Re: BUG #17928: Standby fails to decode WAL on termination of primary

On Sat, Sep 23, 2023 at 11:35:05AM +1200, Thomas Munro wrote:
> Hmm, OK, tricky choices.  Anyway, I guess the next thing to straighten
> out in this area is the OOM policy.
>
> Pushed.  Thanks Alexander and everyone for the report, testing and
> help!  Time to watch the build farm to see if that Perl turns out to
> be robust enough...

Looks pretty stable to me.  I can see one thing in wrasse related to
pack():
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=wrasse&dt=2023-09-22%2023%3A23%3A23
Invalid type 'Q' in pack at t/039_end_of_wal.pl line 139.

I don't have the room to look at that now, perhaps a bit later today
depending on the flow of the wind.
--
Michael