Buildfarm misses running some contrib TAP tests
I noticed by accident that my buildfarm animal indri stopped running
certain contrib TAP tests this morning. In [1]https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=indri&dt=2026-04-07%2013%3A42%3A46 it's running
contrib-amcheck-check (00:00:12)
contrib-auto_explain-check (00:00:02)
contrib-basebackup_to_shell-check (00:00:01)
contrib-bloom-check (00:00:05)
contrib-oid2name-check (00:00:01)
contrib-pg_stat_statements-check (00:00:02)
contrib-postgres_fdw-check (00:00:07)
contrib-test_decoding-check (00:00:07)
contrib-vacuumlo-check (00:00:01)
but in the very next run [2]https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=indri&dt=2026-04-07%2014%3A17%3A01 we see
contrib-amcheck-check (00:00:12)
contrib-auto_explain-check (00:00:01)
contrib-basebackup_to_shell-check (00:00:02)
contrib-bloom-check (00:00:05)
contrib-oid2name-check (00:00:00)
contrib-pg_stash_advice-check (00:00:03)
contrib-pg_stat_statements-check (00:00:02)
contrib-pg_visibility-check (00:00:03)
contrib-test_decoding-check (00:00:08)
What became of postgres_fdw and vacuumlo? And why wasn't
pg_visibility being run before? And why are dblink and pg_prewarm
visible in neither list? Apparently the addition of pg_stash_advice
changed something here, but what?
Looking at some other BF animals shows that it's not just indri: other
autoconf-based animals are showing misbehavior of this sort as well.
I poked at this by adding some debug printouts, and determined that
what is going wrong is the test to see if we built the module:
# can't test it if we haven't built it
next unless scalar glob("$testdir/*.o $testdir/*.obj");
It's failing, sometimes, on modules that definitely do contain object
files. We've had run-ins with "scalar glob()" before [3]/messages/by-id/200085.1741127335@sss.pgh.pa.us, and when
I finally looked at "man perlfunc" what I read is
glob EXPR
glob
In list context, returns a (possibly empty) list of filename
expansions on the value of EXPR such as the standard Unix shell
/bin/csh would do. In scalar context, glob iterates through such
filename expansions, returning undef when the list is exhausted.
I'm not entirely sure what "iterates" means in this context, but
what seems to be happening on my Linux box is that you get undef
unless there is exactly one file matching the glob pattern.
I can't explain why we're not seeing more consistent behavior
out of the buildfarm, like never running postgres_fdw at all.
I wonder if the glob() infrastructure has some buggy internal state.
But the attached patch gives me better results.
regards, tom lane
[1]: https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=indri&dt=2026-04-07%2013%3A42%3A46
[2]: https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=indri&dt=2026-04-07%2014%3A17%3A01
[3]: /messages/by-id/200085.1741127335@sss.pgh.pa.us
Attachments:
misctests.patchtext/x-diff; charset=us-ascii; name=misctests.patchDownload+2-1
I wrote:
I'm not entirely sure what "iterates" means in this context, but
what seems to be happening on my Linux box is that you get undef
unless there is exactly one file matching the glob pattern.
I can't explain why we're not seeing more consistent behavior
out of the buildfarm, like never running postgres_fdw at all.
I wonder if the glob() infrastructure has some buggy internal state.
