kernel leaking memory somewhere
elekktretterr at exemail.com.au
elekktretterr at exemail.com.au
Wed Dec 16 13:33:20 PST 2009
Hi guys,
This is TOP output on FBSD 7 box that runs the same software as the DF
box. Actually the FBSD box also runs MySQL and Ruby on Rails!
last pid: 60325; load averages: 0.00, 0.00, 0.00
up 57+13:36:12 08:11:43
79 processes: 1 running, 78 sleeping
CPU: 0.0% user, 0.0% nice, 0.9% system, 0.0% interrupt, 99.1% idle
Mem: 294M Active, 1906M Inact, 401M Wired, 30M Cache, 399M Buf, 1314M Free
Swap: 4096M Total, 16K Used, 4096M Free
PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
669 root 1 44 0 119M 99M select 1 382:26 0.00% snmpd
844 nobody 1 4 0 7796K 2260K kqread 0 1:59 0.00% memcached
941 root 1 44 0 22880K 2840K select 3 1:44 0.00% sshd
543 root 1 44 0 5692K 1172K select 2 0:26 0.00% syslogd
817 root 1 4 0 5808K 1420K kqread 0 0:16 0.00% master
85668 pgsql 1 44 0 1027M 52340K select 2 0:16 0.00% postgres
948 root 1 8 0 6748K 1200K nanslp 1 0:15 0.00% cron
85671 pgsql 1 44 0 22576K 5784K select 1 0:13 0.00% postgres
825 postfix 1 4 0 5808K 1544K kqread 2 0:11 0.00% qmgr
85669 pgsql 1 44 0 1027M 6972K select 2 0:09 0.00% postgres
85666 pgsql 1 44 0 1027M 5892K select 2 0:08 0.00% postgres
85670 pgsql 1 44 0 1028M 6132K select 2 0:07 0.00% postgres
42312 www 1 20 0 143M 19380K lockf 0 0:02 0.00% httpd
44595 www 1 20 0 143M 19320K lockf 1 0:02 0.00% httpd
42299 root 1 44 0 141M 16304K select 0 0:02 0.00% httpd
899 mysql 9 20 0 61608K 26156K sigwai 2 0:01 0.00% mysqld
612 bind 7 4 0 21420K 14620K kqread 1 0:01 0.00% named
42914 www 1 20 0 143M 19236K lockf 0 0:01 0.00% httpd
42313 www 1 20 0 143M 19240K lockf 1 0:01 0.00% httpd
45188 www 1 20 0 143M 19236K lockf 2 0:01 0.00% httpd
44007 www 1 20 0 143M 19172K lockf 0 0:00 0.00% httpd
41418 sysadmin 1 44 0 33772K 3836K select 2 0:00 0.00% sshd
45711 www 1 20 0 141M 16800K lockf 0 0:00 0.00% httpd
41426 root 1 5 0 10108K 2588K ttyin 0 0:00 0.00% csh
45712 www 1 4 0 141M 16824K kqread 0 0:00 0.00% httpd
48958 www 1 20 0 141M 16808K lockf 3 0:00 0.00% httpd
45966 www 1 20 0 141M 16784K lockf 1 0:00 0.00% httpd
60230 root 1 4 0 33772K 3876K sbwait 0 0:00 0.00% sshd
41415 root 1 4 0 33772K 3848K sbwait 2 0:00 0.00% sshd
58813 postfix 1 4 0 5808K 1444K kqread 2 0:00 0.00% pickup
60237 sysadmin 1 20 0 10108K 2448K pause 2 0:00 0.00% tcsh
832 mysql 1 8 0 7064K 1216K wait 2 0:00 0.00% sh
41420 sysadmin 1 20 0 10108K 2604K pause 2 0:00 0.00% tcsh
60242 root 1 20 0 10108K 2388K pause 2 0:00 0.00% csh
60235 sysadmin 1 44 0 33772K 3864K select 1 0:00 0.00% sshd
993 root 1 5 0 5688K 1016K ttyin 1 0:00 0.00% getty
996 root 1 5 0 5688K 1016K ttyin 0 0:00 0.00% getty
60241 sysadmin 1 8 0 20440K 1588K wait 0 0:00 0.00% su
41425 sysadmin 1 8 0 20440K 1588K wait 0 0:00 0.00% su
As you can see the active memory here is 294MB.
