Re: SIGSEGV from START_REPLICATION 0/XXXXXXX in XLogSendPhysical () at walsender.c:2762

At Thu, 28 May 2020 16:22:33 +0900, Michael Paquier <michael(at)paquier(dot)xyz> wrote in
> On Thu, May 28, 2020 at 09:07:04AM +0300, Vladimir Sitnikov wrote:
> > The CI history shows that HEAD was good at 11 May 13:27 UTC, and it became
> > bad by 19 May 14:00 UTC,
> > so the regression was introduced somewhere in-between.
> >
> > Does that ring any bells?
>
> It does, thanks! This would map with 1d374302 or 850196b6 that
> reworked this area of the code, so it seems like we are not quite done
> with this work yet. Do you still see the problem as of 55ca50d
> (today's latest HEAD)?

I think that's not the case. I think I cause this crash with the
HEAD. I'll post a fix soon.

> Also, just wondering.. If I use more or less the same commands as
> your travis job I should be able to reproduce the problem with a fresh
> JDBC repository, right? Or do you a sub-portion of your regression
> tests to run that easily?

Pgjdbc seems initiating physical replication on a logical replication
session.

regards.

--
Kyotaro Horiguchi
NTT Open Source Software Center