[openstack-dev] [tripleo] CI Squad Meeting Summary (week 7)
adarazs at redhat.com
Tue Feb 21 11:25:13 UTC 2017
On 02/17/2017 07:18 PM, Paul Belanger wrote:
> On Fri, Feb 17, 2017 at 03:39:44PM +0100, Attila Darazs wrote:
>> As always, if these topics interest you and you want to contribute to the
>> discussion, feel free to join the next meeting:
>> Time: Thursdays, 15:30-16:30 UTC
>> Place: https://bluejeans.com/4113567798/
>> Full minutes: https://etherpad.openstack.org/p/tripleo-ci-squad-meeting
> Was this meeting recorded in some manner? I see you are using bluejeans, but
> don't see any recordings of the discussion.
> Additionally, I am a little sad IRC is not being used for these meetings. Some
> of the things tripleo is doing is of interest of me, but I find it difficult to
> join a video session for 1hour just to listen. With IRC, it is easier for me to
> multi task into other things, then come back and review what has been discussed.
We are not recording it for now, sorry. We are tying to keep good
minutes & this summary to bridge the gap for the lack of recording or
We voted about it a few weeks ago and video meeting won. We did agree
about revisiting the IRC option after the transition is done as the bulk
of the meetings are mostly chats about possible technical solutions for
the quickstart transition rather than classic meeting where we decide
stuff. We're bringing those to the weekly TripleO IRC meetings.
>> * We discussed about the state of the Quickstart based update/upgrade jobs
>> upstream. matbu is working on them and the changes for the jobs are under
>> review. Sagi will help with adding project definitions upstream when the
>> changes are merged.
>> * John started to draft out the details of the CI related PTG sessions.
>> * A couple of us brought up reviews that they wanted merged. We discussed
>> the reasons, and agreed that sometimes an encouraging email to the mailing
>> list has the best effect to move important or slow-to-merge changes moving
>> * We talked quite a lot about log collection upstream. Currently Quickstart
>> doesn't collect logs exactly as upstream, and that might be okay, as we
>> collect more, and hopefully in a more easy to digest format.
>> * However we might collect too much, and finding the way around the logs is
>> not that easy. So John suggested to create an entry page in html for the
>> jobs that point to different possible places to find debug output.
> Yes, logging was something of an issue this week. We are still purging data on
> logs.o.o, but it does look like quickstart is too aggressive with log
> collection. We currently only have 12TB of HDD space for logs.o.o and our
> retention policy has dropped from 6 months to 6 weeks.
> I believe we are going have a discussion at the PTG about this for
> openstack-infra and implement some changes (caps) for jobs in the coming future.
> If you are planning on attending the PTG, I encourage you to attend the
I won't be on the PTG this time, but maybe Emilien or John can join.
With regards to space, we're going to comb through the logging and make
sure we're a bit more selective about what we gather.
>> * We also discussed adding back debug output to elastic search, as the
>> current console output doesn't contain everything, we log a lot of
>> deployment output in seperate log files in undercloud/home/stack/*.log
>> * Migration to the new Quickstart jobs will happen at or close to 10th of
>> March, in the beginning of the Pike cycle when the gates are still stable.
>> That was all for this week.
>> Best regards,
>>  https://etherpad.openstack.org/p/tripleo-ci-roadmap
>> OpenStack Development Mailing List (not for usage questions)
>> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
More information about the OpenStack-dev