Oh ... apparently, what it thinks that means is that successive
calls will return file names out of the previous "scalar glob"
call, until it finally returns undef, and then the next call
starts a new scan. This simple test program:
use strict;
use warnings;
my $pgsql = '/home/postgres/pgsql';
foreach my $testdir (glob("$pgsql/contrib/*"))
{
next unless -d "$testdir/t";
print "examining $testdir\n";
my @gresult = glob("$testdir/*.o $testdir/*.obj");
print 'sizeof glob = ' . scalar @gresult . "\n";
# can't test it if we haven't built it
my $scal = scalar glob("$testdir/*.o $testdir/*.obj");
$scal = '<undefined>' if not defined($scal);
print 'scalar glob = ' . $scal . "\n";
}
1;
gives me
examining /home/postgres/pgsql/contrib/amcheck
sizeof glob = 4
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_common.o
examining /home/postgres/pgsql/contrib/auto_explain
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_gin.o
examining /home/postgres/pgsql/contrib/basebackup_to_shell
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_heapam.o
examining /home/postgres/pgsql/contrib/bloom
sizeof glob = 6
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_nbtree.o
examining /home/postgres/pgsql/contrib/dblink
sizeof glob = 1
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/oid2name
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/oid2name/oid2name.o
examining /home/postgres/pgsql/contrib/pg_prewarm
sizeof glob = 2
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/pg_stash_advice
sizeof glob = 3
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/pg_stash_advice.o
examining /home/postgres/pgsql/contrib/pg_stat_statements
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/stashfuncs.o
examining /home/postgres/pgsql/contrib/pg_visibility
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/stashpersist.o
examining /home/postgres/pgsql/contrib/postgres_fdw
sizeof glob = 5
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/sepgsql
sizeof glob = 0
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/test_decoding
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/test_decoding/test_decoding.o
examining /home/postgres/pgsql/contrib/vacuumlo
sizeof glob = 1
scalar glob = <undefined>
So we are getting results that depend mainly on how many .o files
there were in some previous contrib directory. That explains how
come pg_stash_advice managed to change the behavior of later
modules.
regards, tom lane
On 2026-04-07 Tu 7:46 PM, Tom Lane wrote:
I wrote:
I'm not entirely sure what "iterates" means in this context, but
what seems to be happening on my Linux box is that you get undef
unless there is exactly one file matching the glob pattern.
I can't explain why we're not seeing more consistent behavior
out of the buildfarm, like never running postgres_fdw at all.
I wonder if the glob() infrastructure has some buggy internal state.Oh ... apparently, what it thinks that means is that successive
calls will return file names out of the previous "scalar glob"
call, until it finally returns undef, and then the next call
starts a new scan. This simple test program:use strict;
use warnings;my $pgsql = '/home/postgres/pgsql';
foreach my $testdir (glob("$pgsql/contrib/*"))
{
next unless -d "$testdir/t";
print "examining $testdir\n";my @gresult = glob("$testdir/*.o $testdir/*.obj");
print 'sizeof glob = ' . scalar @gresult . "\n";# can't test it if we haven't built it
my $scal = scalar glob("$testdir/*.o $testdir/*.obj");
$scal = '<undefined>' if not defined($scal);
print 'scalar glob = ' . $scal . "\n";
}1;
gives me
examining /home/postgres/pgsql/contrib/amcheck
sizeof glob = 4
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_common.o
examining /home/postgres/pgsql/contrib/auto_explain
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_gin.o
examining /home/postgres/pgsql/contrib/basebackup_to_shell
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_heapam.o
examining /home/postgres/pgsql/contrib/bloom
sizeof glob = 6
scalar glob = /home/postgres/pgsql/contrib/amcheck/verify_nbtree.o
examining /home/postgres/pgsql/contrib/dblink
sizeof glob = 1
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/oid2name
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/oid2name/oid2name.o
examining /home/postgres/pgsql/contrib/pg_prewarm
sizeof glob = 2
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/pg_stash_advice
sizeof glob = 3
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/pg_stash_advice.o
examining /home/postgres/pgsql/contrib/pg_stat_statements
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/stashfuncs.o
examining /home/postgres/pgsql/contrib/pg_visibility
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/pg_stash_advice/stashpersist.o
examining /home/postgres/pgsql/contrib/postgres_fdw
sizeof glob = 5
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/sepgsql
sizeof glob = 0
scalar glob = <undefined>
examining /home/postgres/pgsql/contrib/test_decoding
sizeof glob = 1
scalar glob = /home/postgres/pgsql/contrib/test_decoding/test_decoding.o
examining /home/postgres/pgsql/contrib/vacuumlo
sizeof glob = 1
scalar glob = <undefined>So we are getting results that depend mainly on how many .o files
there were in some previous contrib directory. That explains how
come pg_stash_advice managed to change the behavior of later
modules.
Ugh. That's slightly embarrassing. So I guess the solution is not to use
glob in scalar context here.
Will fix.
cheers
andrew
--
Andrew Dunstan
EDB:https://www.enterprisedb.com