On Fri, Mar 2, 2018 at 1:07 PM, Michael Paquier <michael@paquier.xyz> wrote:
> On Fri, Mar 02, 2018 at 02:29:13AM +0900, Fujii Masao wrote:
>> + * write a backup history file with the same name.
>>
>> So more than one backup history files with the same name
>> but the diffferent content can be created and archived.
>> Isn't this problematic because the backup history file that
>> users want to use later might be overwritten unexpectedly?
>
> Yeah, that's the intention behind the patch. Would that actually happen
> in practice though?
Yes, I think. During recovery, all the pg_basebackup would use the same
backup starting point and create the backup history file with the same name
until the next restartpoint runs. So unless the restartpoint interval is short,
ISTM that such a problematic case can happen even in practice. No?
> We would talk about two backups running
> simultaneously on a standby, which would overlap with each other to
> generate a file aimed only at being helpful for debugging purposes, and
> we provide no information now for backups taken from standbys. We could
> of course make that logic a bit smarter by checking if there is an
> extsing file with the same name and create a new file with a different
> name. But is that worth the complication? That's where I am not
> convinced, and that's the reason why this patch is doing things this
> way.
What about including the backup history file in the base backup instead of
creating it in the standby's pg_wal and archiving it? As the good side effect
of this approach, we can use the backup history file for debugging purpose
even when WAL archiving is not enabled.
Regards,
--
Fujii Masao