消息 [337092]
Sometimes, I can reproduce the crash using:
./python -m test --matchfile=bisect5 test_multiprocessing_spawn --fail-env-changed -F
Using this file:
test.test_multiprocessing_spawn.WithThreadsTestQueue.test_timeout
test.test_multiprocessing_spawn.WithProcessesTestBarrier.test_default_timeout
test.test_multiprocessing_spawn.WithThreadsTestManagerRestart.test_rapid_restart
test.test_multiprocessing_spawn.WithProcessesTestPool.test_release_task_refs
test.test_multiprocessing_spawn.WithManagerTestLock.test_rlock
It seems like the following test is enough to creates a coredump:
test.test_multiprocessing_spawn.WithThreadsTestManagerRestart.test_rapid_restart
Problem: it's really hard to write a *reliable* script/method to trigger the crash. This race condition is very well hidden! |
|
| 日期 |
用户 |
动作 |
参数 |
| 2019-03-04 11:30:12 | vstinner | 修改 | recipients:
+ vstinner, pitrou, lukasz.langa, koobs, pablogsal, scotchka |
| 2019-03-04 11:30:12 | vstinner | 修改 | messageid: <1551699012.65.0.964316273567.issue36114@roundup.psfhosted.org> |
| 2019-03-04 11:30:12 | vstinner | 链接 | issue36114 messages |
| 2019-03-04 11:30:12 | vstinner | 创建 | |
|