We've been assuming that all the "timedout" failures on BF member
fruitcrow were due to some wonkiness in the GNU/Hurd platform.
I got suspicious about that though after noticing that there are
a small number of such failures on other animals, eg [1][2][3].
In each case, the failure message claims it waited a good long
time, which is at variance with the actually observed runtime.
For instance [1] says "timed out after 14400 secs", but the
actual total test runtime is only 01:24:28 according to the
summary at the top of the page.
Looking into the buildfarm client, I realized that it's assuming that
"sleep($wait_time)" is sufficient to wait for $wait_time seconds.
However, the Perl docs point out that sleep() can be interrupted by a
signal. So now I'm suspicious that many of these failures are caused
by a stray signal waking up the wait_timeout thread prematurely.
GNU/Hurd might just be more prone to that than other platforms.
I propose the attached patch to the BF client to try to make this
more robust.
regards, tom lane
[1] https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=ovenbird&dt=2025-11-14%2009%3A21%3A05
[2] https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=conchuela&dt=2025-10-17%2018%3A32%3A07
[3] https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=opaleye&dt=2026-01-08%2023%3A07%3A37
--- run_build.pl.orig 2025-11-25 07:47:25
+++ run_build.pl 2026-01-09 15:02:23
@@ -3415,7 +3415,13 @@ sub wait_timeout
$SIG{$sig} = 'DEFAULT';
}
$SIG{'TERM'} = \&silent_terminate;
- sleep($wait_time);
+ # loop to absorb any unexpected signals without dying early
+ my $end_time = time + $wait_time;
+ while (time < $end_time)
+ {
+ my $delay = $end_time - time;
+ sleep($delay);
+ }
print STDERR "Run timed out, terminating.\n";
my $sig = $usig[0] || 'TERM';
kill $sig, $main_pid;
