francolin just showed a non-reproducing failure in the new pgbench tests:

not ok 211 - transaction count for 001_pgbench_log_1.31583 (3)

#   Failed test 'transaction count for 001_pgbench_log_1.31583 (3)'
#   at t/ line 438.

Apparently, one of the threads ran 3 transactions where the test script
expects it to run at most 2.  Is this a pgbench bug, or is the test
being overoptimistic about how exact the "-T 2" cutoff is?

Probably both? It seems that cutting off on time is not a precise science,
so I suggest to accept 1, 2 and 3 lines, see attached.

diff --git a/src/bin/pgbench/t/ b/src/bin/pgbench/t/
index 3609b9b..1fe3433 100644
--- a/src/bin/pgbench/t/
+++ b/src/bin/pgbench/t/
@@ -463,7 +463,8 @@ pgbench(
 	'pgbench progress');
 # $nthreads threads, 2 seconds, sometimes only one aggregated line is written
-check_pgbench_logs('001_pgbench_log_1', $nthreads, 1, 2,
+# and sometimes 3 lines...
+check_pgbench_logs('001_pgbench_log_1', $nthreads, 1, 3,
 	qr{^\d+ \d{1,2} \d+ \d+ \d+ \d+ \d+ \d+ \d+ \d+ \d+$});
 # with sampling rate
Sent via pgsql-hackers mailing list (
To make changes to your subscription:

Reply via email to