| From: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
|---|---|
| To: | Peter Smith <smithpb2250(at)gmail(dot)com> |
| Cc: | Markus Wanner <markus(dot)wanner(at)enterprisedb(dot)com>, Ajin Cherian <itsajin(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: [PATCH] add concurrent_abort callback for output plugin |
| Date: | 2021-04-26 08:12:32 |
| Message-ID: | CAA4eK1LQ4TbgQ_sehvJQeKmmDp=jWznXKzaLvXpdGAofqn1N1Q@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Mon, Apr 26, 2021 at 7:05 AM Peter Smith <smithpb2250(at)gmail(dot)com> wrote:
>
> Hi,
>
> While testing another WIP patch [1] a clashing GID problem was found,
> which gives us apply worker errors like:
>
> 2021-04-26 10:07:12.883 AEST [22055] ERROR: transaction identifier
> "pg_gid_16403_608" is already in use
> 2021-04-26 10:08:05.149 AEST [22124] ERROR: transaction identifier
> "pg_gid_16403_757" is already in use
>
> These GID clashes were traced back to a problem of the
> concurrent-abort logic: when "streaming" is enabled the
> concurrent-abort logic was always sending "prepare" even though a
> "stream_prepare" had already been sent.
>
> PSA a patch to correct this.
>
Your patch looks good to me, so pushed! Thanks for the report and patch.
--
With Regards,
Amit Kapila.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Michael Paquier | 2021-04-26 08:30:25 | Re: [PATCH] Re: pg_identify_object_as_address() doesn't support pg_event_trigger oids |
| Previous Message | Dilip Kumar | 2021-04-26 07:56:47 | Re: [BUG] "FailedAssertion" reported when streaming in logical replication |