71efb5c74b
add per-priority-level endgame mode
2007-12-26 07:25:50 +00:00
1f3084179e
remove debugging messages that aren't needed anymore.
2007-12-26 07:19:05 +00:00
e870440025
add a per-block request TTL of 120 seconds
2007-12-25 17:22:51 +00:00
819b83d90d
fix another file priority bug.
2007-12-25 06:37:21 +00:00
f8bc11e094
fix a couple of related file priority bugs.
2007-12-25 05:42:33 +00:00
473907d891
a little housekeeping: move tr_torrent stuff into its own header
2007-12-25 05:37:32 +00:00
16e273ed05
fix torrent parsing bug reported by m1b, SoftwareElves, and BentMyWookie
2007-12-25 01:23:07 +00:00
6db02c401b
add message queueing flag to tr_initFull() too
2007-12-24 16:12:23 +00:00
bc84156d4f
add a messageLevel argument to tr_initFull()
2007-12-24 16:02:36 +00:00
35adeec1d0
(gtk) fix startup port issue by hitting it with a damn big hammer
2007-12-24 07:02:40 +00:00
a33f68a344
fix Gimp_'s problem
2007-12-24 05:17:12 +00:00
401cfbdb6b
replace the per-torrent PEX setting with a global one, as per discussion with SoftwareElves and BentMyWookie
2007-12-24 05:03:40 +00:00
b13d8d1dd6
save a copy of the torrent's metadata in tr_torrentInit(), not tr_torrentParse(). remove dead code from metainfo.c.
2007-12-22 18:00:47 +00:00
c946168b1e
(1) fix peer limits in fastresume (2) possible fastresume corruption (3) add `save' field to torrent ctor
2007-12-22 17:30:31 +00:00
3ecb9910e0
save per-torrent peer limits in the fastresume file
2007-12-22 16:11:15 +00:00
2764f9a140
fix indentation errors in r4278.
...
fix consistency errors between ctor and the new functions in r4278.
2007-12-22 04:11:17 +00:00
3449e2d15c
separate getting/setting # connections and # unchoked
2007-12-22 04:00:19 +00:00
7542573044
brush away the last remnants of tr_torrentInit()
2007-12-22 03:51:12 +00:00
9148c9211f
(gtk) tweak the display speed/size units again.
2007-12-22 03:01:14 +00:00
93f69d2b5c
(gtk) add options to set global & per-torrent peer connection limits
2007-12-22 00:06:38 +00:00
8c7434dded
fix a different r4267 bug, also reported by John_Clay
2007-12-21 22:38:01 +00:00
99d6e5deab
fix r4267 crash reported by John_Clay
2007-12-21 22:29:22 +00:00
69f1ee48c1
remove some stubbed code that should've been removed in the last checkin
2007-12-21 22:24:37 +00:00
287975caf5
the torrent ctor is here.
2007-12-21 22:18:40 +00:00
f1baca587b
add torrent-ctor so that xcode can be updated. it's not plugged in yet though.
2007-12-21 16:51:42 +00:00
5c4f841c7e
added tr_torrentSetPeerLimits(), tr_torrentGetPeerLimits(), tr_setGlobalPeerLimit(), tr_getGlobalPeerLimit()
2007-12-20 21:44:16 +00:00
cf30fac9cf
upgrade to miniupnp-20071213
2007-12-20 20:18:22 +00:00
911a6ce32e
remove unused variables
2007-12-20 01:15:13 +00:00
75d4a2b8c2
fix an obscure bug in tr_torrentStat() where (% verified + % unverified + % unavailable) could be > 1.0
2007-12-19 05:57:55 +00:00
e9ca72237f
if tr_torrentStat() is called again in the same second for a torrent, return the cached value.
2007-12-19 02:44:21 +00:00
d58d5210d0
fix tr_stat error: fields desiredSize and desiredAvailable could be too large.
2007-12-18 05:01:00 +00:00
371287136a
fix build error reported by berkut in the forums
2007-12-17 02:21:06 +00:00
36391e02c1
add support for TRANSMISSION_HOME environment variable so I can have multiple copies of Transmission running concurrently in different directories.
