Horton

Re: Assert failure due to "drop schema pg_temp_3 cascade" for temporary tables and \d+ is not showing any info after drooping temp table schema

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
Cc: Robert Haas <robertmhaas@gmail.com>, Mahendra Singh <mahi6run@gmail.com>, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-01-08T00:55:17Z
Lists: pgsql-bugs, pgsql-hackers 
Michael Paquier <michael@paquier.xyz> writes:
> Okay, I am running out of arguments then, so attached is a patch to
> address things.  I would also prefer if we keep the relation name in
> the log even if the namespace is missing.

A couple of thoughts:

* Please revert a052f6c as a separate commit specifically doing that,
so that when it comes time to make the release notes, it's clear that
a052f6c doesn't require documentation.

* I think the check on log_min_messages <= LOG is probably wrong, since
LOG sorts out of order for this purpose.  Compare is_log_level_output()
in elog.c.  I'd suggest not bothering with trying to optimize away the
get_namespace_name call here; we shouldn't be in this code path often
enough for performance to matter, and nobody ever cared about it before.

* I don't greatly like the notation
    dropping orphan temp table \"%s.(null).%s\" ...
and I bet Robert won't either.  Not sure offhand about a better
idea --- maybe
    dropping orphan temp table \"%s\" with OID %u in database \"%s\"

			regards, tom lane

Commits

  1. Avoid failure if autovacuum tries to access a just-dropped temp namespace.

  2. Revert "Forbid DROP SCHEMA on temporary namespaces"

  3. Forbid DROP SCHEMA on temporary namespaces

  4. Make autovacuum more aggressive to remove orphaned temp tables

  5. Code review for early drop of orphaned temp relations in autovacuum.