2b715b89f1
Debug code sometimes copied too much data
2007-10-21 03:37:26 +00:00
ddd85ec7e0
Fixed one leak from struct_http
2007-10-21 01:11:45 +00:00
8098adb48e
Be even more stupidistas-tolerant
2007-10-19 23:15:13 +00:00
84f8738b82
Fix debug output
2007-10-19 22:55:13 +00:00
d768bf6e81
Enable _DEBUG_HTTPERROR to avoid warning...
2007-10-19 22:36:28 +00:00
53bce09735
Include time since startup in full scrape debug output
2007-10-19 22:20:42 +00:00
c59e44ba94
Now limit full scrape to requests sending exactly "GET /scrape HTTP/"
2007-10-19 22:13:59 +00:00
41019c79d8
Introducing multiscrape
2007-10-19 21:56:59 +00:00
d5876ebb6b
Prevent url-decoding for debug by using a temp buffer
2007-10-19 21:00:57 +00:00
271ff156ec
Those stupid "announce ?info_hash" torrents get really annoying. Fix it in announce and scrape.
2007-10-19 20:02:41 +00:00
1b0859411e
be more verbose about what full scrape consumers deliver in their http requests
2007-10-19 15:37:00 +00:00
daf79f694d
After studying opentracker live data, I decided to radically reduce default allocation for vectors and making it grow faster instead: Most pools only had one or two peers, wasting 8*15 or 8*14 bytes.
2007-10-19 14:41:30 +00:00
870c995fbb
Remove an unused variable
2007-10-19 05:23:04 +00:00
cab763d4ee
Release exactly all unused bytes in mmap allocation.
2007-10-19 05:21:01 +00:00
714ea1b88e
give opentracker the missing header file
2007-10-19 03:41:23 +00:00
f56e40c24e
Must not dealloc mmaped buffers with free()
2007-10-19 03:39:04 +00:00
d25cf9307e
Fix misunderstandig of what libowfat thinks are seconds and what I did...
2007-10-19 02:00:53 +00:00
b8b1493127
Increase check interval for timeouted sockets to 10 seconds. That may prevent some disrupted scrapes.
2007-10-19 01:27:23 +00:00
e1f122f7d1
Since array and io_batch are in a union, and we also have time-outing writable sockets, we need to mark, which of the union is used. Also remove the blessed flag from struct http, since its rarely used and can be done in a seperate memcmp.
2007-10-19 01:26:33 +00:00
6eb441d8a0
Cannot access members after they are freed :/
2007-10-19 00:34:56 +00:00
b86e6382a1
Save a lot of work when skipping through uninteresting http request parameters
2007-10-18 23:33:07 +00:00
dba3bb3ae7
Make fromhex() even faster
2007-10-18 11:50:54 +00:00
243d5961d0
fixed one performance bug, where "skipping values from a ¶m=values pair" was requested, the requestor ended up with "values" to be parsed again.
...
improved performance of fromhex
improved performance of is_unreserved() by moving it all into a simple byte array
improved performance of %41 => 'A' conversion by reordering variables
2007-10-18 00:21:22 +00:00
be117f96a2
Increase send timeout. Also use an own variable for that.
2007-10-17 23:41:37 +00:00
32f17413fa
Increase send timeout. Also use an own variable for that.
2007-10-17 22:50:48 +00:00
28de135098
io_timeout is absolute, not relative :/
2007-10-17 21:13:13 +00:00
d1c57d90dd
Free iobatches on timeout
2007-10-17 20:25:18 +00:00
8e78ac7012
HEADS UP: you need to check out the latest libowfat! Huge allocations are now taken from mmap()ed regions, not from heap anymore
2007-10-17 14:43:14 +00:00
1e6b48cbc0
Sockets we write to now do also timeout
2007-10-16 18:23:36 +00:00
c730a24d0a
Fixed 24h clean bug. Allocate less memory for full scrape. Free torrents without any downloaders immediately.
2007-10-16 00:33:31 +00:00
28d88e3380
Exclude empty torrents from full scrape, also try to shrink full scrape buffer after it is full
2007-10-16 00:23:14 +00:00
3dc470d672
fixed line breaks in log
2007-10-15 18:03:13 +00:00
65cc45f129
Log all expansive operations to stderr
2007-10-15 18:01:38 +00:00
ffffd80719
Most open trackers do not have the problem of not being called often enough.. so do not introduce an extra timeout when waiting for activity. That also removes one of the last extraneous gettimeofday calls
2007-10-13 17:58:20 +00:00
6c1adb8fc8
since gettimeofday is rather expansive, we do only fetch time once in a 5 second period, when we are delivered a SIGALRM.
2007-10-13 17:40:37 +00:00
4c15dc7a70
make 64bit safe
2007-10-12 21:57:12 +00:00
dd82db18ad
There is a pool leak somewhere... try to find it
2007-10-06 19:17:21 +00:00
0d36a8754e
fixed some stat output
2007-10-03 22:57:35 +00:00
25e59aeca3
Two more bugs in Makefile. Maybe I learn it one day
2007-10-01 00:31:59 +00:00
d0b5c463d9
Revert debug option in Makefile
2007-10-01 00:29:46 +00:00
fb5a228162
Makefile now runs under bsdmake and gnumake again
2007-10-01 00:27:26 +00:00
c350fa0b3c
New stats for s24s code, this is debug, do not use in real world
2007-09-26 16:49:13 +00:00
c4f924810e
fixed a small bug with 0 torrents and a cosmetical one
2007-09-21 22:24:22 +00:00
6ad00cadc5
Makefile defaults to an _open_ tracker again.
2007-09-10 02:43:30 +00:00
33c9c530d0
An announce with event=stopped now returns correct number of leechers and seeders. In TCP and UDP.
2007-09-10 02:43:11 +00:00
ea276fa0bf
Fixed the inverse interpretation of black/white listing
2007-08-18 12:51:13 +00:00
3e47339b6c
Added whitelisting to reimplement the WANT_CLOSED_TRACKER functionality
2007-08-18 09:56:22 +00:00
992058383a
Under Linux one gets SIG_ALRMs from time to time when using epoll. Ignore those.
2007-08-18 09:29:39 +00:00
77af0b475f
Sort of addressed a concorrency issue when receiving mutliple SIG_HUPs
2007-08-11 13:39:36 +00:00
51e4b179ab
off by -1... or so, strange error class
2007-07-22 18:23:00 +00:00
