On 2014-04-02 20:59:03 +0900, Michael Paquier wrote:
> On Wed, Apr 2, 2014 at 6:58 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> > On 2014-04-02 09:59:28 +0900, Michael Paquier wrote:
> >> On Tue, Apr 1, 2014 at 11:59 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> >> > On 2014-04-01 16:45:46 +0200, Magnus Hagander wrote:
> >> >> On Tue, Apr 1, 2014 at 2:24 PM, Michael Paquier
> >> >> <michael.paquier@gmail.com>wrote:
> >> >> > As of now, pg_basebackup creates an empty repository for pg_replslot/
> >> >> > in a base backup, forcing the user to recreate slots on other nodes of
> >> >> > the cluster with pg_create_*_replication_slot, or copy pg_replslot
> >> >> > from another node. This is not really user-friendly especially after a
> >> >> > failover where a given slave may not have the replication slot
> >> >> > information of the master that it is replacing.
> >> >
> >> > What exactly is your use case for copying the slots?
> >
> >> I had in mind users that want to keep around base backups that could
> >> be used for recovery operations like PITR using a base backup and
> >> archives. It does not apply directly to a live standby, as it would
> >> mean that this standby would be defined to retain WAL for other slaves
> >> connected to the master.
> >
> > I honestly can't follow why that implies copying the slots?
> You simply do not need to recreate manually the slots on the new master.
That doesn't seem like a good justification. The new master won't
necessarily have all the neccessary WAL available, no?
Greetings,
Andres Freund
-- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services