2007-12-15 22:22:30 +00:00
4ff5b472a8
raise the maximum connection limit a bit.
2007-12-15 16:36:43 +00:00
1cb314f0cc
try to make the natpmp error messages yet more helpful. thanks to m1b and Lacrocivious
2007-12-15 06:10:24 +00:00
fe0640caa7
add messages asking the user to check & make sure that port forwarding is enabled on their router
2007-12-15 05:28:16 +00:00
20c5f2a77c
fix off-by-one introduced in r4153
2007-12-15 05:05:23 +00:00
10b6e8ff8f
lessons learned from 0.82:
...
* if we can't bind a socket to a peer, take that peer out of the retry pool.
* also, try to connect with more peers up-front.
2007-12-15 04:26:31 +00:00
4da5fc85bb
fix bad log message
2007-12-15 03:25:13 +00:00
f8f4e1fc32
more helpful error message
2007-12-15 03:17:50 +00:00
99e8b9a19d
two arguments were transposed.
2007-12-15 03:14:16 +00:00
6ded40d360
increase the number of tracker connections we allow during shutdown.
2007-12-13 23:43:50 +00:00
7582f4a929
(maybe) fix shutdown crash in "pulse" reported by pea_
2007-12-13 21:48:39 +00:00
de5c84a52c
allocate more sockets to the tracker during shutdown to help get all the `stopped' commands done
2007-12-13 21:08:05 +00:00
bddcf5e10d
fix crash on shutdown on natpmp routers reported by SoftwareElves
2007-12-13 20:19:52 +00:00
7f335e1bf0
follow Chinstrap's suggestion of waiting a small interval between sending nat commands as per the nat spec
2007-12-13 19:28:51 +00:00
a51d790d81
take pea_'s suggestion of not sending out natpmp/upnp discover messages until port forwarding is enabled
2007-12-13 18:56:22 +00:00
39773104be
accept tiennou's file-checking tweak that clarifies the log messages
2007-12-13 17:28:13 +00:00
c4e6a04131
tweak the debug messages
2007-12-13 17:18:35 +00:00
1b6a8a9504
display the infinity symbol in stats for ratio when there is upload but no download
2007-12-13 04:31:38 +00:00
e1d10bba7e
add extra debugging messages for upnpDiscover()
2007-12-13 02:48:30 +00:00
c8c9012697
fix the 500ms remap reported by Chinstrap
2007-12-13 01:20:16 +00:00
c2e314743f
added more upnp debugging messages for Chinstrap_
2007-12-09 17:50:05 +00:00
7249aed5e4
fix r4109 oops: a debugging console message snuck into svn
2007-12-09 17:17:44 +00:00
ffe9e52e14
I think this one-line change may fix all the "Transmission 0.95 ate my data, corrupted my hard drive, and killed my cat!" complaints
2007-12-09 15:26:27 +00:00
33f284ead1
let the other config directories that we create also respect umask
2007-12-09 15:22:01 +00:00
67a86a2274
finish the `shared' update: bind & listen the incoming port.
...
better logging in network code.
2007-12-09 00:28:34 +00:00
b17d8c34f8
undo r4093 for bmw
2007-12-08 19:47:44 +00:00
008a403463
ensure that the `file' variable passed to tr_msg() is never NULL
2007-12-08 19:42:07 +00:00
73ee11f233
Use libnatpmp for port mapping. rewrite the upnp/natpmp manager.
...
NOTE: this will break the xpjets build until SoftwareElves or a volunteer patches the xcode project file to make a libnatpmp library just like was done for libminiupnp.
2007-12-08 19:34:15 +00:00
d3d1263a84
use fizz' suggestion on from ticket #534 on how to make platform.c more portable wrt header include order
2007-12-07 19:25:54 +00:00
481a210e82
fix tracker connection error reported by Gimp_
2007-12-05 02:06:05 +00:00
39d8c74fcd
small tweak to play nice with old C compilers. Thanks to fizz for reporting & submitting a patch
2007-12-03 19:52:55 +00:00
de9080e9ef
fix tyop
2007-12-03 16:51:22 +00:00
f0d1fa0a76
finish fixing the recursive mutex + cond bug reported by sedatg & Waldorf
2007-12-03 15:27:38 +00:00
c68af92039
fix very annoying recursive mutex + pthread cond bug reported by setatg, Waldorf, and many others
2007-12-03 04:06:45 +00:00
7b068ed926
more mutex work
2007-12-02 18:20:37 +00:00
5e88ab235b
possibly get the pthread changes compiling for old versions of Linux
2007-12-02 17:15:52 +00:00
b14449fe80
possible fix for the crash reported by sedatg in ticket #519
2007-12-02 16:35:44 +00:00
abef433926
improve upload speeds when the speed limits are uncapped.
