On 1st May, I saw this message in my postgres log:
May 2 06:52:02 db10 postgres: [29829-1] 2011-05-02 06:52:02 SGT
ERROR: could not access status of transaction 1573786613
May 2 06:52:02 db10 postgres: [29829-2] 2011-05-02 06:52:02 SGT
DETAIL: Could not open file "pg_clog/05DC": No such file or directory.
May 2 06:52:02 db10 postgres: [29829-3] 2011-05-02 06:52:02 SGT
STATEMENT: SELECT 1 FROM core_bill_id_seq FOR UPDATE
Now, I'm not sure what I should do about it. Database behaves "funny",
some inserts do not work.
Searching the internet suggests that:
1) such errors could happen with PostgreSQL 8.1.x under heavy load -
this server is under constant heavy load, but runs 8.3.14 on Debian Lenny
2) I should simply create a 256k pg_clog/05DC empty file with dd - I
wouldn't like to do it, without first knowing what happened, and if it's
really "good fix"
3) some tables can be corrupted - how can I check that? pg_dump works
fine and doesn't report any errors
4) I may have hardware problems - but this server is running for almost
1 year now, is super stable - servers with hardware issues are likely to
show some issues as well
5) database corrupted due to a server crash - this server never crashed
How should I continue from that (assuming I can't reliably verify if
something wrong is going with the hardware or not - points 4 and 5)?
pgsql-hackers by date
|Next:||From: Alexander Korotkov||Date: 2011-05-04 08:24:21|
|Subject: Re: Extreme bloating of intarray GiST indexes|
|Previous:||From: Johann 'Myrkraverk' Oskarsson||Date: 2011-05-04 05:25:31|
|Subject: Re: DLL export with mingw-w64: currently a no-op|
pgsql-admin by date
|Next:||From: Tobias Schneider||Date: 2011-05-04 09:34:25|
|Subject: Authentification via Kerberos with ProxyTicket|
|Previous:||From: Sam Stearns||Date: 2011-05-03 22:52:06|
|Subject: Re: Unable to Insert Row|