Open Stack

Hi all,

we've been struggling with functional and fullstack jobs for a while
now and we'd like to bring a better stability to it as the failure rate
is still quite high. This is getting annoying, specifically for voting
functional job, which makes it sometimes harder to get some of our
beautiful precious patches merged.

The very basic pillar on the way to improve the jobs is to understand
what is wrong with them. Without having an overview of current failures
it's very hard to make any improvement.

Hence I'd like to ask you for help. When you see a failed job, don't
just 'recheck' with the best hope that your patch won't get hit by gate
failure again. Take a minute, please, open the failure and eventually
report a new bug.

Armando already sent similar email in the past [1], where he
encouraged to have a look at the failure and report a bug. I have also
sent a new policy to our docs with recommended steps [2]. Ideas on
improvement are very welcome!

It's really crucial to get a better grasp of our job failures and I
believe if we can make this a habit into our day to day work, it will
really make a difference at the job stability.

Thank you.
Jakub

[1] 
http://lists.openstack.org/pipermail/openstack-dev/2016-August/100801.html
[2] https://review.openstack.org/#/c/426829/