While on DragonFly:
load averages: 0.00, 0.02, 0.00; up 0+20:31:39
08:13:24
49 processes: 49 running
CPU states: 0.0% user, 0.0% nice, 0.6% system, 0.0% interrupt, 99.4% idle
Memory: 1213M Active, 1836M Inact, 320M Wired, 25M Cache, 199M Buf, 114M Free
Swap: 4096M Total, 2692K Used, 4093M Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND
1756 pgsql 152 0 833M 13M select 1 0:09 0.00% 0.00% postgres
1757 pgsql 152 0 833M 8324K select 0 0:08 0.00% 0.00% postgres
1759 pgsql 152 0 8536K 1252K poll 3 0:04 0.00% 0.00% postgres
1758 pgsql 152 0 833M 1724K select 3 0:02 0.00% 0.00% postgres
712 root 152 0 7880K 712K select 0 0:02 0.00% 0.00% httpd
677 root 152 0 5484K 1600K select 2 0:01 0.00% 0.00% sendmail
1754 pgsql 153 0 832M 3204K select 0 0:00 0.00% 0.00% postgres
578 root 152 0 2276K 680K nanslp 1 0:00 0.00% 0.00% dntpd
897 sysadmin 152 0 7800K 888K select 3 0:00 0.00% 0.00% sshd
1638 root 152 0 928K 4K nfsd 3 0:00 0.00% 0.00% nfsd
464 root 152 0 2864K 516K poll 1 0:00 0.00% 0.00% rpcbind
903 root 152 0 3968K 660K ttyin 0 0:00 0.00% 0.00% csh
663 root 152 0 4984K 836K select 3 0:00 0.00% 0.00% sshd
331 root 152 0 1756K 608K select 3 0:00 0.00% 0.00% syslogd
696 root 152 0 2556K 820K nanslp 3 0:00 0.00% 0.00% cron
895 root 157 0 7928K 788K sbwait 2 0:00 0.00% 0.00% sshd
6325 root 152 0 7928K 2712K sbwait 0 0:00 0.00% 0.00% sshd
1417 root 152 0 2396K 4K select 0 0:00 0.00% 0.00% mountd
6350 root 152 0 3644K 1544K CPU1 1 0:00 0.00% 0.00% top
1637 root 152 0 1720K 40K select 3 0:00 0.00% 0.00% nfsd
6333 sysadmin 154 0 2904K 1380K wait 1 0:00 0.00% 0.00% su
6334 root 153 0 3484K 1296K pause 0 0:00 0.00% 0.00% csh
801 root 153 0 2936K 576K wait 3 0:00 0.00% 0.00% login
822 root 152 0 3740K 284K ttyin 2 0:00 0.00% 0.00% csh
898 sysadmin 165 0 3612K 284K pause 3 0:00 0.00% 0.00% tcsh
804 root 159 0 1804K 464K ttyin 2 0:00 0.00% 0.00% getty
6328 sysadmin 153 0 3612K 1308K pause 0 0:00 0.00% 0.00% tcsh
681 smmsp 153 0 5356K 916K pause 2 0:00 0.00% 0.00% sendmail
6327 sysadmin 152 0 7800K 2000K select 2 0:00 0.00% 0.00% sshd
902 sysadmin 156 0 2904K 576K wait 2 0:00 0.00% 0.00% su
The difference is clear. The 2 servers have same amount of physical memory.
Im gonna try the memory program, but Im telling you all I really need to
do is start doing some heavy IO and get those postgres processes going
(they usually only use about 30MB RES memory) and the box starts swapping.
Petr
More information about the Kernel
mailing list