Re: "pg_ctl: the PID file ... is empty" at end of make check

2020-11-13 Thread Tom Lane
Thomas Munro writes: > My bug report got an automated-looking message telling me to retest in > Big Sur beta 6 back in September, and I've just now upgraded an old > x86 Mac to Big Sur 11.01 and I can no longer reproduce the problem, so > it looks like it was fixed! Thanks, Apple. Hah! I've occ

Re: "pg_ctl: the PID file ... is empty" at end of make check

2020-11-13 Thread Thomas Munro
On Fri, Oct 18, 2019 at 2:26 PM Thomas Munro wrote: > On Fri, Oct 18, 2019 at 1:21 AM Tom Lane wrote: > > Thomas Munro writes: > > > On Tue, Oct 15, 2019 at 1:55 PM Tom Lane wrote: > > >> and now prairiedog has shown it too: > > >> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=prairie

Re: "pg_ctl: the PID file ... is empty" at end of make check

2019-10-17 Thread Thomas Munro
On Fri, Oct 18, 2019 at 1:21 AM Tom Lane wrote: > Thomas Munro writes: > > On Tue, Oct 15, 2019 at 1:55 PM Tom Lane wrote: > >> and now prairiedog has shown it too: > >> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=prairiedog&dt=2019-10-14%2021%3A45%3A47 > >> which is positively fasci

Re: "pg_ctl: the PID file ... is empty" at end of make check

2019-10-17 Thread Tom Lane
Thomas Munro writes: > On Tue, Oct 15, 2019 at 1:55 PM Tom Lane wrote: >> and now prairiedog has shown it too: >> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=prairiedog&dt=2019-10-14%2021%3A45%3A47 >> which is positively fascinating, because prairiedog is running a >> bronze-age versi

Re: "pg_ctl: the PID file ... is empty" at end of make check

2019-10-14 Thread Thomas Munro
On Tue, Oct 15, 2019 at 1:55 PM Tom Lane wrote: > Thomas Munro writes: > > Agreed. Secret non-shareable bug report filed. Fingers crossed. > > Since that conversation, longfin has shown the same symptom > just once more: > > longfin | REL_11_STABLE | 2019-07-28 22:29:03 | recoveryCheck | waiti

Re: "pg_ctl: the PID file ... is empty" at end of make check

2019-10-14 Thread Tom Lane
[ blast from the past dept. ] Thomas Munro writes: > On Thu, Nov 29, 2018 at 3:30 AM Tom Lane wrote: >> Thomas Munro writes: >>> https://github.com/macdice/unlinktest >> Bleah. But you can do better than ask whether it's a bug: you can >> quote POSIX: >> ... >> Not a lot of wiggle room there.

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-28 Thread Thomas Munro
On Thu, Nov 29, 2018 at 3:30 AM Tom Lane wrote: > Thomas Munro writes: > > On Wed, Nov 28, 2018 at 6:47 PM Tom Lane wrote: > >> Is it possible that unlink() on APFS is not atomic? > > > I think you might be right. > > https://github.com/macdice/unlinktest > > Bleah. But you can do better than a

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-28 Thread Tom Lane
Thomas Munro writes: > On Wed, Nov 28, 2018 at 6:47 PM Tom Lane wrote: >> Is it possible that unlink() on APFS is not atomic? > I think you might be right. > https://github.com/macdice/unlinktest Bleah. But you can do better than ask whether it's a bug: you can quote POSIX: The unlink() f

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-28 Thread Thomas Munro
On Wed, Nov 28, 2018 at 6:47 PM Tom Lane wrote: > Is it possible that unlink() on APFS is not atomic? That is, the > sequence of events is something like > > pg_ctl: open("postmaster.pid") > postmaster: unlink("postmaster.pid") > pg_ctl: reads file, gets zero

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Tom Lane
Thomas Munro writes: > On Wed, Nov 28, 2018 at 5:28 PM Tom Lane wrote: >> So my theory is we broke something in HEAD a couple weeks ago. But what? > Hmm. Not seeing it. I'm trying to do it again, with a make check loop. >> The fsync changes you made are suspiciously close to this issue (ie o

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Thomas Munro
On Wed, Nov 28, 2018 at 5:28 PM Tom Lane wrote: > Thomas Munro writes: > > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce > > it, but I noticed exactly the same thing on longfin the other day: > > https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Tom Lane
Thomas Munro writes: > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce > it, but I noticed exactly the same thing on longfin the other day: > https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%2005%3A39%3A04 I trawled the buildfarm logs and discover

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Thomas Munro
On Wed, Nov 28, 2018 at 4:10 PM Tom Lane wrote: > Thomas Munro writes: > > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce > > it, but I noticed exactly the same thing on longfin the other day: > > https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%

Re: "pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Tom Lane
Thomas Munro writes: > Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce > it, but I noticed exactly the same thing on longfin the other day: > https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%2005%3A39%3A04 > Anyone know what that's about? No :-(.

"pg_ctl: the PID file ... is empty" at end of make check

2018-11-27 Thread Thomas Munro
Hello, Today I saw a one-off case of $SUBJECT, on macOS. I can't reproduce it, but I noticed exactly the same thing on longfin the other day: https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=longfin&dt=2018-11-25%2005%3A39%3A04 Anyone know what that's about? -- Thomas Munro http://www.