Re: Autovacuum worker doesn't immediately exit on postmaster death

Greetings,

* Stephen Frost (sfrost(at)snowman(dot)net) wrote:
> * Michael Paquier (michael(at)paquier(dot)xyz) wrote:
> > On Mon, Mar 22, 2021 at 04:07:12PM -0400, Robert Haas wrote:
> > > On Mon, Mar 22, 2021 at 1:48 PM Stephen Frost <sfrost(at)snowman(dot)net> wrote:
> > >> Thanks for that. Attached is just a rebased version with a commit
> > >> message added. If there aren't any other concerns, I'll commit this in
> > >> the next few days and back-patch it. When it comes to 12 and older,
> > >> does anyone want to opine about the wait event to use? I was thinking
> > >> PG_WAIT_TIMEOUT or WAIT_EVENT_PG_SLEEP ...
> > >
> > > I'm not sure if we should back-patch this, but I think if you do you
> > > should just add a wait event, rather than using a generic one.
> >
> > I would not back-patch that either, as this is an improvement of the
> > current state. I agree that this had better introduce a new wait
> > event. Even if this stuff gets backpatched, you won't introduce an
> > ABI incompatibility with a new event as long as you add the new event
> > at the end of the existing enum lists, but let's keep the wait events
> > ordered on HEAD.
>
> Adding CFI's in places that really should have them is something we
> certainly have back-patched in the past, and that's just 'an improvement
> of the current state' too, so I don't quite follow the argument being
> made here that this shouldn't be back-patched.
>
> I don't have any problem with adding into the older releases, at the end
> of the existing lists, the same wait event that exists in 13+ for this
> already.
>
> Any other thoughts on this, particularly about back-patching or not..?

We seem to be at a bit of an impasse on this regarding back-patching,
which seems unfortunate to me, but without someone else commenting it
seems like it's stalled.

I'll go ahead and push the change to HEAD soon, as there doesn't seem to
be any contention regarding that.

Thanks,

Stephen