Re: tests against running server occasionally fail, postgres_fdw & tenk1

Hi,

On 2023-02-27 12:42:00 -0500, Tom Lane wrote:
> I wrote:
> > Hah - I thought of a solution. We can avoid this race condition if
> > we make the remote session itself inspect pg_stat_activity and
> > return its displayed application_name. Just need a foreign table
> > that maps onto pg_stat_activity.

Sounds reasonable. I guess you could also do it with a function that is
allowed to be pushed down. But given that you already solved it this way...

I think it's worth having an example for checks like this in the postgres_fdw
tests, even if it's perhaps not worth it for the application_name GUC on its
own. We saw that the GUC test copied the debug_discard_caches use of another
test...

> I went ahead and coded it that way, and it doesn't look too awful.
> Any objections?

Looks good to me.

I think it'd be an indication of a bug around the invalidation handling if the
terminations were required. So even leaving other things aside, I prefer this
version.

Greetings,

Andres Freund