Horton

Re: BUG #17268: Possible corruption in toast index after reindex index concurrently

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Noah Misch <noah@leadboat.com>
Cc: Andres Freund <andres@anarazel.de>, Alexey Ermakov <alexey.ermakov@dataegret.com>, Robert Haas <robertmhaas@gmail.com>, Maxim Boguk <maxim.boguk@gmail.com>, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>, Peter Geoghegan <pg@bowt.ie>
Date: 2021-11-11T09:09:49Z
Lists: pgsql-bugs 
On Mon, Nov 08, 2021 at 10:42:17PM -0800, Noah Misch wrote:
> On Mon, Nov 08, 2021 at 10:31:19PM -0800, Andres Freund wrote:
>> On 2021-11-08 21:46:25 -0800, Noah Misch wrote:
>>> Doing both sounds harmless.  Regarding REINDEX CONCURRENTLY on a system
>>> catalog, I bet that can still reach bugs even if we do both, considering this:
>> 
>> Normal catalog aren't REINDEX CONCURRENTLY. It's just toast tables that are
>> kind of system-y, but also kind of not, that can be reindexed concurrently.
> 
> Oh, okay.

To be clear on this point, users cannot reindex concurrently catalog
indexes and toast indexes associated to catalog tables, just toast
indexes of normal tables.  I don't know if any of you have been
working on a patch, but I was cooking something.  It would be worth
checking if an isolation test could be written.
--
Michael

Commits

  1. Fix corruption of toast indexes with REINDEX CONCURRENTLY