On 01/09/2014 10:16 PM, Tom Lane wrote:
> Heikki Linnakangas <hlinnakangas@vmware.com> writes:
>> Actually, why is the partially-filled 000000010000000000000002 file
>> archived in the first place? ...
>
>> So, the rationale is that otherwise it would take a long time until that
>> segment is archived. To be precise, I don't think the segment with the
>> old TLI would ever be archived without the above, but the same segment
>> on the new timeline would, after it fills up.
>
>> Wouldn't it be better to not archive the old segment, and instead switch
>> to a new segment after writing the end-of-recovery checkpoint, so that
>> the segment on the new timeline is archived sooner?
>
> Don't we want to archive both? If you want to recover to the end of the
> old timeline, you're going to need that file too, no?
Hmm. It should be the responsibility of the original server to archive
the segment on the old timeline. Mind you, partial segments are never
archived, except for this one case, so how did the old segment find its
way to the new server? A few possibilities come to mind: the DBA
manually copied it from the old server to pg_xlog, it was streamed by
streaming replication, or it was included in a base backup. The OP's
test script resembles the base backup case.
In all of those cases, I don't think it's the new server's
responsibility to archive it. If it was copied to pg_xlog manually, the
administrator may also copy it to the archive if he feels like it. If it
was streamed from a live server, the original server should take care of
it. If it was included in a backup, well, it's included in the backup so
it doesn't necessarily need to be archived.
- Heikki