Re: Question about switchover with PG9 replication

From: Cyril Scetbon <cyril(dot)scetbon(at)free(dot)fr>
To: Wouter D'Haeseleer <wdh(at)vasco(dot)com>
Cc: Jean-Armel Luce <jaluce06(at)gmail(dot)com>, "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: Question about switchover with PG9 replication
Date: 2011-02-27 08:46:13
Message-ID: 4D6A0F55.4060300@free.fr
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Le 07/02/2011 09:57, Wouter D'Haeseleer a écrit :
> Question 1 : is it possible to have such a replication configuration
> with the streaming replication of PG9 (cascaded replication) ?
>
> Nope, as far as I have tested pg only has 1 master and can have a
> number of slaves, so having 2 masters is not possible.
>
The second host named master2 is a slave which has slaves too, not a
master on which you can write.
>
>
> Question 2 : All the procedures I have seen describing a switchover
> between 2 PG servers require to copy (or rsync) the database from the
> new master (old slave) to the new slave (old master).
> Is it possible to do switchover between sites (between PGMaster1 and
> PGMaster2) whithout copying all the database from the new PG master to
> the new PG slave ?
> If it is not possible yet, shall it be possible in future releases ?
>
> Nope this is not possible, pg requires to have an updates
> basebackup at the slave, this is because if you loose a lot of
> streamed wal files it is impossible for the slave to catch up and
> have consistent data.
>
>

--
Cyril SCETBON

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Aleksey Tsalolikhin 2011-02-27 08:49:48 Re: database is bigger after dump/restore - why? (60 GB to 109 GB)
Previous Message Amitabh Kant 2011-02-27 00:27:17 Re: Partitions and indexes