Postgres fails to start
Hi All,
I have experienced that Postgres services was not able to start, I tried from Services and command, and both were failed. Here are the logs I got, please help. (I am using postgres 9.1 and there are plenty disk space left).
2015-04-07 09:01:28 EST LOG: database system was interrupted while in recovery at 2015-04-07 08:53:41 EST
2015-04-07 09:01:28 EST HINT: This probably means that some data is corrupted and you will have to use the last backup for recovery.
2015-04-07 09:01:28 EST FATAL: the database system is starting up
2015-04-07 09:01:28 EST LOG: database system was not properly shut down; automatic recovery in progress
2015-04-07 09:01:28 EST LOG: redo starts at CD1/3C1414D0
2015-04-07 09:01:29 EST FATAL: the database system is starting up
2015-04-07 09:01:30 EST FATAL: the database system is starting up
2015-04-07 09:01:31 EST FATAL: the database system is starting up
2015-04-07 09:01:32 EST FATAL: the database system is starting up
2015-04-07 09:01:33 EST FATAL: the database system is starting up
2015-04-07 09:01:34 EST FATAL: the database system is starting up
2015-04-07 09:01:35 EST FATAL: the database system is starting up
2015-04-07 09:01:36 EST FATAL: the database system is starting up
2015-04-07 09:01:37 EST FATAL: the database system is starting up
2015-04-07 09:01:38 EST FATAL: the database system is starting up
2015-04-07 09:01:39 EST FATAL: the database system is starting up
2015-04-07 09:01:40 EST FATAL: the database system is starting up
2015-04-07 09:01:41 EST FATAL: the database system is starting up
2015-04-07 09:01:43 EST FATAL: the database system is starting up
2015-04-07 09:01:43 EST LOG: unrecognized win32 error code: 1392
2015-04-07 09:01:43 EST CONTEXT: xlog redo split_r: rel 1663/2008723533/2107262657 left 375921, right 379817, next 376394, level 0, firstright 95
2015-04-07 09:01:43 EST FATAL: could not extend file "base/2008723533/2107262657.2": Invalid argument
2015-04-07 09:01:43 EST HINT: Check free disk space.
2015-04-07 09:01:43 EST CONTEXT: xlog redo split_r: rel 1663/2008723533/2107262657 left 375921, right 379817, next 376394, level 0, firstright 95
2015-04-07 09:01:44 EST LOG: startup process (PID 2652) exited with exit code 1
2015-04-07 09:01:44 EST LOG: aborting startup due to startup process failure
Regards,
Haiming Zhang
Software Engineer
________________________________
If you are not an authorised recipient of this e-mail, please contact me at Redflex immediately by return phone call or by email. In this case, you should not read, print, retransmit, store or act in reliance on this e-mail or any attachments, and should destroy all copies of them. This e-mail and any attachments are confidential and may contain privileged information and/or copyright material of Redflex or third parties. You should only retransmit, distribute or commercialise the material if you are authorised to do so. This notice should not be removed.
On Tue, Apr 7, 2015 at 9:25 AM, Haiming Zhang
<Haiming.Zhang@redflex.com.au> wrote:
2015-04-07 09:01:43 EST LOG: unrecognized win32 error code: 1392
[...]
In Windows world, 1392 = ERROR_FILE_CORRUPT = the file or directory is
corrupted and unreadable:
https://msdn.microsoft.com/en-us/library/windows/desktop/ms681385%28v=vs.85%29.aspx
So first check your file system, and be sure that you have a backup around.
--
Michael
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
Hi Michael,
Thank you for replying. The file was there, is that ok to remove the corrupted file to recover postgres? "base/2008723533/2107262657.2"
Unfortunately, I only have a backup on February. Is there a way I can recover it without losing the recent data?
Thanks and regards,
Haiming
-----Original Message-----
From: Michael Paquier [mailto:michael.paquier@gmail.com]
Sent: Tuesday, 7 April 2015 11:36 AM
To: Haiming Zhang
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] Postgres fails to start
On Tue, Apr 7, 2015 at 9:25 AM, Haiming Zhang <Haiming.Zhang@redflex.com.au> wrote:
2015-04-07 09:01:43 EST LOG: unrecognized win32 error code: 1392
[...]
In Windows world, 1392 = ERROR_FILE_CORRUPT = the file or directory is corrupted and unreadable:
https://msdn.microsoft.com/en-us/library/windows/desktop/ms681385%28v=vs.85%29.aspx
So first check your file system, and be sure that you have a backup around.
--
Michael
________________________________
If you are not an authorised recipient of this e-mail, please contact me at Redflex immediately by return phone call or by email. In this case, you should not read, print, retransmit, store or act in reliance on this e-mail or any attachments, and should destroy all copies of them. This e-mail and any attachments are confidential and may contain privileged information and/or copyright material of Redflex or third parties. You should only retransmit, distribute or commercialise the material if you are authorised to do so. This notice should not be removed.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
On Tue, Apr 7, 2015 at 10:46 AM, Haiming Zhang
<Haiming.Zhang@redflex.com.au> wrote:
Thank you for replying. The file was there, is that ok to remove the corrupted file to recover postgres? "base/2008723533/2107262657.2"
Be careful here, I would recommend taking a file-level snapshot before
going on and do perhaps-stupid things. As that's a btree right split,
perhaps you could recover your data by ignoring this index...
Unfortunately, I only have a backup on February. Is there a way I can recover it without losing the recent data?
What is lost is lost. A good backup strategy is essential.
--
Michael
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
Hi Michael,
Thank you again. It seems hard to recover, I will go the hard way (lost data). Learned a lesson.
Regards,
Haiming
-----Original Message-----
From: Michael Paquier [mailto:michael.paquier@gmail.com]
Sent: Tuesday, 7 April 2015 1:04 PM
To: Haiming Zhang
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] Postgres fails to start
On Tue, Apr 7, 2015 at 10:46 AM, Haiming Zhang <Haiming.Zhang@redflex.com.au> wrote:
Thank you for replying. The file was there, is that ok to remove the corrupted file to recover postgres? "base/2008723533/2107262657.2"
Be careful here, I would recommend taking a file-level snapshot before going on and do perhaps-stupid things. As that's a btree right split, perhaps you could recover your data by ignoring this index...
Unfortunately, I only have a backup on February. Is there a way I can recover it without losing the recent data?
What is lost is lost. A good backup strategy is essential.
--
Michael
________________________________
If you are not an authorised recipient of this e-mail, please contact me at Redflex immediately by return phone call or by email. In this case, you should not read, print, retransmit, store or act in reliance on this e-mail or any attachments, and should destroy all copies of them. This e-mail and any attachments are confidential and may contain privileged information and/or copyright material of Redflex or third parties. You should only retransmit, distribute or commercialise the material if you are authorised to do so. This notice should not be removed.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general