2007-12-01 23:08:34 +00:00
f534d0929c
use PMP_MCAST_ADDR, instead of a hardcoded address, in net.c
2007-12-01 05:11:30 +00:00
78f3c40484
get nat-pmp working again in Leopard. Thanks to Chinstrap for tracking down the problem & submitting a patch. See the comment in this changeset for more details.
2007-12-01 05:01:02 +00:00
0e1e7efd3b
take inspiration from Solomon when handling odd tracker errors that confuse tracker's request queue: stop the torrent.
...
when the tracker gives us errors that confuse the tracker work queue,
2007-11-30 23:37:44 +00:00
f2a08d272f
be more forgiving of tracker failure
2007-11-30 17:10:33 +00:00
332019593d
speed tweaks based on whitty's feedback
2007-11-29 02:56:31 +00:00
f9670dc7ac
throw away all pending `scrape' requests on shutdown... we don't need them then, so they just slow us down
2007-11-29 02:31:21 +00:00
827dc86bb4
rewrite the tracker code. this should improve and/or fix a number of bugs, including "too many open files", "router death", "slow internet", and the mutex release crash.
2007-11-29 00:43:58 +00:00
1e4b3a0577
make the peer stats a little easier to understand
2007-11-27 21:39:41 +00:00
398bc04bcf
housekeeping. revert r3984 at tiennou's request
2007-11-27 15:39:59 +00:00
27f08f325d
fix upload speed limit bug tracked down by makira_
2007-11-27 04:27:03 +00:00
d0047acc86
reimplement tr_buildPath() using struct evbuffer
2007-11-27 04:11:39 +00:00
0663c4aef7
possible fix for Zeph's "too many open files" bug report.
2007-11-26 21:26:01 +00:00
81ea9b4abd
sloppy, sloppy. what should've been 1 commit has been 5.
2007-11-26 20:44:53 +00:00
033ee7ff58
fix crash-on-exit introduced by stats code
2007-11-26 20:44:19 +00:00
cf85da1759
plug in the stats.filesAdded field
2007-11-26 20:37:07 +00:00
548e0324ab
fix stupid bug introduced in the last stats commit
2007-11-26 20:27:07 +00:00
56b71bd144
remove dead code
2007-11-26 20:22:08 +00:00
07f2e6365a
move tr_getDefaultRoute() to natpmp.c because it's the only code that uses it
2007-11-26 20:21:52 +00:00
c3c4b699b9
stats
2007-11-26 20:20:46 +00:00
2185081fcb
more stats
2007-11-26 20:05:46 +00:00
bf3bd6bd5d
avoid divide-by-0 in stats
2007-11-26 19:18:00 +00:00
6b5f1839aa
get last commit building
2007-11-26 19:10:53 +00:00
3aae5534bd
work on the stats a little bit.
2007-11-26 18:55:36 +00:00
0ed3610468
fix crash on startup reported by Gimp_
2007-11-26 05:03:28 +00:00
e354b82452
disable SWIFT on private trackers, because they are typically ratio-based sites where you *want* to upload as much as possible to deadbeats
2007-11-26 04:05:12 +00:00
63cb8506ca
fix crash when loading fastresume files
2007-11-26 03:54:20 +00:00
ec042e7278
enable SWIFT in trunk
2007-11-25 17:09:09 +00:00
a720545256
fix peer-mgr.c shear between 0.9x and trunk
2007-11-25 17:07:12 +00:00
d735989aec
fix clientIsSnubbedBy() to work correctly when we're a seed
2007-11-25 16:57:08 +00:00
bd2de0888e
fix Wyzo
2007-11-24 19:36:26 +00:00
4a53c761a3
add Wyzo to clients list
2007-11-24 19:19:22 +00:00
18b3b388c0
fix oops from last commit
2007-11-23 03:01:13 +00:00
5ad14e4abc
fix another 0.94 prerelease crash report... this one from when finishing verifying a torrent that's running.
