| From: | Jeff Boes <jboes(at)nexcerpt(dot)com> |
|---|---|
| To: | |
| Cc: | pgsql-admin(at)postgresql(dot)org |
| Subject: | Re: Drop and reload table to clear corruption? |
| Date: | 2002-10-22 21:40:38 |
| Message-ID: | 1035322838.26402.107.camel@takin.private.nexcerpt.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
On Tue, 2002-10-22 at 17:02, Tom Lane wrote:
> jboes(at)nexcerpt(dot)com (Jeff Boes) writes:
> > We have a few tables that seem to have suffered some kind of
> > corruption in their TOAST files. Symptoms include:
>
> > + Runaway CPU usage during VACUUM ANALYZE (load average
> > spikes of 20x normal for 10-15 minutes at a time) and
> > more rarely during other operations
>
> This seems odd. Can you attach to one of the runaway backends with a
> debugger and get a stack trace? That might give some clue what the
> problem is.
No; we can't do *anything* while this is happening. The load average
(on a Linux system) runs up to 30+ within a few minutes. By the time we
notice it happening, no command prompts are responsive.
--
Jeff Boes vox 616.226.9550 ext 24
Database Engineer fax 616.349.9076
Nexcerpt, Inc. http://www.nexcerpt.com
...Nexcerpt... Extend your Expertise
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Nils-Erik Svangård | 2002-10-23 00:03:59 | Recover data from crashed HD |
| Previous Message | Tom Lane | 2002-10-22 21:02:16 | Re: Drop and reload table to clear corruption? |