Re: pgsql: Allow multiple xacts during table sync in logical replication.

On Fri, Feb 12, 2021 at 9:00 AM Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
> On Fri, Feb 12, 2021 at 8:47 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> >
> > Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> writes:
> > > I noticed one failure:
> > > https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=thorntail&dt=2021-02-12%2002%3A28%3A12
> > > Checking the same.
> >
> > I think thorntail runs with nondefault prevailing wal_level, might be
> > related.
> >
>
> Right, it is due to max_wal_senders nondefault config parameter. I
> think we need to slightly tweak the test so that we don't initially
> try to connect, then enable the subscription later with different
> command and then perform the rest of the commands.
>

I have pushed the fix for this and waiting for new results of
'thorntail'. In the meantime, I see another failure:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=walleye&dt=2021-02-12%2003%3A15%3A43

select count(*) > 0 as ok from pg_ls_waldir();
- ok
-----
- t
-(1 row)
-
+ERROR: could not stat file "pg_wal/000000010000000000000072":
Permission denied

I see the exact same failure 66 days ago on the same machine:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=walleye&dt=2020-12-07%2016%3A15%3A46

This doesn't look to be related to this commit and I am not sure why
it is failing occasionally, maybe due to some machine issues?

--
With Regards,
Amit Kapila.