests (blue mrtg line)denis
2007-02-01el cheapo vector debuggingerdgeist
2007-02-01Use multiple output buffers to try to avoid copy on write when unnecessaryerdgeist
2007-01-31Try to shake off too stupid clients by sending an Bittorrent error rather than an HTTP 400erdgeist
2007-01-31Use signed size_t wherever appropriateerdgeist
2007-01-31size_t cant get < 0 and thus wont report errorserdgeist
2007-01-29provide mrtg with more sane stringserdgeist
2007-01-29We now answer as soon as the first \n is seenerdgeist
We do not copy input buffer to socket's input array anymore, if we find \n in first buffer We simplified parsing of input string We removed asserts that were needed to debug negative seeder counts We split static input and output buffers for obvious reasons We enlarged static output buffer to prepare multi scrape functionality We removed -lm from Makefile, this lib was only needed, when our random algorithm worked on floats We close fds now that come with an empty iocookie into write loop, those don't make sense We documented query string parser in opentracker.c a lot better to help auditing
2007-01-27complete_D_ ... ECODERTOODUMBerdgeist
2007-01-27account downloaded before early returnserdgeist
2007-01-26announce.php and scrape.php are now being handlederdgeist
2007-01-26Introduce another unspec'd command: sync which syncs the most recent peer pool for a torrent out. Do some variable type clean up. Do some code structure ↵erdgeist
clean up.
2007-01-25return random check back intervals to our clientserdgeist
2007-01-24Code cleanup reindentingerdgeist
2007-01-24backout Revision 1.67denis
2007-01-24workaround for some weird requestsdenis
2007-01-24Denis wanted scrapes and erdgeist didnt give it :)denis
2007-01-24Copy request before expanding url-encoded bytesdenis
2007-01-24Debugging for failed requests.erdgeist
2007-01-23Count successful announces extraerdgeist
2007-01-20FD-debug also in new iob_send patherdgeist
2007-01-20Full scrape, kinda testederdgeist
2007-01-20increase the timeout to make the ride a little smoother for some clientsdenis
2007-01-20-m64 is not usually a correct compile optionerdgeist
2007-01-19Debug file descriptorserdgeist
2007-01-19*seuftz*denis
2007-01-19set the interval to 1800 for announce (last change was only for scrape)denis
2007-01-18From now the OT_CLIENT_TIMEOUT will be enforced, even if some data arrived inside this intervalerdgeist
2007-01-18Tidied main loop do accept() and read() before timeout()erdgeist
2007-01-18Give our server some air to check for timedout connectionserdgeist
2007-01-18Rearranged code to try to read data as soon as socket is accepted, also raise backlog for listening socketerdgeist
2007-01-17Simple top5 by peers/seeders tableerdgeist
2007-01-16Rudimentary statserdgeist
2007-01-16Do some timeout checking on the first possible momenterdgeist
2007-01-16Previous answer to event=stopped irritated some clients. Now reply with a static nothing-messageerdgeist
2007-01-14Cast types for mrtg_scrape sprintferdgeist
2007-01-14Drop permission, if we canerdgeist
2007-01-11Documentation improved, some reindenting (again), variable types checked, unnecessary defines removederdgeist
2007-01-10Ignore SIGPIPE. This is safe, since it only raises, when the remote site died before we could send our paket back and is catched by write() returning -1erdgeist
2007-01-09Fixed a bug where I didn't replace new buffer pointer after realloc. Fixed a bug where I didnt copy enough memory when shrinking vectors. Now save some ↵erdgeist
extra bytes in header.
2007-01-08Now we also check, if no io happenserdgeist
2007-01-08Some code spacifying. Added actually checking timeouts.erdgeist
2007-01-08Indent. Sorry. Plus timeout on every connection, for every byte we got.erdgeist
2007-01-08This one breaks all ;) Lets see, what happenserdgeist
2007-01-07add timeout for clientserdgeist
2007-01-06shutdown was unnecessaryerdgeist