Message330560
The bug is still alive:
https://buildbot.python.org/all/#/builders/168/builds/294
I found a bug using attached mp_debug.patch.
I wrote PR 10755 to try to fix it.
I have access to FreeBSD CURRENT buildbot. I only reproduced the bug once using:
./python -m test -v test_multiprocessing_fork -m WithThreadsTestPool --fail-env-changed -F
I ran this command 4 times in parallel in 4 terminals, but I failed to reliably reproduce the bug. Even if I stress the computer using "./python -m test -j4 -r" run in different shell, or using my "system_load.py 3" script, I fail to trigger the bug. |
|
Date |
User |
Action |
Args |
2018-11-27 23:38:37 | vstinner | set | recipients:
+ vstinner |
2018-11-27 23:38:37 | vstinner | set | messageid: <1543361917.77.0.788709270274.issue33676@psf.upfronthosting.co.za> |
2018-11-27 23:38:37 | vstinner | link | issue33676 messages |
2018-11-27 23:38:37 | vstinner | create | |
|