Re: Unexpected changes of CurrentResourceOwner and CurrentMemoryContext

Álvaro Herrera <alvherre(at)kurilemu(dot)de> wrote:

> On 2025-Sep-12, Antonin Houska wrote:
>
> > Euler Taveira <euler(at)eulerto(dot)com> wrote:
> >
> > > Interesting. I'm wondering that if this patch is applied we could remove the
> > > following code [...] from pg_logical_slot_get_changes_guts and
> > > LogicalSlotAdvanceAndCheckSnapState functions too. IIUC the referred
> > > code is a band-aid that will be improved someday.
> >
> > Even though we're fixing the likely reason of this problem, we cannot be 100%
> > sure that no other problem like this still exists. So I'd not remove this
> > assignment. Maybe add Assert(CurrentResourceOwner == old_resowner) in front of
> > that, and adjust the comment?
>
> Yeah, I'm going to risk removing it, because if we don't do it now,
> we're never going to do it. We can mitigate the risk of missing
> remaining bugs by having that assertion you suggest, so that if anyone
> actually hits a problem here, we'll know soon enough.
>
> I have pushed it with that change. I'll also add an open item for pg19
> so that we remember to come back to remove the assertions, if we feel we
> no longer need them.

I was worried about removing those workarounds because it was not trivial to
diagnose the issue. But it should be ok for the master branch. Thanks.

--
Antonin Houska
Web: https://www.cybertec-postgresql.com