| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Bruce Momjian <bruce(at)momjian(dot)us> |
| Cc: | Matthias van de Meent <boekewurm+postgres(at)gmail(dot)com>, Jehan-Guillaume de Rorthais <jgdr(at)dalibo(dot)com>, Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>, pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: pg_walfile_name_offset can return inconsistent values |
| Date: | 2023-11-09 23:25:35 |
| Message-ID: | ZU1qb5NBAKpcIUU_@paquier.xyz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Nov 09, 2023 at 04:14:07PM -0500, Bruce Momjian wrote:
> Attached is the full patch that changes pg_walfile_name_offset() and
> pg_walfile_name(). There is no need for doc changes. We need to
> document this as incompatible in case users are realying on the old
> behavior for WAL archiving purposes. If they want the old behavior they
> need to check for an offset of zero and subtract one from the file name.
FWIW, I am not really convinced that there is a strong need to
backpatch any of that. There's a risk that some queries relying on
the old behavior suddenly break after a minor release, and that's
always annoying. A HEAD-only change seems like a safer bet to me.
> Can someone check that all other calls to XLByteToPrevSeg() are correct?
On a quick check, all the other calls use that for end record LSNs, so
that looks fine.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2023-11-09 23:35:42 | Re: pg_walfile_name_offset can return inconsistent values |
| Previous Message | Andres Freund | 2023-11-09 23:12:50 | Re: pg_upgrade failing for 200+ million Large Objects |