9f00edc22888 has disabled a permutation due to failures in the CI for
FreeBSD environments, but this is a matter of timing. Let's document
properly why this type of permutation is a bad idea if relying on a wait
done in a SQL function, so as this can be avoided when implementing new
tests (this spec is also a template).
Reviewed-by: Bertrand Drouvot
Discussion: https://postgr.es/m/
[email protected]
step detach2 { SELECT injection_points_detach('injection-points-wait'); }
# Detach after wait and wakeup.
-# This permutation is proving to be unstable on FreeBSD, so disable for now.
+# Permutations like the following one commented out should be avoided, as
+# the detach may finish before the SQL function doing the wait returns
+# its result. It is recommended to use wakeups as the last permutation
+# should a wait be done within an SQL function.
#permutation wait1 wakeup2 detach2
# Detach before wakeup. s1 waits until wakeup, ignores the detach.