Re: Determining what is causing load when server is idle.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Apr 20, 2006 at 11:28:50AM +0200, George Magklaras wrote:
> Hi Ray,
> 
> Sorry, I lost the thread on my browser. I have looked at the figures and 
> no I do not see an excessive number of page faults. The vmstat you 
> indicated on a very quiet system does indeed point to an unusually high 
> number of interrupts.
> 
> If you still have this issue, try booting on a uniprocessor kernel or 
> boot with the 'noapic' option and see if you face the symptom or your 
> system boots properly. I also note that the kernel you mentioned did not 
> appear to be the latest. So, I would up2date the system, see if the 
> problem goes away and then on the up-to-date kernel (2.6.9-34.ELsmp as 
> we speak) boot with the noapic option and see if you see a difference.

I have tried noapic mode, but I do not remember the results unfortunately.
Will give it a shot again and document.

Also will upgrade the kernel, but just a side note -- I built a custom
kernel based on 2.6.16 and the load is a *lot* closer to where I would
expect it.  Well not quite -- but the I/O "glitches" are gone although the
load still never gets to 0.00 even on a completely idle box.

Need to compare vmstat output from that kernel to what I'm using now.

I suspect the combo of the ata_piix driver and software raid, but need a way
to prove or disprove that connection. :)

Thanks for the help,
Ray

-- 
redhat-list mailing list
unsubscribe mailto:redhat-list-request@xxxxxxxxxx?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/redhat-list

[Index of Archives]     [CentOS]     [Kernel Development]     [PAM]     [Fedora Users]     [Red Hat Development]     [Big List of Linux Books]     [Linux Admin]     [Gimp]     [Asterisk PBX]     [Yosemite News]     [Red Hat Crash Utility]


  Powered by Linux