Operating System - Tru64 Unix
1751720 Members
5733 Online
108781 Solutions
New Discussion юеВ

Re: ES47 - strange cpu usage

 
BG Jeong
Advisor

ES47 - strange cpu usage

Hi all.
Our system is ES47.
When I monitor system resource using vmstat,
sys part is so high.

Do anyone know this case?

I'll attach a log file.

Thanks in advance.
Tru64 from Korea
6 REPLIES 6
Kapil Jha
Honored Contributor

Re: ES47 - strange cpu usage

Can you please post output of top and see which process is taking most of CPU.
BR,
Kapil
I am in this small bowl, I wane see the real world......
Venkatesh BL
Honored Contributor

Re: ES47 - strange cpu usage

What do you mean by "sys" part in the output?
Vladimir Fabecic
Honored Contributor

Re: ES47 - strange cpu usage

Hello BG
Can you supply more informations, such as what kind of application is this machine used for, is this machine in cluster, how many RAM do have, any swapping.
Looks like you may have heavy I/O utilization.
In vino veritas, in VMS cluster
Ivan Ferreira
Honored Contributor

Re: ES47 - strange cpu usage

The system is not paging so the high sys is not caused by this activiy.

Use collect to identify the top processes and the disk activity, is this a stand alone host or part of a cluster?

This could be caused by a poor written application.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
BG Jeong
Advisor

Re: ES47 - strange cpu usage

sorry lately appling.

ES47 + smartarray 5402a
2 internal disk mirror
Oracle and Java base web appliction

Top monitoring output just display cpu%.
it's ok.

attach top and collect outputs.
Tru64 from Korea
Hein van den Heuvel
Honored Contributor

Re: ES47 - strange cpu usage

That seems wrong.
No obvious user task to tie the system time to.
Has it always been like this?
When did you first notice?
What changed?
The system has only been runing for a month, but if it was mine, then I would first take a 'therapeutic reboot' and see if the problem goes away, only to start worrying if it comes back again.

If I wanted to know what was going on, then I'd go aske the kernel where it is spending its time. Use kprofile or dcpi.

Just 2 cpus, one in each RAD huh?
8GB is not much, but enough it seems.

Whuich Unix version.
Recent hardware suggest 5.1B-x

7 LAN's huh?
My guess at this time is that the network has a configuration error.
Maybe TCP, but more likely sysconfigtab : inet or maybe ipc.


hth,
Hein.