Horton

Re: stress test for parallel workers

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andrew Dunstan <andrew.dunstan@2ndquadrant.com>
Cc: mark@2ndquadrant.com, Thomas Munro <thomas.munro@gmail.com>, Justin Pryzby <pryzby@telsasoft.com>, Andres Freund <andres@anarazel.de>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2019-10-11T19:11:57Z
Lists: pgsql-hackers 
Andrew Dunstan <andrew.dunstan@2ndquadrant.com> writes:
> On 10/11/19 11:45 AM, Tom Lane wrote:
>> FWIW, I'm not excited about that as a permanent solution.  It requires
>> root privilege, and it affects the whole machine not only the buildfarm,
>> and making it persist across reboots is even more invasive.

> OK, but I'm not keen to have to tussle with coredumpctl. Right now our
> logic says: for every core file in the data directory try to get a
> backtrace. Use of systemd-coredump means that gets blown out of the
> water, and we no longer even have a simple test to see if our program
> caused a core dump.

I haven't played that much with this software, but it seems you can
do "coredumpctl list <path-to-executable>" to find out what it has
for a particular executable.  You would likely need a time-based
filter too (to avoid regurgitating previous runs' failures),
but that seems do-able.

			regards, tom lane

Commits

  1. In the postmaster, rely on the signal infrastructure to block signals.

  2. Paper over regression failures in infinite_recurse() on PPC64 Linux.

  3. Hack pg_ctl to report postmaster's exit status.

  4. Re-order some regression test scripts for more parallelism.