Re: More issues with pg_verify_checksums and checksum verification in base backups

Stephen Frost <sfrost@snowman.net>

From: Stephen Frost <sfrost@snowman.net>
To: Michael Paquier <michael@paquier.xyz>
Cc: Andres Freund <andres@anarazel.de>, David Steele <david@pgmasters.net>, Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>, pgsql-hackers@postgresql.org, andrew@dunslane.net, daniel@yesql.se, magnus@hagander.net, tgl@sss.pgh.pa.us
Date: 2018-11-27T22:45:41Z
Lists: pgsql-hackers
Greetings,

* Michael Paquier (michael@paquier.xyz) wrote:
> On Mon, Nov 19, 2018 at 10:17:19PM -0500, Stephen Frost wrote:
> > Let's try to not conflate these two issues though, they're quite
> > independent.
> 
> This is a poke about a recent issue raised by Michael Banck here:
> https://www.postgresql.org/message-id/f1543332405.17247.9.camel@credativ.de
> And for which I am suggesting a minimal fix, which is extracted from a
> patch at the most-top of this thread:
> https://www.postgresql.org/message-id/20181127213625.GM1716@paquier.xyz
> 
> If there are no objections, I would like to fix the actual issue.  Then
> I'll rebase a patch on top of what has been fixed for this thread for
> what I proposed in the base backup code.

So the as-committed "whitelist" approach did, in fact, end up excluding
huge portions of the database from being checked, that is, everything
inside of tablespaces.

This doesn't exactly change my opinion regarding this discussion and I'd
rather we revert the "whitelist" patch and use the very minimal patch
from here:

https://www.postgresql.org/message-id/20181012013918.GA30064%40paquier.xyz

Thanks!

Stephen

Commits

  1. Fix various checksum check problems for pg_verify_checksums and base backups

  2. Switch pg_verify_checksums back to a blacklist