Re: basebackup: do not verify checksums on pages written before enabling checksums

From: Daniel Gustafsson <daniel(at)yesql(dot)se>
To: Bertrand Drouvot <bertranddrouvot(dot)pg(at)gmail(dot)com>
Cc: Zsolt Parragi <zsolt(dot)parragi(at)percona(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: basebackup: do not verify checksums on pages written before enabling checksums
Date: 2026-08-17 13:15:43
Message-ID: DAE694FC-B549-4A15-A8AB-C35768D3A50F@yesql.se
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> On 17 Aug 2026, at 14:54, Bertrand Drouvot <bertranddrouvot(dot)pg(at)gmail(dot)com> wrote:

> +$node->stop;
> +$node->corrupt_page_checksum($fcorrupt, 0);
> +$node->start;
>
> The test stops and restarts the node before taking the backup, which resets
> lastChecksumChangeRecPtr.
>
> Should we also test that verification resumes while lastChecksumChangeRecPtr is
> still set?

corrupt_page_checksum() only works reliably when the cluster is turned off, a
reliable corrupt-checksum mechanism for an online cluster could for sure be
done but seems a bit more invasive at this point. Or am I missing something
obvious?

--
Daniel Gustafsson

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Mihail Nikalayeu 2026-08-17 13:16:37 Re: Re-read conindid under the referenced table's lock in the RI fast path
Previous Message Zsolt Parragi 2026-08-17 13:11:17 pg_stat_database.checksum_failures misses single-page failures in backups