Re: 100% CPU pg processes that don't die. - Mailing list pgsql-general

From Scott Marlowe
Subject Re: 100% CPU pg processes that don't die.
Date
Msg-id dcc563d10808091240j607390a7i707a58abfdcabbd3@mail.gmail.com
Whole thread Raw
In response to Re: 100% CPU pg processes that don't die.  ("Scott Marlowe" <scott.marlowe@gmail.com>)
Responses Re: 100% CPU pg processes that don't die.
List pgsql-general
On Sat, Aug 9, 2008 at 1:38 PM, Scott Marlowe <scott.marlowe@gmail.com> wrote:
> On Sat, Aug 9, 2008 at 1:31 PM, Stephen Frost <sfrost@snowman.net> wrote:
>> * Scott Marlowe (scott.marlowe@gmail.com) wrote:
>>> Aug  9 13:13:21 engelberg kernel: [71242.735046]
>>>
>>> Does this look like a kernel bug or a pgsql bug to most people?
>>
>> It's certainly something kernel-related.  It might be the OOM killer
>> though..  You might want to disable that.  Is the box running out of
>> memory (run 'free')?
>
> Hehe, no.  It's got 32 Gig of ram and is using 25G of that for kernel
> cache.  There are no entries in any log for oom killer that I can see.
>
> Hmmm.  I'm running the latest Ubuntu LTS with 2.6.24-19-server kernel.
>
> Just ran the latest apt-get upgrade and only was missing an update to
> pciutils.
>
> btw, here's the output of free:
>
>             total       used       free     shared    buffers     cached
> Mem:      33080292   32983004      97288          0      87992   24882632
> -/+ buffers/cache:    8012380   25067912
> Swap:      7815580        144    7815436
>
> Interesting thing, I've got a machine with the same configuration
> sitting next to it that's not doing this.  But it seems odd it could
> be hardware induced.

I take that back.  This problem followed the RAID card from one
machine to another.

pgsql-general by date:

Previous
From: "Scott Marlowe"
Date:
Subject: Re: 100% CPU pg processes that don't die.
Next
From: Stephen Frost
Date:
Subject: Re: 100% CPU pg processes that don't die.