Subject: Re: load average going to stars...
To: RJ45 <rj45@slacknet.com>
From: Andrey Petrov <petrov@netbsd.org>
List: port-sparc64
Date: 02/12/2003 19:07:43
On Wed, Feb 12, 2003 at 03:59:17PM -0700, RJ45 wrote:
>
> on my ultra 10 after some heavy file transfer both via scp and ftp using
> the ex0 NIC the load average stay fixed to 3.
> There is no CPU consuming process:
>
> ps -auwx | less
> USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND
> root 3 2.1 15.6 0 81592 ?? DK 8:18PM 3:55.14 [pagedaemon]
> root 754 0.0 0.1 960 416 p0 R+ 11:52PM 0:00.00 ps -auwx
> root 212 0.0 0.4 1624 1944 p0 Ds 8:32PM 0:-1.83 -tcsh
> root 210 0.0 0.4 976 2232 ?? Ss 8:32PM 0:-1.77 sshd: root@ttyp0
> root 206 0.0 0.0 96 856 ?? SWs+ - 0:00.00 /usr/libexec/getty
> suncons
> console
> root 191 0.0 0.2 312 880 ?? Ss 8:18PM 0:00.08 /usr/sbin/cron
> root 186 0.0 0.2 136 968 ?? Ss 8:18PM 0:00.01 /usr/sbin/inetd -l
> root 173 0.0 0.0 784 1664 ?? SWs - 0:00.00 /usr/sbin/sshd
> root 93 0.0 0.2 248 864 ?? Ss 8:18PM 0:-1.22 /usr/sbin/syslogd -s
> root 6 0.0 15.6 0 81592 ?? DK 8:18PM 0:31.79 [aiodoned]
> root 5 0.0 15.6 0 81592 ?? DK 8:18PM 0:26.54 [ioflush]
> root 4 0.0 15.6 0 81592 ?? DK 8:18PM 0:00.25 [reaper]
> root 2 0.0 15.6 0 81592 ?? DK 8:18PM 0:00.00 [atapibus0]
> root 1 0.0 0.0 880 136 ?? SWs - 0:00.00 init
> root 755 0.0 0.4 1624 1944 p0 RV 11:52PM 0:00.00 -tcsh
> root 0 0.0 15.6 0 81592 ?? DKs 8:18PM 0:00.50 [swapper]
>
> but the load average stays always high:
>
> 11:52PM up 3:34, 1 user, load averages: 3.20, 3.15, 3.02
> USER TTY FROM LOGIN@ IDLE WHAT
> root p0 sword.hidden.org 8:32PM 0 w
>
I'd use top or vmstat or systat because ps is a snapshot thing and loadaverage
is historical. Are there any other anomalies or just this?
Andrey