An open and free bittorrent tracker https://erdgeist.org/gitweb/opentracker
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

216 lines
6.4 KiB

13 years ago
13 years ago
13 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
13 years ago
13 years ago
13 years ago
13 years ago
13 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
13 years ago
13 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
13 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
** struct ot_workstruct gets ritcher (and will become even ritcher soon). This is where we encapsulate all per-request data from peer to hash to peer_id, so that it is available everywhere without passing hundreds of pointers down the stack. Most functions that do work down the stack now accept an ot_workstruct and some flags. So it can end up in the stats/event-handler where it will be the default parameter in the future. ** peer_id is now being copied by default and moved to ot_workstruct So it is available in stats and subsequent functions. ** sync scrape madness is gone SYNC_SCRAPE was intended to sync tracker state that would normally be lost on restarts i.e. downloaded counts per torrent. The way was to push it in the tracker cloud after finding all neighbouring trackers. This is madness. It never was tested and can be done per tracker by fetching stats/mode=statedump from time to time and starting opentracker with the -l option later. ** livesync thread has its own ot_workstruct now So it can behave like ot_udp and ot_http against trackerlogic.c and get rid of the first half of the embarrassing global variables. The sending half will be fixed soon [tm]. ** stats can log completed events The author recognizes the needs of original content distributors to keep track of the amount of times a work has been downloaded. While not feasible and used on openbittorrent and other open and anonymous tracker installations, a tracker user can now choose to send those events to syslog.
11 years ago
3 years ago
  1. /* This software was written by Dirk Engling <erdgeist@erdgeist.org>
  2. It is considered beerware. Prost. Skol. Cheers or whatever.
  3. $id$ */
  4. /* System */
  5. #include <sys/types.h>
  6. #include <sys/uio.h>
  7. #include <string.h>
  8. #include <pthread.h>
  9. #include <unistd.h>
  10. #include <stdlib.h>
  11. /* Libowfat */
  12. #include "socket.h"
  13. #include "ndelay.h"
  14. #include "byte.h"
  15. #include "ip6.h"
  16. /* Opentracker */
  17. #include "trackerlogic.h"
  18. #include "ot_livesync.h"
  19. #include "ot_accesslist.h"
  20. #include "ot_stats.h"
  21. #include "ot_mutex.h"
  22. #ifdef WANT_SYNC_LIVE
  23. char groupip_1[4] = { 224,0,23,5 };
  24. #define LIVESYNC_INCOMING_BUFFSIZE (256*256)
  25. #define LIVESYNC_OUTGOING_BUFFSIZE_PEERS 1480
  26. #define LIVESYNC_OUTGOING_WATERMARK_PEERS (sizeof(ot_peer)+sizeof(ot_hash))
  27. #define LIVESYNC_MAXDELAY 15 /* seconds */
  28. enum { OT_SYNC_PEER };
  29. /* Forward declaration */
  30. static void * livesync_worker( void * args );
  31. /* For outgoing packets */
  32. static int64 g_socket_in = -1;
  33. /* For incoming packets */
  34. static int64 g_socket_out = -1;
  35. static pthread_mutex_t g_outbuf_mutex = PTHREAD_MUTEX_INITIALIZER;
  36. char g_outbuf[LIVESYNC_OUTGOING_BUFFSIZE_PEERS];
  37. static size_t g_outbuf_data;
  38. static ot_time g_next_packet_time;
  39. static pthread_t thread_id;
  40. void livesync_init( ) {
  41. if( g_socket_in == -1 )
  42. exerr( "No socket address for live sync specified." );
  43. /* Prepare outgoing peers buffer */
  44. memcpy( g_outbuf, &g_tracker_id, sizeof( g_tracker_id ) );
  45. uint32_pack_big( g_outbuf + sizeof( g_tracker_id ), OT_SYNC_PEER);
  46. g_outbuf_data = sizeof( g_tracker_id ) + sizeof( uint32_t );
  47. g_next_packet_time = g_now_seconds + LIVESYNC_MAXDELAY;
  48. pthread_create( &thread_id, NULL, livesync_worker, NULL );
  49. }
  50. void livesync_deinit() {
  51. if( g_socket_in != -1 )
  52. close( g_socket_in );
  53. if( g_socket_out != -1 )
  54. close( g_socket_out );
  55. pthread_cancel( thread_id );
  56. }
  57. void livesync_bind_mcast( ot_ip6 ip, uint16_t port) {
  58. char tmpip[4] = {0,0,0,0};
  59. char *v4ip;
  60. if( !ip6_isv4mapped(ip))
  61. exerr("v6 mcast support not yet available.");
  62. v4ip = ip+12;
  63. if( g_socket_in != -1 )
  64. exerr("Error: Livesync listen ip specified twice.");
  65. if( ( g_socket_in = socket_udp4( )) < 0)
  66. exerr("Error: Cant create live sync incoming socket." );
  67. ndelay_off(g_socket_in);
