[ 
https://issues.apache.org/jira/browse/PROTON-326?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13676849#comment-13676849
 ] 

Philip Harvey commented on PROTON-326:
--------------------------------------

Note that I would normally have captured a core dump to analyse where the tests 
were stuck, but this was on a bare-bones machine without gdb installed :-(
                
> proton-c ServerTest.testIdleTimeout sometimes fails, causing test suite to 
> hang forever
> ---------------------------------------------------------------------------------------
>
>                 Key: PROTON-326
>                 URL: https://issues.apache.org/jira/browse/PROTON-326
>             Project: Qpid Proton
>          Issue Type: Bug
>          Components: proton-c
>    Affects Versions: 0.4
>         Environment: Linux
>            Reporter: Philip Harvey
>
> I ran the python test suite against proton-c today and noticed that 
> ServerTest.testIdleTimeout failed with the following output:
> {noformat}
> proton_tests.engine.ServerTest.testIdleTimeout 
> ..........................ERROR amqp:resource-limit-exceeded 
> local-idle-timeout expired
>  fail
> Error during test:  Traceback (most recent call last):
>     File "./tests/python/proton-test", line 340, in run
>       phase()
>     File "/.../workspace/proton/tests/python/proton_tests/engine.py", line 
> 1412, in testIdleTimeout
>       assert self.conn.state == (Endpoint.LOCAL_ACTIVE | 
> Endpoint.REMOTE_ACTIVE), "Connection terminated"
>   AssertionError: Connection terminated
> {noformat}
> This test usually succeeds so I guess it contains a race condition.
> Something that is arguably more problematic is that this failure causes the 
> test suite to hang. The suite proceeds to run the other tests, but does not 
> terminate when it reaches the end. On Linux, I actually had to kill the 
> process rather than ctrl-c-ing it.  
> I cannot reproduce the test failure, but I do find that if I make it fail 
> (for example, by modifying the assert on line 1412 to always fail) then the 
> suite hangs every time.
> It's obviously undesirable for a failing test to cause the entire suite to 
> hang.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to