You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@qpid.apache.org by Keith W <ke...@gmail.com> on 2012/08/15 12:20:29 UTC
Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Hi all
I notice that the Java Client/CPP Broker job on Jenkins began to fail
(timeout) 5 days ago and has timed out in the same place
(AddressBasedDestinationTest) on every build since . Normally the job
completes successfully. The commit that triggered the build suggests that
this was caused by a CPP Broker change.
The first failure was:
https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-Java-Cpp-Test/902/
HTH, Keith
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Rajith Attapattu <ra...@gmail.com>.
Gordon if the original commit was included in RC3, then we should
backport your fix as well.
Rajith
On Wed, Aug 15, 2012 at 11:11 AM, Gordon Sim <gs...@redhat.com> wrote:
> On 08/15/2012 12:07 PM, Gordon Sim wrote:
>>
>> On 08/15/2012 11:20 AM, Keith W wrote:
>>>
>>> I notice that the Java Client/CPP Broker job on Jenkins began to fail
>>> (timeout) 5 days ago and has timed out in the same place
>>> (AddressBasedDestinationTest) on every build since . Normally the job
>>> completes successfully. The commit that triggered the build suggests
>>> that
>>> this was caused by a CPP Broker change.
>>
>>
>> Very plausible! I will take a look...
>
>
> I've checked in a fix for the hang (bug causing no-local to be turned on
> when it wasn't supposed to be).
>
> I am however seeing a failure in
> org.apache.qpid.systest.disttest.controllerandclient.ControllerAndClientTest.testIteratingFeature
>
> But its not clear to me what the problem is. The controller reports missing
> responses:
>
> Testcase: testIteratingFeature took 10.93 sec
> Caused an ERROR
> After 10000ms ... Timed out waiting for command responses ... Expecting 2
> more responses.
> org.apache.qpid.disttest.DistributedTestException: After 10000ms ... Timed
> out waiting for command responses ... Expecting 2 more responses.
>
> and indeed I can see on one connection, command messages are sent out for
> CREATE_CONNECTION, CREATE_SESSION and CREATE_CONSUMER, but only a RESPONSE
> for the first is received back (and then forwarded on to the controller).
>
> I can't see any errors elsewhere, but perhaps I'm not looking in all the
> right places. Any ideas?
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
> For additional commands, e-mail: dev-help@qpid.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Keith Wall <ke...@gmail.com>.
We'll look into testIteratingFeature failure, it is actual a test testing a perf test framework, so should not be considered a blocked for the release. It has no relation to the cpp broker.
Thanks, Keith.
Sent from my iPhone
On 15 Aug 2012, at 16:29, Gordon Sim <gs...@redhat.com> wrote:
> testIteratingFeature
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Gordon Sim <gs...@redhat.com>.
On 08/15/2012 04:11 PM, Gordon Sim wrote:
> On 08/15/2012 12:07 PM, Gordon Sim wrote:
>> On 08/15/2012 11:20 AM, Keith W wrote:
>>> I notice that the Java Client/CPP Broker job on Jenkins began to fail
>>> (timeout) 5 days ago and has timed out in the same place
>>> (AddressBasedDestinationTest) on every build since . Normally the job
>>> completes successfully. The commit that triggered the build suggests
>>> that
>>> this was caused by a CPP Broker change.
>>
>> Very plausible! I will take a look...
>
> I've checked in a fix for the hang (bug causing no-local to be turned on
> when it wasn't supposed to be).
>
> I am however seeing a failure in
> org.apache.qpid.systest.disttest.controllerandclient.ControllerAndClientTest.testIteratingFeature
>
>
> But its not clear to me what the problem is. The controller reports
> missing responses:
>
> Testcase: testIteratingFeature took 10.93 sec
> Caused an ERROR
> After 10000ms ... Timed out waiting for command responses ... Expecting
> 2 more responses.
> org.apache.qpid.disttest.DistributedTestException: After 10000ms ...
> Timed out waiting for command responses ... Expecting 2 more responses.
>
> and indeed I can see on one connection, command messages are sent out
> for CREATE_CONNECTION, CREATE_SESSION and CREATE_CONSUMER, but only a
> RESPONSE for the first is received back (and then forwarded on to the
> controller).
>
> I can't see any errors elsewhere, but perhaps I'm not looking in all the
> right places. Any ideas?
Seems to pass (or get skipped?) under jenkins...
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Gordon Sim <gs...@redhat.com>.
On 08/15/2012 12:07 PM, Gordon Sim wrote:
> On 08/15/2012 11:20 AM, Keith W wrote:
>> I notice that the Java Client/CPP Broker job on Jenkins began to fail
>> (timeout) 5 days ago and has timed out in the same place
>> (AddressBasedDestinationTest) on every build since . Normally the job
>> completes successfully. The commit that triggered the build suggests
>> that
>> this was caused by a CPP Broker change.
>
> Very plausible! I will take a look...
I've checked in a fix for the hang (bug causing no-local to be turned on
when it wasn't supposed to be).
I am however seeing a failure in
org.apache.qpid.systest.disttest.controllerandclient.ControllerAndClientTest.testIteratingFeature
But its not clear to me what the problem is. The controller reports
missing responses:
Testcase: testIteratingFeature took 10.93 sec
Caused an ERROR
After 10000ms ... Timed out waiting for command responses ... Expecting
2 more responses.
org.apache.qpid.disttest.DistributedTestException: After 10000ms ...
Timed out waiting for command responses ... Expecting 2 more responses.
and indeed I can see on one connection, command messages are sent out
for CREATE_CONNECTION, CREATE_SESSION and CREATE_CONSUMER, but only a
RESPONSE for the first is received back (and then forwarded on to the
controller).
I can't see any errors elsewhere, but perhaps I'm not looking in all the
right places. Any ideas?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Gordon Sim <gs...@redhat.com>.
On 08/15/2012 11:20 AM, Keith W wrote:
> I notice that the Java Client/CPP Broker job on Jenkins began to fail
> (timeout) 5 days ago and has timed out in the same place
> (AddressBasedDestinationTest) on every build since . Normally the job
> completes successfully. The commit that triggered the build suggests that
> this was caused by a CPP Broker change.
Very plausible! I will take a look...
> The first failure was:
>
> https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-Java-Cpp-Test/902/
Thanks!
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org
Re: Qpid-Java-Cpp-Test failing (timing out) for last 5 days
Posted by Keith W <ke...@gmail.com>.
>> Did you guys receive any emails indicating the failures ?
>> For some reason I didn't. Maybe I accidentally deleted them.
> The job was timing out, rather than failing. I'll check the Jenkins
options and see how these are handled.
I've ticked the "fail the build" on timeout option on all our Jenkin jobs.
My excpectation is that this will cause a notification to be sent to
notifications@.. if the job ever times-out, as it did in this case and
should me we are all alerted sooner.