Horton

Re: Improve behavior of concurrent TRUNCATE

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: "Bossart, Nathan" <bossartn@amazon.com>
Cc: Postgres hackers <pgsql-hackers@postgresql.org>
Date: 2018-08-09T16:30:02Z
Lists: pgsql-hackers

Attachments

On Thu, Aug 09, 2018 at 03:27:04PM +0000, Bossart, Nathan wrote:
> Thanks!  This patch builds cleanly, the new tests pass, and my manual
> testing hasn't uncovered any issues.  Notably, I cannot reproduce the
> originally reported authentication issue by using TRUNCATE after this
> change.  Beyond a few small comment wording suggestions below, it
> looks good to me.

Thanks, I have updated the patch as you suggested.  Any more
improvements to it that you can foresee?

> The second and fourth tests don't seem to actually block.

Yeah, that's intentional.
--
Michael

Commits

  1. Improve TRUNCATE by avoiding early lock queue