Known memory leak in 6-STABLE from April 1st?
Ulrich Spoerlein
uspoerlein at gmail.com
Tue May 15 08:35:47 UTC 2007
On 5/14/07, Marc G. Fournier <freebsd at hub.org> wrote:
> > Now after doing some heavy IMAP testing (cyrus reconstruct of big
> > maildirs) the system froze to a complete halt. Stupid me already
> > rebooted the machine, tomorrow I'll try to break into DDB when it
> > happens again. I also started recording top(1) memory output and
> > sysctl vm.zone output.
> >
> > The main questions is: Were there any known memory leaks at the start
> > of April? Any patches I should blindly try before spending several
> > days on debugging this?
>
> Hrmmm ... long shot here, but what does:
>
> sysctl kern.ipc.numopensockets
>
> show over that period of time ... just wondering if we are somehow related on
> problems here, just different symptoms ...
Sorry no, nothing suspicous there. It bounces up and down, after
killing all amavis, cyrus and postfix processes it came down to about
80. Right now it's at 280 again, and the memory is so small, that I
can no longer grep(1) a 600MB file or do other useful stuff.
This is the last vm.zone output, anything suspicous? What commands
should I run (in DDB?) to see where the memory is going?
Tue May 15 10:33:38 CEST 2007
vm.zone:
ITEM SIZE LIMIT USED FREE REQUESTS
FFS2 dinode: 256, 0, 83286, 10179, 1273390
FFS1 dinode: 128, 0, 0, 0, 0
FFS inode: 132, 0, 83286, 10761, 1273390
Mountpoints: 664, 0, 7, 11, 8
SWAPMETA: 276, 121576, 11, 17, 22
pfosfp: 28, 0, 0, 0, 0
pfospfen: 108, 0, 0, 0, 0
pfiaddrpl: 92, 0, 0, 0, 0
pfstatescrub: 28, 0, 0, 0, 0
pffrcent: 12, 50141, 0, 0, 0
pffrcache: 48, 10062, 0, 0, 0
pffrag: 48, 0, 0, 0, 0
pffrent: 16, 5075, 0, 0, 0
pfrkentry2: 156, 0, 0, 0, 0
pfrkentry: 156, 0, 0, 0, 0
pfrktable: 1240, 0, 0, 0, 0
pfpooladdrpl: 68, 0, 0, 0, 0
pfaltqpl: 128, 0, 0, 0, 0
pfstatepl: 260, 10005, 0, 0, 0
pfrulepl: 604, 0, 0, 0, 0
pfsrctrpl: 100, 0, 0, 0, 0
rtentry: 132, 0, 37, 108, 3848
ripcb: 180, 25608, 0, 110, 33
sackhole: 20, 0, 0, 676, 3672
tcpreass: 20, 1690, 0, 676, 1365
hostcache: 76, 15400, 5, 245, 317
syncache: 100, 15366, 0, 195, 27209
tcptw: 48, 5148, 0, 624, 34253
tcpcb: 464, 25600, 29, 203, 212261
inpcb: 180, 25608, 29, 389, 212261
udpcb: 180, 25608, 19, 179, 493092
ipq: 32, 904, 0, 0, 0
unpcb: 144, 25623, 213, 3405, 300681
socket: 356, 25608, 263, 5567, 1006076
KNOTE: 68, 0, 0, 616, 724767
PIPE: 408, 0, 10, 818, 1544936
DIRHASH: 1024, 0, 861, 247, 4143
NFSNODE: 460, 0, 1, 23, 7
NFSMOUNT: 480, 0, 1, 15, 2
L VFS Cache: 291, 0, 402, 326, 6124
S VFS Cache: 68, 0, 87410, 16638, 1831436
NAMEI: 1024, 0, 0, 660, 105375628
VNODEPOLL: 76, 0, 0, 250, 8
VNODE: 272, 0, 83341, 12405, 1273709
ata_composit: 196, 0, 0, 0, 0
ata_request: 204, 0, 0, 76, 34
g_bio: 132, 0, 0, 696, 16904466
ACL UMA zone: 388, 0, 0, 0, 0
mbuf_jumbo_1: 16384, 0, 0, 0, 0
mbuf_jumbo_9: 9216, 0, 0, 0, 0
mbuf_jumbo_p: 4096, 0, 0, 0, 0
mbuf_cluster: 2048, 25600, 929, 325, 1276052
mbuf: 256, 0, 931, 554, 141685396
mbuf_packet: 256, 0, 806, 679, 104799211
VMSPACE: 296, 0, 127, 328, 1617479
UPCALL: 44, 0, 5, 229, 10
KSEGRP: 88, 0, 386, 174, 556
THREAD: 376, 0, 394, 286, 292710
PROC: 536, 0, 170, 215, 1617545
Files: 72, 0, 580, 4296, 40984983
4096: 4096, 0, 229, 542, 3118296
2048: 2048, 0, 229, 547, 4814216
1024: 1024, 0, 347, 389, 9329214
512: 512, 0, 186, 398, 1319246
256: 256, 0, 919, 3941, 5484210
128: 128, 0, 2516, 6034, 28789911
64: 64, 0, 5339, 7110, 21344537
32: 32, 0, 2687, 7257, 10126384
16: 16, 0, 2861, 996, 9327126
mt_zone: 1024, 0, 189, 3, 189
DP fakepg: 72, 0, 0, 0, 0
PV ENTRY: 24, 2067410, 126063, 334892, 1426534413
MAP ENTRY: 68, 0, 8613, 7795, 104088221
KMAP ENTRY: 68, 57344, 1001, 1127, 208001
MAP: 192, 0, 7, 13, 7
VM OBJECT: 132, 0, 82483, 14058, 43364599
128 Bucket: 524, 0, 2145, 865, 8052
64 Bucket: 268, 0, 88, 80, 274
32 Bucket: 140, 0, 69, 99, 180
16 Bucket: 76, 0, 42, 58, 118
UMA Hash: 128, 0, 3, 27, 6
UMA RCntSlab: 104, 0, 627, 2, 7918
UMA Slabs: 64, 0, 1933, 427, 31248
UMA Zones: 480, 0, 83, 5, 83
UMA Kegs: 140, 0, 83, 13, 83
Uli
More information about the freebsd-stable
mailing list