Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 26 Dec 2002 06:07:32 -0800
From:      "Mexx" <mexx@telus.net>
To:        <freebsd-questions@FreeBSD.ORG>
Subject:   All mbuf clusters exhausted, please see tuning(7)
Message-ID:  <000501c2ace8$227b6ee0$7700b38e@bc.hsia.telus.net>

next in thread | raw e-mail | index | archive | help
sysctl kern.ipc.somaxconn
kern.ipc.somaxconn: 8192

sysctl kern.ipc.nmbclusters
kern.ipc.nmbclusters: 65535

sysctl kern.maxfiles
kern.maxfiles: 65536

sysctl kern.maxfilesperproc
kern.maxfilesperproc: 32768

How much higher can I go with dual 1.13 Ghz + 2 gigs of ram?

That's that main problem that causes the server to crash and require
powercycle to come back up. It would only stay up for a few seconds on high
traffic.

mbuf clusters have been increased to 65535

131070 was attempted but server did not like it and won't boot up.

Server has 2 gigs of ram

FreeBSD 4.6-RELEASE

fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0x1400-0x143f mem
0xfeaa0000-0xfeabffff,0xfeae0000-0xfeae0fff irq 7 at device 3.0 on pc
i0
fxp0: Ethernet address 00:03:47:a4:fb:23
inphy0: <i82555 10/100 media interface> on miibus0
inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
fxp1: <Intel Pro 10/100B/100+ Ethernet> port 0x1440-0x147f mem
0xfea60000-0xfea7ffff,0xfea80000-0xfea80fff irq 9 at device 4.0 on pc
i0
fxp1: Ethernet address 00:03:47:a4:fb:24
inphy1: <i82555 10/100 media interface> on miibus1
inphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto

netstat -in
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
fxp0 1500 <Link#1> 00:03:47:a4:fb:23 7943967 2 11631732 1 0
fxp0 1500 66.28.252/xx 66.28.252.4 1175 - 11638120 - -
fxp0 1500 66.28.252.xx/ 66.28.252.17 1346 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.18 90206 - 0 - -
fxp0 1500 66.28.252.x/x 66.28.252.7 2696087 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.11 2 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.13 13 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.25 13 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.28 5152357 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.29 13 - 0 - -
fxp0 1500 66.28.252.xx/ 66.28.252.30 56 - 0 - -
fxp1* 1500 <Link#2> 00:03:47:a4:fb:24 0 0 0 0 0
lo0 16384 <Link#3> 148 0 148 0 0
lo0 16384 127 127.0.0.1 148 - 148 - -

Load and Cpu is good but crashes on Mbuff error. I can't increase it anymore
cause it would not bootup with new changes =/ What could I do? Could it be
bad motherboard or harddrive?


last pid: 694; load averages: 0.08, 0.04, 0.01 up 0+00:02:24 09:39:09
483 processes: 1 running, 482 sleeping
CPU states: 1.2% user, 0.0% nice, 9.1% system, 3.3% interrupt, 86.4% idle
Mem: 139M Active, 62M Inact, 205M Wired, 37M Cache, 18M Buf, 1568M Free
Swap: 2048M Total, 2048M Free
Dec 7 09:39:08 redapple /kernel: All mbuf clusters exhausted, please see
tuning(7).
Dec 7 09:39:08 redapple /kernel: All mbuf clusters exhausted, please see
tuning(7).
411 www -18 0 119M 57764K mclalc 1 0:03 6.83% 5.81% thttpd
445 www -18 0 50964K 32620K mclalc 0 0:01 2.60% 2.15% thttpd
443 www -18 0 56276K 37576K mclalc 0 0:01 2.54% 2.10% thttpd
447 www -18 0 64532K 10292K mclalc 1 0:01 1.19% 0.98% thttpd
693 root 28 0 2484K 1712K CPU1 0 0:00 1.85% 0.78% top
108 root 2 0 976K 684K select 1 0:02 0.26% 0.24% syslogd
183 root 2 0 4428K 3672K select 0 0:00 0.00% 0.00% httpd
449 www -18 0 25212K 5652K mclalc 1 0:00 0.00% 0.00% thttpd
451 www -18 0 19460K 6584K mclalc 1 0:00 0.00% 0.00% thttpd
123 root 2 0 2232K 1676K select 1 0:00 0.00% 0.00% sshd
635 root 18 0 1372K 956K pause 0 0:00 0.00% 0.00% csh
111 bind 2 0 2380K 2008K select 0 0:00 0.00% 0.00% named
558 root 2 0 2316K 1936K select 0 0:00 0.00% 0.00% sshd
362 www 2 0 4508K 3752K sbwait 0 0:00 0.00% 0.00% httpd
207 mysql 2 0 26960K 4724K poll 1 0:00 0.00% 0.00% mysqld
378 www 2 0 4508K 3752K sbwait 0 0:00 0.00% 0.00% mysqld
493 www 2 0 4520K 3752K sbwait 1 0:00 0.00% 0.00% httpd
487 www 2 0 4520K 3752K sbwait 0 0:00 0.00% 0.00% httpd
227 www 2 0 4508K 3752K sbwait 0 0:00 0.00% 0.00% httpd
258 www 2 0 4508K 3752K accept 1 0:00 0.00% 0.00% httpd
378 www 2 0 4508K 3752K sbwait 1 0:00 0.00% 0.00% httpd

Could an attack cause my server to run out of mbufs? Thanks



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?000501c2ace8$227b6ee0$7700b38e>