RecoveryIT failing ...

[Till Westmann <ti...@apache.org>]

Hi,

it seems that I have RecoveryIT consistently succeeding on my box and 
consistently failing on Jenkins [1][2].
Any hints what to look for from somebody who was in a similar state 
before?

Thanks,
Till

[1] 
https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integration-tests/309/
[2] 
https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integration-tests/311/

Re: RecoveryIT failing ...

[Till Westmann <ti...@apache.org>]

On 14 Aug 2016, at 1:17, abdullah alamoudi wrote:

> Clearly, this problem is caused by the first failure only. Subsequent
> failures are caused by the state left behind by the first test but you
> already know that.

Well, I kind of do, but I\u2019ve never had to deal with failures of the 
ITs
that happened on Jenkins but work on my machine. So far the failures I
produced were clearly mine and I could diagnose/fix them locally. :)

> I think the kill command failed to kill the processes and so I would 
> say
> this probably should be looked at by Mike Blow :)

But the kill commands don\u2019t really explain why the tests fail in the
first place. Or do they (killing certainly could be different between
OS X and Linux)?

Anyway, what I\u2019m looking for are hints on the tools and way to debug
these kind of failures (and one of the reasons for asking on the list
is to share the responses with everybody else).

Cheers,
Till

> On Sat, Aug 13, 2016 at 9:06 PM, Till Westmann <ti...@apache.org> 
> wrote:
>
>> Hi,
>>
>> it seems that I have RecoveryIT consistently succeeding on my box and
>> consistently failing on Jenkins [1][2].
>> Any hints what to look for from somebody who was in a similar state 
>> before?
>>
>> Thanks,
>> Till
>>
>> [1] https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integ
>> ration-tests/309/
>> [2] https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integ
>> ration-tests/311/
>>

Re: RecoveryIT failing ...

[abdullah alamoudi <ba...@gmail.com>]

Clearly, this problem is caused by the first failure only. Subsequent
failures are caused by the state left behind by the first test but you
already know that.
I think the kill command failed to kill the processes and so I would say
this probably should be looked at by Mike Blow :)

Cheers,
Abdullah.

On Sat, Aug 13, 2016 at 9:06 PM, Till Westmann <ti...@apache.org> wrote:

> Hi,
>
> it seems that I have RecoveryIT consistently succeeding on my box and
> consistently failing on Jenkins [1][2].
> Any hints what to look for from somebody who was in a similar state before?
>
> Thanks,
> Till
>
> [1] https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integ
> ration-tests/309/
> [2] https://asterix-jenkins.ics.uci.edu/job/asterix-gerrit-integ
> ration-tests/311/
>