Date: Wed, 31 May 2006 10:29:16 -0300 From: Alexandre Biancalana <ale@seudns.net> To: freebsd-net@FreeBSD.ORG Subject: requests for mbufs denied increasing Message-ID: <447D9A2C.7010407@seudns.net>
next in thread | raw e-mail | index | archive | help
Hi list,
I've a NFS/FTP server used as backup server, receiving few
connections with high volume of data (100GB/Day). During this period the
"requests for mbufs denied (mbufs/clusters/mbuf+clusters)" of netstat -m
is increasing constantly, now is:
151/404/555 mbufs in use (current/cache/total)
65/141/206/65456 mbuf clusters in use (current/cache/total/max)
65/79 mbuf+clusters out of packet secondary zone in use (current/cache)
0/0/0/0 4k (page size) jumbo clusters in use (current/cache/total/max)
0/0/0/0 9k jumbo clusters in use (current/cache/total/max)
0/0/0/0 16k jumbo clusters in use (current/cache/total/max)
167K/383K/550K bytes allocated to network (current/cache/total)
9053239/2872380/2872048 requests for mbufs denied
(mbufs/clusters/mbuf+clusters)
0/0/0 requests for jumbo clusters denied (4k/9k/16k)
0/5/4432 sfbufs in use (current/peak/max)
0 requests for sfbufs denied
0 requests for sfbufs delayed
0 requests for I/O initiated by sendfile
2220 calls to protocol drain routines
I tried to increase the kern.ipc.nmbclusters from 16192 to 32728 and
65456, without success. Before increase this I'he to boot the machine
because this stopped to accept new network connections and a simple ping
returned "sendto: *No buffer* space available".
The machine is running 6.1-STABLE updated yesterday, is a P4 3Ghz HT
enabled, 512MB Ram, 4 IDE 200GB discs and 4 SCSI 73GB, all of then are
striped with geom_stripe, building an 1 TB volume used to store all
backup data.
top output:
last pid: 3350; load averages: 0.96, 0.87,
0.78 up 0+00:50:25 10:21:39
60 processes: 2 running, 58 sleeping
CPU states: 27.3% user, 0.0% nice, 7.1% system, 0.4% interrupt, 65.2%
idle
Mem: 151M Active, 147M Inact, 155M Wired, 27M Cache, 59M Buf, 968K Free
Swap: 512M Total, 224K Used, 512M Free, 4K In
vmstat 5 output:
procs memory page disks faults cpu
r b w avm fre flt re pi po fr sr ad0 ad1 in sy cs us
sy id
1 4 1 458640 26264 136 8 1 0 3073 2879 0 0 8045 7062 19992
19 11 71
1 4 0 458668 21048 2 23 0 0 3680 3438 34 42 1153 3082 4492
40 6 54
0 5 0 458696 18648 2 4 0 1 3551 3414 34 37 1144 3524 5274
33 5 62
1 4 0 458724 23492 64 5 5 0 3266 3426 60 51 1253 2125 4234
28 5 67
Best Regards,
Alexandre
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?447D9A2C.7010407>