Please help

Started by Nonameabout 22 years ago3 messages
#1Noname
ohp@pyrenet.fr

I've have a MAJOR crash an hour ago and postgresql doesn't start anymoe
(version 7.3.4).

Here's the error log:

Oct 30 17:16:20 server postgres[4135]: [1-2] This probably means that some data blocks are corrupted
Oct 30 17:16:20 server postgres[4135]: [1-3] and you will have to use the last backup for recovery.
Oct 30 17:16:20 server postgres[4135]: [2] LOG: checkpoint record is at 36/C27C14C0
Oct 30 17:16:20 server postgres[4135]: [3] LOG: redo record is at 36/C2782998; undo record is at 0/0; shutdown FALSE
Oct 30 17:16:20 server postgres[4135]: [4] LOG: next transaction id: 203794305; next oid: 32417798
Oct 30 17:16:20 server postgres[4135]: [5] LOG: database system was not properly shut down; automatic recovery in progress
Oct 30 17:16:20 server postgres[4135]: [6] LOG: redo starts at 36/C2782998
Oct 30 17:16:21 server postgres[4135]: [7] PANIC: Invalid page header in block 6157 of 29135442
Oct 30 17:16:21 server postgres[4132]: [1] LOG: startup process (pid 4135) was terminated by signal 6
Oct 30 17:16:21 server postgres[4132]: [2] LOG: aborting startup due to startup process failure

Is there anything I can do not to reload all backups?

Regards
--
Olivier PRENANT Tel: +33-5-61-50-97-00 (Work)
6, Chemin d'Harraud Turrou +33-5-61-50-97-01 (Fax)
31190 AUTERIVE +33-6-07-63-80-64 (GSM)
FRANCE Email: ohp@pyrenet.fr
------------------------------------------------------------------------------
Make your life a dream, make your dream a reality. (St Exupery)

#2Tom Lane
tgl@sss.pgh.pa.us
In reply to: Noname (#1)
Re: Please help

ohp@pyrenet.fr writes:

I've have a MAJOR crash an hour ago and postgresql doesn't start anymoe
(version 7.3.4).

Oct 30 17:16:21 server postgres[4135]: [7] PANIC: Invalid page header in block 6157 of 29135442
Oct 30 17:16:21 server postgres[4132]: [1] LOG: startup process (pid 4135) was terminated by signal 6
Oct 30 17:16:21 server postgres[4132]: [2] LOG: aborting startup due to startup process failure

Is there anything I can do not to reload all backups?

You could try turning on zero_damaged_pages in postgresql.conf. If you
are lucky, the page in question is going to be rewritten from WAL anyway.

regards, tom lane

#3Noname
ohp@pyrenet.fr
In reply to: Tom Lane (#2)
Re: Please help

Thanks Tom,

The answer came too late and I could'nt wait. pg_resetlog did nearly the
trick, Only one database was really hurt. So I reloaded all but this one
from pg_dumpall then the last one from backup...

I'm cursed
On Thu, 30 Oct 2003, Tom Lane wrote:

Date: Thu, 30 Oct 2003 17:25:02 -0500
From: Tom Lane <tgl@sss.pgh.pa.us>
To: ohp@pyrenet.fr
Cc: pgsql-hackers list <pgsql-hackers@postgresql.org>
Subject: Re: [HACKERS] Please help

ohp@pyrenet.fr writes:

I've have a MAJOR crash an hour ago and postgresql doesn't start anymoe
(version 7.3.4).

Oct 30 17:16:21 server postgres[4135]: [7] PANIC: Invalid page header in block 6157 of 29135442
Oct 30 17:16:21 server postgres[4132]: [1] LOG: startup process (pid 4135) was terminated by signal 6
Oct 30 17:16:21 server postgres[4132]: [2] LOG: aborting startup due to startup process failure

Is there anything I can do not to reload all backups?

You could try turning on zero_damaged_pages in postgresql.conf. If you
are lucky, the page in question is going to be rewritten from WAL anyway.

regards, tom lane

--
Olivier PRENANT Tel: +33-5-61-50-97-00 (Work)
6, Chemin d'Harraud Turrou +33-5-61-50-97-01 (Fax)
31190 AUTERIVE +33-6-07-63-80-64 (GSM)
FRANCE Email: ohp@pyrenet.fr
------------------------------------------------------------------------------
Make your life a dream, make your dream a reality. (St Exupery)