On 11/27/18 4:25 PM, Michael Paquier wrote:
> On Tue, Nov 27, 2018 at 09:45:04AM -0500, Stephen Frost wrote:
>> If you don't consider your recovery scripts and your backup scripts to
>> be related then I've really got to wonder how you're regularly testing
>> your backups to make sure that they're actually valid.
>
> Base backups can be perfectly self-contained as long as they include all
> the WAL segments needed to recover up to the end-of-backup record.
> That's what pg_basebackup does with its default options
> (--wal-method=stream in particular).
This is true, of course, but it misses one of the major benefits of
file-level backups which is PITR.
>> If you aren't regularly testing your backups then I've got little
>> sympathy.
>
> Fortunately they do, hundreds of time on a daily basis ;)
Nice!
>> To be clear, pgbackrest doesn't have any dependency here- but it, like
>> all of the other 3rd party backup solutions and any restore solution
>> that a user has come up with, are going to have to be changed to deal
>> with the changes in how recovery works, so this is a good time to make
>> these changes.
>
> My point is that base backups do not have a mandatory dependency with
> recovery.conf all the time as they can perfectly be restored if they are
> standalone backups. I can see a dependency with recovery.conf once you
> have a base backup which needs to be fed with WAL segments from an
> external archive, or when using a base backup to create a standby.
If you want to do PITR -- which is the default in most situations --
then some interaction with recovery.conf is needed. I think you would
be hard-pressed to find a prominent HA or backup solution that doesn't
do so.
--
-David
david@pgmasters.net