server went down

[Brian Cluff]

You might want to try running a 2.4 kernel if you aren't already.  It's
supposed to take care problems like you are describing, and you'll get a lot
better bang for your buck from the CPU's.
Just be care if you are running Reiser on RAID, there are some curruption
problems with that setup... otherwise..  good stuff!

Brian Cluff
----- Original Message -----
>   5:23pm  up  2:15, 191 users,  load average: 11.46, 11.16, 14.25
> 872 processes: 858 sleeping, 14 running, 0 zombie, 0 stopped
> CPU0 states: 18.3% user, 81.5% system,  0.0% nice,  0.0% idle
> CPU1 states: 17.0% user, 83.0% system,  0.0% nice,  0.0% idle
> CPU2 states: 18.7% user, 81.1% system,  0.0% nice,  0.0% idle
> CPU3 states: 17.1% user, 82.7% system,  0.0% nice,  0.0% idle
> Mem:  3954672K av, 2090204K used, 1864468K free,  573700K shrd,  331720K
> buff
> Swap: 1056092K av,    1684K used, 1054408K free                 1453604K
> cached
>
> this quad system constantly has 80% useage on all 4 procs??? is this
> normal???
>
> We hit 650 users today and had to reboot.  Not good.  The system was
> still
> running but the load was so high that it would not move.  There seem to
> be atleast
> 2-3 process left hanging after someone logs out.  usually -sh and our
> bussness
> application.
>
> What in a program can cause it not to die when your session is over? and
> what can i do to help clean these up rather than just haveing greping
> them out
> of a ps and killing them?
>
> --
> --
> Bill Warner
> Direct Alliance Corp.
> Unix/Linux Admin.
>
> ________________________________________________
> See http://PLUG.phoenix.az.us/navigator-mail.shtml if your mail doesn't
post to the list quickly and you use Netscape to write mail.
>
> Plug-discuss mailing list  -  Plug-discuss@lists.PLUG.phoenix.az.us
> http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
>