Horton

Re: Subscription tests fail under CLOBBER_CACHE_ALWAYS

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Amit Langote <amitlangote09@gmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, Amit Kapila <amit.kapila16@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2021-05-22T15:32:43Z
Lists: pgsql-hackers 
I wrote:
> Amit Langote <amitlangote09@gmail.com> writes:
>> BTW, I think we'd need to cherry-pick f3b141c4825 (or maybe parts of
>> it) into v13 branch for back-patching this.

> I already did a fair amount of that yesterday, cf 84f5c2908 et al.
> But that does raise the question of how far we need to back-patch this.
> I gather that the whole issue might've started with 1375422c, so maybe
> we don't really need a back-patch at all?

... wrong.  Running v13 branch tip under CLOBBER_CACHE_ALWAYS provokes
a core dump in 013_partition.pl, so 1375422c is not to blame.  Now
I'm wondering how far back there's a live issue.

			regards, tom lane

Commits

  1. Be more verbose when the postmaster unexpectedly quits.

  2. Fix access to no-longer-open relcache entry in logical-rep worker.