Thanks Clark,

I’m pretty sure the UTs would conflict with each other so raising the 
concurrency is probably a big undertaking. Though it’s definitely worth looking 
into in the future. Shifting the job timeout a little bit maybe a the simplest 
solution.

Thanks again,
Andras

_____________________________
From: Clark Boylan <[email protected]>
Sent: Tuesday, May 1, 2018 6:12 PM
Subject: Re: [openstack-dev] [mistral] Help with test run
To: <[email protected]>


On Fri, Apr 27, 2018, at 2:22 AM, András Kövi wrote:
> Hi,
>
> Can someone please help me with why this build ended with TIMED_OUT?
> http://logs.openstack.org/85/527085/8/check/mistral-tox-unit-mysql/3ffae9f/

Reading the job log the job setup only took a few minutes. Then the unittests 
start and are running continuously until the timeout happens at 30 minutes. 
Chances are that the default 30 minute timeout is not sufficient for this job. 
Runtime may vary based on cloud region and presence of noisy neighbors.

As for making this more reliable you can increase the timeout in the job 
configuration for that job. Another approach would be to make the unittests run 
more quickly. I notice the job is hard coded to use concurrency=1 when invoking 
the test runner so you are only using ~1/8 of the available cpus. You might try 
increasing this value though will likely need to make sure the tests don't 
conflict with each other.

Clark

__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: [email protected]?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: [email protected]?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to