[openstack-dev] [tripleo][heat] Summit session clashes

Zane Bitter zbitter at redhat.com
Wed Apr 20 01:26:25 UTC 2016


On 19/04/16 18:04, Steve Baker wrote:
> On 19/04/16 20:29, Steven Hardy wrote:
>> On Tue, Apr 19, 2016 at 04:24:46PM +1200, Steve Baker wrote:
>>>     All of the TripleO design summit sessions are on Thursday afternoon in
>>>     slots which clash with Heat sessions. Heat is a core component of TripleO
>>>     and as a contributor to both projects I was rather hoping to attend as
>>>     many of both sessions as possible - I don't think I'm alone in this
>>>     desire.
>>>
>>>     Is it possible that some horse trading could take place to reduce the
>>>     clashes? Maybe TripleO sessions could move to Wednesday morning?
>> Yes I agree this is unfortunate.  I already queried the clashes wrt the
>> contributor meetups, and was told we can only adjust if we can find another
>> project willing to switch - I'm Open to negotiation if any other PTLs wish
>> to change sessions at this late stage.
>>
>> I see the current Heat schedule has SoftwareDeployment improvements [1] and
>> Issues with very large stacks [2] at non-conflicting times, which is good
>> as these are probably amongst the top priorities for TripleO (other than
>> performance improvements, which relates to very-large-stacks).
>>
>> One observation I would make is that Heat does have a pretty large number
>> of sessions (12 in total plus meetup), this is always going to present
>> challenges from a scheduling point of view - perhaps we can ask for a
>> volunteer or two (other than myself) from the Heat community who is willing
>> to cover at least the Upgrades fishbowl[3] and Composable Services
>> workroom[4] sessions if we can't resolve the conflicts.
>>
>> Thanks,
>>
>> Steve
>>
>> [1]https://www.openstack.org/summit/austin-2016/summit-schedule/events/9115
>> [2]https://www.openstack.org/summit/austin-2016/summit-schedule/events/9117
>> [3]https://www.openstack.org/summit/austin-2016/summit-schedule/events/9118
>> [4]https://www.openstack.org/summit/austin-2016/summit-schedule/events/9292
>
> I think we could reduce the topic overlap just by shuffling the heat
> sessions.
>
> Thomas, what do you think of the following?
> Swap work sessions "Release model and versioning" and "Validation
> improvements"
> https://www.openstack.org/summit/austin-2016/summit-schedule/events/9240
> https://www.openstack.org/summit/austin-2016/summit-schedule/events/9247
> This would let me attend the tripleo CI work session, and the release
> model session which I proposed ;)

I'm driving the validation improvements session, but I don't care about 
the conflicting TripleO session "Reducing the CI pain" so that's OK.

> Swap work sessions "Performance improvements" and "hot-parser"
> https://www.openstack.org/summit/austin-2016/summit-schedule/events/9236
> https://www.openstack.org/summit/austin-2016/summit-schedule/events/9248

I think this is a no-go because it would put the hot-parser session on 
at the same time as the Tacker/heat-translator joint session that will 
necessarily have a bunch of the same people in it.

> This will make the performance work session before the performance
> fishbowl, but I'm sure we could make that work. Its not like we need the
> fishbowl to solicit areas of improvement.

Umm, I thought that was the idea.

I need to be at both of those Heat ones anyway, so this doesn't really 
help me. I'd rather have the DLM session in this slot instead. (The only 
sessions I can really skip are the Release Model, Functional Tests and 
DLM.) That would give us:

           Heat                        TripleO

  Wed 3:30 Release Model
  Wed 4:30 HOT Parser
  Wed 5:20 Functional Tests

  Thu 1:30 DLM                         Upgrades
  Thu 2:20 Convergence switchover      Containers
  Thu 3:10 Convergence cleanup         Composable Roles
  Thu 4:10 Performance                 API
  Thu 5:00 Validation                  CI

I think that way Steve and I could probably both cover upgrades, and he 
could cover the rest.

I'd like to get to the composable roles and containers sessions too, but 
we'd have to rejig basically every Heat session and I think it's too 
late to be doing that.

cheers,
Zane.



More information about the OpenStack-dev mailing list