Horton

Re: Online enabling of checksums

Andres Freund <andres@anarazel.de>

View thread View on postgr.es
From: Andres Freund <andres@anarazel.de>
To: Magnus Hagander <magnus@hagander.net>
Cc: Andrey Borodin <x4mmm@yandex-team.ru>, Tomas Vondra <tomas.vondra@2ndquadrant.com>, Heikki Linnakangas <hlinnaka@iki.fi>, Daniel Gustafsson <daniel@yesql.se>, Michael Banck <michael.banck@credativ.de>, Robert Haas <robertmhaas@gmail.com>, Greg Stark <stark@mit.edu>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2018-04-05T21:23:37Z
Lists: pgsql-hackers 
Hi,

On 2018-04-05 22:06:36 +0200, Magnus Hagander wrote:
> I have now pushed this latest version with some minor text adjustments and
> a catversion bump.

Is there any sort of locking that guarantees that worker processes see
an up2date value of
DataChecksumsNeedWrite()/ControlFile->data_checksum_version? Afaict
there's not. So you can afaict end up with checksums being computed by
the worker, but concurrent writes missing them.  The window is going to
be at most one missed checksum per process (as the unlocking of the page
is a barrier) and is probably not easy to hit, but that's dangerous
enough.

Just plonking a barrier into DataChecksumsNeedWrite() etc is a
possibility, but it's also not free...

Greetings,

Andres Freund

Commits

  1. Online enabling and disabling of data checksums

  2. Deactive flapping checksum isolation tests.

  3. Add support for coordinating record typmods among parallel workers.