2007-11-23 02:49:06 +00:00
3b04f824e0
fix libT crash reported in the forums
2007-11-23 02:14:21 +00:00
8caa113d4b
fix LTEP PEX error reported by Adrian Ulrich in the BitTorrent mailing list
2007-11-22 06:48:08 +00:00
6df942138b
fix portmapping crash.
2007-11-22 06:13:57 +00:00
87d4bcd278
Change the default port from 9090 to 51413 to get in compliance with IANA's port number assignment scheme. (1024 through 49151 are reserved)
2007-11-22 04:15:22 +00:00
91e672d428
some progress on the overall statistics, though probably not visible to end users yet
2007-11-21 20:03:53 +00:00
260033847f
fix ticket #451 (Files remain in use after removal)
2007-11-21 16:16:59 +00:00
a7671a3afc
update clients list to identify test builds
2007-11-21 15:20:07 +00:00
9f9901b3fa
tweak the peer limits a bit for faster uploading
2007-11-20 17:29:56 +00:00
172efd42fa
fix "manual update" crash reported by Gimp_ and BentMyWookie
2007-11-20 16:50:48 +00:00
0a61e04757
now that libevent has a stable 1.4 release out with the features & fixes we wanted, aim third-party/libevent/ at libevent's 1.4 maintainance branch instead of their trunk.
2007-11-20 03:11:50 +00:00
9cec81b524
fix the crash reported by pea_, Gimp, and John_Clay [ref: http://pastebin.ca/784834 ]
2007-11-19 21:44:38 +00:00
9f0ca608f6
fix r3892 oops
2007-11-19 21:24:53 +00:00
b61dcfdbff
add "this session" stats stubs for BentMyWookie
2007-11-19 20:07:21 +00:00
f6371c99a3
add "time running" to tr_global_stats
2007-11-19 19:30:24 +00:00
4cfa1ca6b5
added tr_global_stats prototypes for BentMyWookie
2007-11-19 17:29:17 +00:00
41b1f76d9b
remove more dead code. this scaffolding was to let libevent play nice back when we had one thread per torrent
2007-11-19 16:30:31 +00:00
3929b37da3
denis is right about r3880; we should respect umask
2007-11-19 12:59:26 +00:00
41cee06b66
oops, this was supposed to be in 3879
2007-11-19 05:52:05 +00:00
3704854654
add more asserts up the call chain to find the source of pea_ and Gimp's bug
2007-11-19 04:51:06 +00:00
a654f808e3
change TR_MKDIR_PERM from 0777 to 0755
...
change TR_CREAT_PERM from 0666 to 0644
2007-11-19 04:46:49 +00:00
aa11440755
*maybe* fix the crash reported by pea_ in the forums and by Gimp in http://pastebin.ca/777952
2007-11-19 04:44:14 +00:00
b0f5ff96f3
back to the "router death" issue: throttle how many connections T will try to open at any one time
2007-11-18 17:35:28 +00:00
ed72b1b795
fix choking bug that could penalize good peers.
2007-11-18 06:15:13 +00:00
a277382578
more download speed tweaks for situations where T is both uploading and downloading.
2007-11-18 04:24:17 +00:00
e371009606
this should improve download speeds. test it!
2007-11-18 03:18:26 +00:00
9a81cd7637
reintroduce the "SWIFT" algorithm
2007-11-18 01:00:49 +00:00
37969406b2
Add "Status" column to tr_peer_stat.
2007-11-17 23:43:33 +00:00
54a2bb978b
fix timeout error that caused us to drop connections prematurely
2007-11-17 17:49:30 +00:00
a2609ae2a7
fix permissions issue reported by Jaybird in http://transmission.m0k.org/forum/viewtopic.php?t=2944
2007-11-17 16:20:08 +00:00
ef3d0df0d9
try to tickle some more information out of John_Clay's bug report
2007-11-17 07:48:51 +00:00
77d58126f4
remove yet more dead code
2007-11-17 02:47:47 +00:00
135da6efca
tweaking the speed limit settings a little more
2007-11-17 00:30:36 +00:00
8d08db716d
speed limit code cleanup
2007-11-17 00:16:26 +00:00
cc728b0470
progress on the "speed limits kill my transfer rate" bug.
