[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 2.7 high load averages.
- To: misc_(_at_)_openbsd_(_dot_)_org
- Subject: Re: 2.7 high load averages.
- From: Seth Arnold <sarnold_(_at_)_willamette_(_dot_)_edu>
- Date: Thu, 5 Oct 2000 00:29:38 -0700
- Mail-followup-to: misc_(_at_)_openbsd_(_dot_)_org
Rick, I have one thought; you seem to have a lot of processes in 'sleep'
state... the load average shows how many processes are waiting to use
the CPU, and if this many processes are waiting for, say, disk IO, or
network IO, then probably the only thing to bring the load average down
is to have your machine service disk and network IO faster.
I'm not sure I would worry about it if you don't notice performance
problems..
Corrections welcome. :)
* OpenBSD mailing list <obsd_(_at_)_righi_(_dot_)_df_(_dot_)_unibo_(_dot_)_it> [001005 00:22]:
>
> Hello,
> I noticed that load average on my system seems quite high and I think they
> should not.
> For example now ther eis only a few of activity on the system and here are
> the load averages and the list of processes
>
> load averages: 0.73, 0.56, 0.35 09:10:07
> 37 processes: 1 running, 34 idle, 2 stopped
> CPU states: 1.7% user, 0.0% nice, 0.9% system, 0.2% interrupt, 97.2% idle
> Memory: Real: 31M/43M act/tot Free: 16M Swap: 1844K/342M used/tot
>
> PID USERNAME PRI NICE SIZE RES STATE WAIT TIME CPU COMMAND
> 21116 root 2 0 6444K 12M sleep select 2:12 0.05% XF86_SVGA
> 28456 rupert 2 0 16M 11M sleep select 0:14 0.05% netscape
> 6947 root 2 0 252K 572K idle select 0:12 0.00% sshd
> 13276 rj45 2 0 532K 1160K sleep select 0:12 0.00% screen-3.9.8
> 26617 root 2 0 396K 1232K stop - 0:05 0.00% ssh
> 29878 root 2 0 272K 1164K sleep select 0:04 0.00% sshd
> 32497 root 2 0 636K 644K idle select 0:02 0.00% sendmail
> 17070 root -6 0 808K 940K stop piperd 0:01 0.00% cvs
> 18597 root 10 0 224K 368K sleep nanosl 0:01 0.00% cron
> 15917 root 2 0 96K 328K sleep select 0:00 0.00% syslogd
> 8245 rupert 2 0 340K 1384K sleep select 0:00 0.00% fvwm
> 29474 rupert 2 0 732K 2380K sleep select 0:00 0.00% xterm
> 12697 root 2 0 72K 224K idle select 0:00 0.00% inetd
> 17742 root 18 0 884K 716K sleep pause 0:00 0.00% tcsh
> 29162 rj45 18 0 900K 728K idle pause 0:00 0.00% tcsh
> 9930 rj45 18 0 288K 824K sleep pause 0:00 0.00% screen-3.9.8
> 5628 rj45 3 0 900K 728K idle ttyin 0:00 0.00% tcsh
> 15766 root 10 0 312K 1712K idle wait 0:00 0.00% xdm
>
> and yesterday I WAs downloading the sourcetree with cvs. That was the only
> relevant process and here are hte loads
>
> load averages: 1.21, 1.35, 1.16 17:56:06
> 30 processes: 1 running, 27 idle, 2 stopped
> CPU states: 0.5% user, 0.0% nice, 0.5% system, 0.2% interrupt, 98.9% idle
> Memory: Real: 15M/29M act/tot Free: 29M Swap: 1844K/342M used/tot
>
> PID USERNAME PRI NICE SIZE RES STATE WAIT TIME CPU COMMAND
> 21116 root 2 0 6444K 9880K idle select 1:59 0.00% XF86_SVGA
> 21753 root 2 0 408K 1232K sleep select 0:35 0.00% ssh
> 6947 root 2 0 252K 572K idle select 0:11 0.00% sshd
> 13276 rj45 2 0 532K 1160K sleep select 0:08 0.00% screen-3.9.8
> 26617 root 2 0 396K 1232K stop - 0:05 0.00% ssh
> 28844 root -6 0 828K 960K sleep piperd 0:04 0.00% cvs
> 29878 root 2 0 272K 1164K sleep select 0:04 0.00% sshd
> 32497 root 2 0 636K 644K idle select 0:01 0.00% sendmail
> 17070 root -6 0 808K 940K stop piperd 0:01 0.00% cvs
> 18597 root 10 0 224K 368K sleep nanosl 0:01 0.00% cron
> 15917 root 2 0 96K 328K sleep select 0:00 0.00% syslogd
> 12697 root 2 0 72K 224K idle select 0:00 0.00% inetd
> 29162 rj45 18 0 900K 728K idle pause 0:00 0.00% tcsh
> 17742 root 18 0 880K 712K idle pause 0:00 0.00% tcsh
> 5628 rj45 3 0 900K 728K idle ttyin 0:00 0.00% tcsh
> 8692 rj45 18 0 892K 716K sleep pause 0:00 0.00% tcsh
> 11390 rj45 3 0 896K 716K idle ttyin 0:00 0.00% tcsh
>
> My system is a AMD 400 with 64MB RAM.
>
> The load averages should be lower.Anyone has a suggestion or some hints
> about this behaviour ??
>
> thanks
>
> Rick
>
>
>
Visit your host, monkey.org