On 06/19/2012 12:51 PM, Danek Duvall wrote:
Rich Burridge wrote:
/net/stard.us.oracle.com/tank/ws/UL/7177355/components/python/python27/test-trans.txt
So the test_multiprocessing failure is new:
test_multiprocessing
Process Process-17:
Traceback (most recent call last):
File ".../Python-2.7.3/Lib/multiprocessing/process.py", line 258, in
_bootstrap
self.run()
File ".../Python-2.7.3/Lib/multiprocessing/process.py", line 114, in run
self._target(*self._args, **self._kwargs)
File ".../Python-2.7.3/Lib/test/test_multiprocessing.py", line 1558, in
_writefd
os.dup2(conn.fileno(), i)
OSError: [Errno 9] Bad file number
test test_multiprocessing failed -- Traceback (most recent call last):
File ".../Python-2.7.3/Lib/test/test_multiprocessing.py", line 1613, in
test_large_fd_transfer
self.assertEqual(f.read(), b"bar")
AssertionError: '' != 'bar'
That's a bit disconcerting, but that code is new, so it could just be a
problem with the test, and not with Python.
A quick google suggests it could be the same as:
http://mail.python.org/pipermail//python-bugs-list/2011-September/147715.html
Following the thread shows a potential patch at:
http://hg.python.org/cpython/rev/cd15473a9de2
which (if I'm reading things correctly), looks like it's been accepted.
I'll investigate more tomorrow and try to work out an equivalent patch
for us.
I'm not happy about the test_sys failure, either, though I suspect that may
have something to do with the dtrace patch.
The other failures simply say you need to run in verbose mode to know what
went wrong. Boo.
Given that there doesn't appear to be a regression here, I wouldn't
consider it a requirement to analyze these failures prior to putback, but
it'd be nice if you could do that, or at least file some bugs (maybe check
to make sure they're not already filed in the python bug tracker).
If I just fixup the test_multiprocessing one, is that enough for an
integration? (I can also open up other bug(s) to try to investigate the
other
failures if you like.)
Thanks.
Thanks,
Danek
_______________________________________________
userland-discuss mailing list
userland-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/userland-discuss