2007-11-16 21:58:11 +00:00
521d32af9d
tweak the libevent read buffer size
2007-11-16 21:47:55 +00:00
a1805875ee
try to fix the platform.c:278 issue reported by dave in http://transmission.m0k.org/forum/viewtopic.php?t=2949
2007-11-16 20:41:41 +00:00
0ac302b302
progress on the "speed limits kill my transfer rate" bug.
2007-11-16 20:40:03 +00:00
d2a5dd50d3
remove dead code
2007-11-16 20:38:51 +00:00
73680b6263
fix tr_peerMsgsAddRequest memory leak reported by tiennou
2007-11-16 15:45:26 +00:00
1ebaf597a4
fix trunk memory leak reported by tiennou
2007-11-16 15:33:57 +00:00
787d2b800c
fix benocde crash reported by Gimp_
2007-11-16 06:37:01 +00:00
1464dc7962
fix r3830 bug reported by Gimp_
2007-11-16 05:43:47 +00:00
c3dce15a9e
if we haven't advertised that we support pex, then even if a peer sends us a pex message, don't send one back.
2007-11-16 04:42:51 +00:00
b6b339e25a
fix bencode crash reported earlier today by Gimp_, and possibly also by tiennou
2007-11-16 03:32:42 +00:00
807eff8416
remove dead code.
2007-11-15 22:03:22 +00:00
a048204380
remove dead code.
2007-11-15 16:43:46 +00:00
815fb89b91
fix torrent metafile creation error introduced in trunk in r3735 and reported by tiennou in ticket #478
2007-11-15 15:07:05 +00:00
cc42c8aaf6
fix tracker redirect bug reported by sunny
2007-11-15 14:34:48 +00:00
257eaee631
more of the same
2007-11-15 05:47:23 +00:00
ea3a1387f2
more of the same.
2007-11-15 05:20:38 +00:00
8019f8cd2c
raise the number of unchoked peers per torrent
2007-11-15 05:04:21 +00:00
a14fe99e00
another download speed tweak
2007-11-15 04:59:37 +00:00
c4f15263a0
since many people seem to be having a hard time holding onto good peers, be a little more lenient on how quickly we hang up on peers that have transferred piece data with us.
2007-11-14 05:02:03 +00:00
a9e5fff2e6
* fix the `req->length == (uint32_t)( ((block)==((msgs->torrent)->blockCount-1))' bug.
...
* there seems to be a pattern for peers that were (intentionally?) giving incomplete data to trigger the bug above. when a peer does this, give them a strike on its three-strikes-and-you're-banned count
2007-11-13 05:36:43 +00:00
6242508fe2
minor tweaks.
2007-11-12 22:26:10 +00:00
66c9119a84
tiennou's fix for #461
2007-11-12 15:15:51 +00:00
dfba59b97a
tiennou's fix for #463
2007-11-12 15:08:58 +00:00
f1e85f5206
Increase the intervals on timers that won't affect performance. This reduces most of Transmission's PowerTOP footprint.
2007-11-11 20:27:13 +00:00
8516860fec
fix a bug in tr_stat's availability numbers, reported by BentMyWookie
2007-11-11 16:33:04 +00:00
d89d6faad4
don't require clients to call tr_torrentClearStatusCallback() before tr_torrentClose().
2007-11-11 03:14:01 +00:00
adeee3d886
silence some Fast Peer console messages
2007-11-10 22:29:55 +00:00
0aee22aaff
have tr_close() call tr_torrentClose() on all running torrents
2007-11-10 22:29:28 +00:00
c32f893038
commit tiennou's fast-peers-final patch from ticket 149
2007-11-10 16:06:00 +00:00
4cb16901b8
fix tracker wart reported by tiennou
2007-11-10 16:00:59 +00:00
76626796f3
make the hard drive less noisy than it was in 0.92
2007-11-10 05:09:40 +00:00
e7e479d758
fix r3777 compiler warning reported by SoftwareElves
2007-11-10 05:01:13 +00:00
ca03b62a58
* fix a bug that could corrupt peer connections, causing slower speeds and/or fewer connections
...
