[Python-checkins] bpo-37244: Fix test_multiprocessing.test_resource_tracker() (GH-14288)
Victor Stinner
webhook-mailer at python.org
Tue Jun 25 18:30:21 EDT 2019
https://github.com/python/cpython/commit/e1a63c4f21011a3ae77dff624196561070c83446
commit: e1a63c4f21011a3ae77dff624196561070c83446
branch: master
author: Pierre Glaser <pierreglaser at msn.com>
committer: Victor Stinner <vstinner at redhat.com>
date: 2019-06-26T00:30:17+02:00
summary:
bpo-37244: Fix test_multiprocessing.test_resource_tracker() (GH-14288)
Increase robustness of test_resource_tracker(): retry for 60 seconds.
files:
M Lib/test/_test_multiprocessing.py
diff --git a/Lib/test/_test_multiprocessing.py b/Lib/test/_test_multiprocessing.py
index 07cf09aaa4ab..eef262d723da 100644
--- a/Lib/test/_test_multiprocessing.py
+++ b/Lib/test/_test_multiprocessing.py
@@ -5010,12 +5010,21 @@ def create_and_register_resource(rtype):
_resource_unlink(name1, rtype)
p.terminate()
p.wait()
- time.sleep(2.0)
- with self.assertRaises(OSError) as ctx:
- _resource_unlink(name2, rtype)
- # docs say it should be ENOENT, but OSX seems to give EINVAL
- self.assertIn(
- ctx.exception.errno, (errno.ENOENT, errno.EINVAL))
+
+ deadline = time.monotonic() + 60
+ while time.monotonic() < deadline:
+ time.sleep(.5)
+ try:
+ _resource_unlink(name2, rtype)
+ except OSError as e:
+ # docs say it should be ENOENT, but OSX seems to give
+ # EINVAL
+ self.assertIn(e.errno, (errno.ENOENT, errno.EINVAL))
+ break
+ else:
+ raise AssertionError(
+ f"A {rtype} resource was leaked after a process was "
+ f"abruptly terminated.")
err = p.stderr.read().decode('utf-8')
p.stderr.close()
expected = ('resource_tracker: There appear to be 2 leaked {} '
More information about the Python-checkins
mailing list