Re: BUG #3484: Missing pg_clog file / corrupt index

From: Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc>
To: Feng Chen <fchen(at)covergence(dot)com>
Cc: Gregory Stark <stark(at)enterprisedb(dot)com>, Marc Schablewski <ms(at)clickware(dot)de>, "Decibel!" <decibel(at)decibel(dot)org>, pgsql-bugs(at)postgresql(dot)org
Subject: Re: BUG #3484: Missing pg_clog file / corrupt index
Date: 2007-08-24 13:57:08
Message-ID: 46CEE3B4.4060004@kaltenbrunner.cc
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Feng Chen wrote:
> We also have the same exact problem - every 5 to 10 days when the data
> get to some size, PostgreSQL complains about missing pg_clog files, and
> invalid page headers during either vacuum or reindex operations.
>
> The problem happens on different customer sites with Linux 2.6.11.
>
> There is one particular table that is much more heavily used - 99%
> inserts, some selects. And this table turns to be the one having
> problems.
>
> If this happens at different companies so frequently, I'd doubt it's a
> hardware problem. And we did check the hardware but did not find any
> problems.
>
[...]
> 2007-08-13T10:12:07+00:00 ERROR: invalid page header in block 4018 of
> relation "calllegstart_sessionid"
> 2007-08-13T10:12:15+00:00 ERROR: could not access status of
> transaction 0
> 2007-08-13T10:12:15+00:00 DETAIL: could not create file
> "pg_subtrans/0201": File exists

what version of postgresql is this exactly ? there is a problem in older
8.1 versions that could cause this error under high transaction rates.

Stefan

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Alvaro Herrera 2007-08-24 14:01:58 Re: BUG #3484: Missing pg_clog file / corrupt index
Previous Message Andrew Sullivan 2007-08-24 13:54:01 Re: BUG #3484: Missing pg_clog file / corrupt index