* add more assertions to smoke out any remaining BT protocol bugs
2007-11-10 04:56:27 +00:00
fd430fde90
undoing the r3773-r3774 experiment.
2007-11-09 20:07:52 +00:00
fda79c7eef
oops, wrong #include for NSCParameterAssert in r3773.
2007-11-09 19:51:48 +00:00
67b719c2de
on Darwin, use NSCParameterAssert() instead of assert().
2007-11-09 19:50:40 +00:00
4792788d24
move the gateway discovery from init() to start() so that tr_init() doesn't block.
2007-11-09 16:10:48 +00:00
e168cf3f69
don't let incoming peer connections sneak past MAX_CONNECTED_PEERS_PER_TORRENT.
2007-11-09 15:19:12 +00:00
b382628933
have libtransmission be in charge of checking if nat traversal is disabled when quitting
2007-11-09 15:06:32 +00:00
5da8a977c6
* add O_LARGEFILE support to fdlimit for the benefit of 32-bit posix machines
...
* minor tweaks
2007-11-09 04:32:19 +00:00
42e70ad53e
fix John_Clay's assertion bugreport
2007-11-09 01:25:34 +00:00
99ea26c1c9
* try to get peers to scale up to speed faster.
...
* remove unnecessary steps when receiving block data from peers.
2007-11-09 01:22:15 +00:00
39f9f14751
minor tweaks.
2007-11-08 21:25:17 +00:00
125bd34bbb
added `plaintext preferred' encryption option.
2007-11-08 21:20:08 +00:00
0d819bd67e
more work on the "router death" issue.
2007-11-08 19:43:48 +00:00
bc3eff899f
* some work on lessening the way we hammer the router.
...
* move strlcpy and strlcat out of their own files and into utils.c
2007-11-08 04:11:09 +00:00
71f4efcbad
update our #includes now that libevent has cleaned up event.h
2007-11-07 18:26:19 +00:00
b43bb1483d
fix libevent #include quirk reported by SoftwareElves
2007-11-07 03:13:04 +00:00
906c5d6f1f
dead code removal: azId
2007-11-07 01:27:32 +00:00
dc6a7b9485
fix crash reported by hussam in #transmission
2007-11-07 00:42:19 +00:00
229102a196
more libevent fun. good thing the mac build is broken or I'd've been getting bug reports on this instead :)
2007-11-06 23:23:47 +00:00
476de56e45
the libevent header file comments LIED about the return value of evbuffer_add_vprintf()!
2007-11-06 23:10:04 +00:00
60aab6c433
remove libtransmission code that duplicates functionality in libevent
2007-11-06 21:14:30 +00:00
9df263326d
use Thomas Bernard's miniupnp library instead of rolling our own.
2007-11-06 16:02:50 +00:00
5e2b2d34c3
ensure that tr_close() exits gracefully even when the trackers aren't responding.
2007-11-05 05:42:25 +00:00
14d8abb323
maybe fix portmapping
2007-11-05 03:56:54 +00:00
0f92cb7f30
fix bug reported in the forums by "grabman"
2007-11-05 03:09:27 +00:00
7ce8312ddc
fix memory error reported by John_Clay
2007-11-02 20:27:03 +00:00
e3764e732b
fix possible benc leak reported by MishaS ( #441 )
2007-11-01 18:44:37 +00:00
7cb1791f2a
fix handshake memory leak reported by MishaS ( #442 )
2007-11-01 18:41:13 +00:00
3e0e462b10
finish cleaning up the code that was churned during the "corruption" fix.
2007-11-01 13:47:32 +00:00
e625b10d6d
"corruption" fix part 3: cleanup. remove unneeded possible points of failure.
2007-10-31 18:10:55 +00:00
a6206b036c
"corruption" fix part 2: lots of assertions, and try to complete pieces & ban bad peers sooner
2007-10-31 04:23:51 +00:00
Charles Kerr