[
https://issues.apache.org/jira/browse/PROTON-687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14144305#comment-14144305
]
Andrew Stitcher commented on PROTON-687:
----------------------------------------
At the point of writing to freed delivery context he python stack looks like:
2899 def __del__(self):
>2900 pn_delivery_set_context(self._dlv, pn_py2void(None))
2901
cleanup
1098 def cleanup(self):
1099 # release resources created by this class
1100 super(CreditTest, self).cleanup()
1101 self.c1 = None
>1102 self.snd = None
1103 self.c2 = None
1104 self.rcv2 = None
1105 self.snd2 = None
teardown
1107 def teardown(self):
>1108 self.cleanup()
1109
run
347 # stored with the exception info in the sys module
348 if opts.bare:
349 phase()
350 else:
351 try:
>352 phase()
353 except KeyboardInterrupt:
354 raise
355 except:
356 self.exception_phase_name = phase_name
357 self.exception = sys.exc_info()
_run
499 runner.run("test", test)
500
501 if hasattr(inst, "tearDown"):
502 runner.run("teardown", inst.tearDown)
503 elif hasattr(inst, "teardown"):
>504 runner.run("teardown", inst.teardown)
505
506 return runner
507
508 def __repr__(self):
509 return "MethodTest(%r, %r)" % (self.cls, self.method)
run_test
416 sys.stdout.write(output)
417 sys.stdout.flush()
418 interceptor.begin()
419 start = time.time()
420 try:
>421 runner = test()
422 finally:
423 interceptor.reset()
424 end = time.time()
425 if interceptor.dirty:
426 if interceptor.last != "\n":
> Memory corruption in proton-test
> --------------------------------
>
> Key: PROTON-687
> URL: https://issues.apache.org/jira/browse/PROTON-687
> Project: Qpid Proton
> Issue Type: Bug
> Components: proton-c, python-binding
> Affects Versions: 0.8
> Reporter: Cliff Jansen
> Assignee: Rafael H. Schloming
> Priority: Blocker
> Attachments: segv.patch
>
>
> proton-test will fail with memory violations. Deallocated memory is
> accessed, usually with benign results but less often on Windows.
> The attached crude patch allows the bug to be "discovered" on Linux as well.
> It is not clear to me if the problem is with the swig wrapper, or whether
> this just exposes a reference counting bug in proton-c.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
