From: | "Jolles, Peter M (GE Energy)" <peter(dot)jolles(at)ge(dot)com> |
---|---|
To: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Cannot start Postgres- FATAL: invalid cache id: 19 |
Date: | 2010-12-01 16:41:31 |
Message-ID: | E66B1BA008C7374390C6AE9EF01E584A03B5CC01@ALPMLVEM08.e2k.ad.ge.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Wednesday, December 01, 2010 9:58 AM Vick Khera wrote:
> On Wed, Dec 1, 2010 at 9:20 AM, Jolles, Peter M (GE Energy) wrote:
> > 2010-11-29 12:39:17 EST HINT: Stop the postmaster and use a
> > standalone backend to vacuum that database.
>
> so did you do what the HINT told you to do? does it still not startup?
When I realized what was happening I stopped the server and attempted to restart it after making a backup. I was under the impression the database had to be able to start up in order to vacuum it. If there is a way to do it with the database not running, I don't know how to do that.
> > I had multiple auto-vacuum instances running, but it looks like they
> > were unable to keep up.
> >
> are there any log lines complaining that autovacuum terminated any jobs?
> anything from autovacuum?
I don't see anything in the log that the autovacuum specifically terminated any jobs or gave any error. Approximately 4 hours before the database started refusing transactions the following was present, but I wasn't able to shut everything down in time:
EST WARNING: database "fps_data" must be vacuumed within 11000000 transactions
I know now that I'm going to have to slow down the transaction rate in order to let the autovacuum keep up, but at this point I am just trying to salvage the data present in the database.
Thanks,
Peter Jolles
From | Date | Subject | |
---|---|---|---|
Next Message | Anibal David Acosta | 2010-12-01 17:11:22 | How to recover a postgres installation from files |
Previous Message | Gary Chambers | 2010-12-01 16:10:39 | Re: Grant command help -- postgres |