Re: Backup history file should be replicated in Streaming Replication?

Fujii Masao wrote:
> On Thu, Dec 24, 2009 at 1:39 PM, Fujii Masao <masao(dot)fujii(at)gmail(dot)com> wrote:
>> On Wed, Dec 23, 2009 at 7:50 PM, Heikki Linnakangas
>> <heikki(dot)linnakangas(at)enterprisedb(dot)com> wrote:
>>> Ok. How about writing the history file in pg_stop_backup() for
>>> informational purposes only. Ie. never read it, but rely on the WAL
>>> records instead.
>> Sounds good. I'll make such change as a self-contained patch.
>
> Done. Please see the attached patch.
>
> Design:
>
> * pg_stop_backup writes the backup-end xlog record which contains
> the backup starting point.
>
> * In archive recovery, the startup process doesn't mark the database
> as consistent until it has read the backup-end record.
>
> * A backup history file is still created as in the past, but is never
> used.

As the patch stands, reachedBackupEnd is never set to true if starting
from a restore point after the end-of-backup. We'll need to store the
information that we've reached end-of-backup somewhere on disk.

Here's is modified patch that adds a new backupStartPoint field to
pg_control for that + some other minor editorialization.

--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com