Re: remove more archiving overhead - Mailing list pgsql-hackers

From David Steele
Subject Re: remove more archiving overhead
Date
Msg-id 58b2ae42-5378-778f-e198-3dcaf5ecc47a@pgmasters.net
Whole thread Raw
In response to Re: remove more archiving overhead  (Nathan Bossart <nathandbossart@gmail.com>)
Responses Re: remove more archiving overhead
List pgsql-hackers
On 7/7/22 12:18, Nathan Bossart wrote:
> On Thu, Jul 07, 2022 at 10:46:23AM -0400, David Steele wrote:
> 
>> There are plenty of ways that already-archived WAL might get archived again
>> and this is just one of them.
> 
> What are some of the others?  I was aware of the case that was fixed in
> ff9f111, where we might try to re-archive a file with different contents,
> but I'm curious what other ways you've seen this happen.

On the PG side, crashes and (IIRC) immediate shutdown.

In general, any failure in the archiver itself. Storage, memory, 
network, etc. There are plenty of ways that the file might make it to 
storage but postgres never gets notified, so it will retry.

Any archiver that is not tolerant of this fact is not going to be very 
useful and this patch only makes it slightly more true.

Regards,
-David



pgsql-hackers by date:

Previous
From: Greg Stark
Date:
Subject: Re: explain analyze rows=%.0f
Next
From: Nathan Bossart
Date:
Subject: Re: remove more archiving overhead