  68. if( socket_bind4_reuse( g_socket_in, tmpip, port ) == -1 )
  69. exerr("Error: Cant bind live sync incoming socket." );
  70. if( socket_mcjoin4( g_socket_in, groupip_1, v4ip ) )
  71. exerr("Error: Cant make live sync incoming socket join mcast group.");
  72. if( ( g_socket_out = socket_udp4()) < 0)
  73. exerr("Error: Cant create live sync outgoing socket." );
  74. if( socket_bind4_reuse( g_socket_out, v4ip, port ) == -1 )
  75. exerr("Error: Cant bind live sync outgoing socket." );
  76. socket_mcttl4(g_socket_out, 1);
  77. socket_mcloop4(g_socket_out, 0);
  78. }
  79. /* Caller MUST hold g_outbuf_mutex. Returns with g_outbuf_mutex unlocked */
  80. static void livesync_issue_peersync( ) {
  81. char mycopy[LIVESYNC_OUTGOING_BUFFSIZE_PEERS];
  82. size_t data = g_outbuf_data;
  83. memcpy( mycopy, g_outbuf, data );
  84. g_outbuf_data = sizeof( g_tracker_id ) + sizeof( uint32_t );
  85. g_next_packet_time = g_now_seconds + LIVESYNC_MAXDELAY;
  86. /* From now this thread has a local copy of the buffer and
  87. has modified the protected element */
  88. pthread_mutex_unlock(&g_outbuf_mutex);
  89. socket_send4(g_socket_out, mycopy, data, groupip_1, LIVESYNC_PORT);
  90. }
  91. static void livesync_handle_peersync( struct ot_workstruct *ws ) {
  92. int off = sizeof( g_tracker_id ) + sizeof( uint32_t );
  93. /* Now basic sanity checks have been done on the live sync packet
  94. We might add more testing and logging. */
  95. while( off + (ssize_t)sizeof( ot_hash ) + (ssize_t)sizeof( ot_peer ) <= ws->request_size ) {
  96. memcpy( &ws->peer, ws->request + off + sizeof(ot_hash), sizeof( ot_peer ) );
  97. ws->hash = (ot_hash*)(ws->request + off);
  98. if( !g_opentracker_running ) return;
  99. if( OT_PEERFLAG(&ws->peer) & PEER_FLAG_STOPPED )
  100. remove_peer_from_torrent( FLAG_MCA, ws );
  101. else
  102. add_peer_to_torrent_and_return_peers( FLAG_MCA, ws, /* amount = */ 0 );
  103. off += sizeof( ot_hash ) + sizeof( ot_peer );
  104. }
  105. stats_issue_event(EVENT_SYNC, 0,
  106. (ws->request_size - sizeof( g_tracker_id ) - sizeof( uint32_t ) ) /
  107. ((ssize_t)sizeof( ot_hash ) + (ssize_t)sizeof( ot_peer )));
  108. }
  109. /* Tickle the live sync module from time to time, so no events get
  110. stuck when there's not enough traffic to fill udp packets fast
  111. enough */
  112. void livesync_ticker( ) {
  113. /* livesync_issue_peersync sets g_next_packet_time */
  114. pthread_mutex_lock(&g_outbuf_mutex);
  115. if( g_now_seconds > g_next_packet_time &&
  116. g_outbuf_data > sizeof( g_tracker_id ) + sizeof( uint32_t ) )
  117. livesync_issue_peersync();
  118. else
  119. pthread_mutex_unlock(&g_outbuf_mutex);
  120. }
  121. /* Inform live sync about whats going on. */
  122. void livesync_tell( struct ot_workstruct *ws ) {
  123. pthread_mutex_lock(&g_outbuf_mutex);
  124. memcpy( g_outbuf + g_outbuf_data, ws->hash, sizeof(ot_hash) );
  125. memcpy( g_outbuf + g_outbuf_data + sizeof(ot_hash), &ws->peer, sizeof(ot_peer) );
  126. g_outbuf_data += sizeof(ot_hash) + sizeof(ot_peer);
  127. if( g_outbuf_data >= LIVESYNC_OUTGOING_BUFFSIZE_PEERS - LIVESYNC_OUTGOING_WATERMARK_PEERS )
  128. livesync_issue_peersync();
  129. else
  130. pthread_mutex_unlock(&g_outbuf_mutex);
  131. }
  132. static void * livesync_worker( void * args ) {
  133. struct ot_workstruct ws;
  134. ot_ip6 in_ip; uint16_t in_port;
  135. (void)args;
  136. /* Initialize our "thread local storage" */
  137. ws.inbuf = ws.request = malloc( LIVESYNC_INCOMING_BUFFSIZE );
  138. ws.outbuf = ws.reply = 0;
  139. memcpy( in_ip, V4mappedprefix, sizeof( V4mappedprefix ) );
  140. while( 1 ) {
  141. ws.request_size = socket_recv4(g_socket_in, (char*)ws.inbuf, LIVESYNC_INCOMING_BUFFSIZE, 12+(char*)in_ip, &in_port);
  142. /* Expect at least tracker id and packet type */
  143. if( ws.request_size <= (ssize_t)(sizeof( g_tracker_id ) + sizeof( uint32_t )) )
  144. continue;
  145. if( !accesslist_isblessed(in_ip, OT_PERMISSION_MAY_LIVESYNC))
  146. continue;
  147. if( !memcmp( ws.inbuf, &g_tracker_id, sizeof( g_tracker_id ) ) ) {
  148. /* TODO: log packet coming from ourselves */
  149. continue;
  150. }
  151. switch( uint32_read_big( sizeof( g_tracker_id ) + (char *)ws.inbuf ) ) {
  152. case OT_SYNC_PEER:
  153. livesync_handle_peersync( &ws );
  154. break;
  155. default:
  156. break;
  157. }
  158. }
  159. /* Never returns. */
  160. return NULL;
  161. }
  162. #endif
  163. const char *g_version_livesync_c = "$Source: /home/cvsroot/opentracker/ot_livesync.c,v $: $Revision: 1.20 $\n";