Horton

Re: Weird failure with latches in curculio on v15

Nathan Bossart <nathandbossart@gmail.com>

View thread View on postgr.es
From: Nathan Bossart <nathandbossart@gmail.com>
To: Andres Freund <andres@anarazel.de>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Thomas Munro <thomas.munro@gmail.com>, Fujii Masao <fujii@postgresql.org>, Michael Paquier <michael@paquier.xyz>, Postgres hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-02-01T22:35:55Z
Lists: pgsql-hackers

Attachments

On Wed, Feb 01, 2023 at 09:58:06AM -0800, Nathan Bossart wrote:
> On Wed, Feb 01, 2023 at 08:58:01AM -0800, Andres Freund wrote:
>> On 2023-02-01 10:12:26 -0500, Tom Lane wrote:
>>> The fundamental issue is that we have no good way to break out
>>> of system(), and I think the original idea was that
>>> in_restore_command would be set *only* for the duration of the
>>> system() call.  That's clearly been lost sight of completely,
>>> but maybe as a stopgap we could try to get back to that.
>> 
>> We could push the functions setting in_restore_command down into
>> ExecuteRecoveryCommand(). But I don't think that'd end up necessarily
>> being right either - we'd now use the mechanism in places we previously
>> didn't (cleanup/end commands).
> 
> Right, we'd only want to set it for restore_command.  I think that's
> doable.

Here is a first draft for the proposed stopgap fix.  If we want to proceed
with this, I can provide patches for the back branches.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com

Commits

  1. Avoid calling proc_exit() in processes forked by system().

  2. Move extra code out of the Pre/PostRestoreCommand() section.

  3. Revert refactoring of restore command code to shell_restore.c

  4. Refactor code in charge of running shell-based recovery commands

  5. Clean up inconsistent use of fflush().

  6. Report wait events for local shell commands like archive_command.