[issue40500] test_multiprocessing_fork leaks processes on PPC64LE RHEL8 LTO + PGO 3.x

[STINNER Victor]

STINNER Victor <vstinner at python.org> added the comment:

Another example "PPC64LE RHEL8 LTO 3.x":
https://buildbot.python.org/all/#/builders/356/builds/347

0:02:49 load avg: 9.54 [415/423/1] test_multiprocessing_spawn failed (env changed) (2 min 15 sec) -- running: test_concurrent_futures (58.1 sec), test_peg_generator (1 min 39 sec), test_gdb (47.8 sec)
Warning -- multiprocessing.Manager still has [<SpawnProcess name='SyncManager-140' pid=2395988 parent=2389351 started>, <SpawnProcess name='QueueManager-141' pid=2395999 parent=2389351 started>] active children after 5.1163996559916995 seconds
Warning -- Dangling processes: {<SpawnProcess name='QueueManager-141' pid=2395999 parent=2389351 started>}
Warning -- multiprocessing.Manager still has [<SpawnProcess name='SyncManager-143' pid=2396044 parent=2389351 started>, <SpawnProcess name='QueueManager-141' pid=2395999 parent=2389351 started>] active children after 5.115814491000492 seconds

----------

_______________________________________
Python tracker <report at bugs.python.org>
<https://bugs.python.org/issue40500>
_______________________________________