Re: Postgres 9.0.4 replication issue: FATAL: requested WAL segment 0000000100000B110000000D has already been removed

Hi Tomas,

Thanks. Increasing the wal_keep_segments to 4000 did the trick.

I will set up WAL archive as well.
Thanks again.

Ashish.
On Sat, Nov 19, 2011 at 5:59 PM, Tomas Vondra <tv(at)fuzzy(dot)cz> wrote:

> Hi,
>
> On 19 Listopad 2011, 10:44, Ashish Gupta wrote:
> > I searched on various forums, where people encountered similar error,
> > however in all such issues WAL file existed on Master. In this case
> Master
> > is not retaining the WAL file required by the Slave.
> >
> > I am unable to understand as to why Master is not retaining the WAL
> files.
> > Any pointer/suggestions would be helpful.
> > Thanks for attention.
>
> The cause is very simple - the standby needs all WAL segments created
> since the backup started, but the master removes some of them. There are
> two ways to fix this:
>
> 1) increase the wal_keep_segments so that enough segments is kept
>
> It seems that the slave asked for B110000000D when master already created
> B110000000D. That's almost 4000 segments if I'm counting correctly. That
> means your database is either quite busy or the backup takes very long
> time.
>
> This stores all the data on master, so you'll have to keep that in mind
> when planning the capacity. For example the 4000 segments are almost 64GB.
>
> 2) Set up a WAL archive - a separate instance where the WAL segments are
> kept. See how the archive_command works. And there's pg_archivecleanup for
> maintenance of the archive.
>
> Tomas
>
>