Horton

Re: Allow non-superuser to cancel superuser tasks.

x4mmm@yandex-team.ru

View thread View on postgr.es
From: "Andrey M. Borodin" <x4mmm@yandex-team.ru>
To: Nathan Bossart <nathandbossart@gmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, Kirill Reshke <reshkekirill@gmail.com>, "Leung, Anthony" <antholeu@amazon.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2024-06-14T07:06:36Z
Lists: pgsql-hackers 

> On 13 Jun 2024, at 02:04, Nathan Bossart <nathandbossart@gmail.com> wrote:
> 
> I adjusted 0001 based on my upthread feedback.

This patch looks good to me. Spellchecker is complaining about “signaling” instead of “signalling”, but ISTM it’s OK.

I’ve tried to dig into the test.
The problem is CV is allocated in

inj_state = GetNamedDSMSegment("injection_points”,

which seems to be destroyed in

shmem_exit() calling dsm_backend_shutdown()

This happens before we broadcast that sleep is over.
I think this might happen with any wait on injection point if it is pg_terminate_backend()ed.

Is there way to wake up from CV sleep before processing actual termination?

Thanks!


Best regards, Andrey Borodin.

Commits

  1. Look up backend type in pg_signal_backend() more cheaply.

  2. Add tap test for pg_signal_autovacuum role

  3. Introduce pg_signal_autovacuum_worker.

  4. Add a slot synchronization function.

  5. Ban role pg_signal_backend from more superuser backend types.