pg.ddx.io  pgsql-admin@postgresql.org mailing list archive  
help / color / mirror / Atom feed
WAL corruption while replication
4+ messages / 2 participants
[nested] [flat]

* WAL corruption while replication
@ 2012-10-17 19:33 Parkirat Bagga <parkiratbagga@gmail.com>
  0 siblings, 0 replies; 4+ messages in thread

From: Parkirat Bagga @ 2012-10-17 19:33 UTC (permalink / raw)
  To: pgsql-admin

Hi,

How to recover, if master postgres WAL got corrupted before it is applied to
the standby server?

Will I loose all the data present in the WAL segment?

How can I partially apply the WAL?

Regards,
Parkirat Singh Bagga.



--
View this message in context: http://postgresql.1045698.n5.nabble.com/WAL-corruption-while-replication-tp5728689.html
Sent from the PostgreSQL - admin mailing list archive at Nabble.com.




^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* WAL corruption while replication
@ 2012-10-17 20:36 Parkirat Bagga <parkiratbagga@gmail.com>
  2012-10-18 07:50 ` Re: WAL corruption while replication Albe Laurenz <laurenz.albe@wien.gv.at>
  0 siblings, 1 reply; 4+ messages in thread

From: Parkirat Bagga @ 2012-10-17 20:36 UTC (permalink / raw)
  To: pgsql-admin

Hi,

How to recover, if master postgres WAL got corrupted before it is applied to
the standby server?

Will I loose all the data present in the WAL segment?

How can I partially apply the WAL?

Regards,
Parkirat Singh Bagga. 



--
View this message in context: http://postgresql.1045698.n5.nabble.com/WAL-corruption-while-replication-tp5728705.html
Sent from the PostgreSQL - admin mailing list archive at Nabble.com.




^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* Re: WAL corruption while replication
  2012-10-17 20:36 WAL corruption while replication Parkirat Bagga <parkiratbagga@gmail.com>
@ 2012-10-18 07:50 ` Albe Laurenz <laurenz.albe@wien.gv.at>
  2012-10-18 17:34   ` Re: WAL corruption while replication Parkirat Bagga <parkiratbagga@gmail.com>
  0 siblings, 1 reply; 4+ messages in thread

From: Albe Laurenz @ 2012-10-18 07:50 UTC (permalink / raw)
  To: Parkirat Bagga *EXTERN* <parkiratbagga@gmail.com>; pgsql-admin

Parkirat Bagga wrote:
> How to recover, if master postgres WAL got corrupted before it is
applied to
> the standby server?
> 
> Will I loose all the data present in the WAL segment?
> 
> How can I partially apply the WAL?

During normal recovery:
-----------------------

WAL will be replayed until the point of corruption,
then the server will stop recovering and come up.
The last WAL segment will be applied until the
point of corruption.

There is no way to recover past a corruption in
a WAL segment.

On a standby:
-------------

If you have streaming replication, you may not even
notice the problem if the standby is already connected
to the primary, because it does not read WAL archives in that case.

If you don't have streaming replication, the standby will
keep trying to apply the corrupted WAL segment and will
get stuck.  There's no way to get the standby past the
point of corruption.
The same will happen in streaming replication if the primary
does not have the required information any more and the
standby has to use the WAL archives.

Yours,
Laurenz Albe




^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* Re: WAL corruption while replication
  2012-10-17 20:36 WAL corruption while replication Parkirat Bagga <parkiratbagga@gmail.com>
  2012-10-18 07:50 ` Re: WAL corruption while replication Albe Laurenz <laurenz.albe@wien.gv.at>
@ 2012-10-18 17:34   ` Parkirat Bagga <parkiratbagga@gmail.com>
  0 siblings, 0 replies; 4+ messages in thread

From: Parkirat Bagga @ 2012-10-18 17:34 UTC (permalink / raw)
  To: pgsql-admin

Thanks Albe for the clarification.

Regards,
Parkirat Singh Bagga.



--
View this message in context: http://postgresql.1045698.n5.nabble.com/WAL-corruption-while-replication-tp5728705p5728878.html
Sent from the PostgreSQL - admin mailing list archive at Nabble.com.




^ permalink  raw  reply  [nested|flat] 4+ messages in thread


end of thread, other threads:[~2012-10-18 17:34 UTC | newest]

Thread overview: 4+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2012-10-17 19:33 WAL corruption while replication Parkirat Bagga <parkiratbagga@gmail.com>
2012-10-17 20:36 WAL corruption while replication Parkirat Bagga <parkiratbagga@gmail.com>
2012-10-18 07:50 ` Albe Laurenz <laurenz.albe@wien.gv.at>
2012-10-18 17:34   ` Parkirat Bagga <parkiratbagga@gmail.com>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox