| openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements: Updated from generate-constraints https://review.openstack.org/320792 | 06:36 |
|---|---|---|
| coolsvap | we need to have a sync between the proposal bot and dhellmann update constraint bot/patches | 06:58 |
| *** jpena|off is now known as jpena | 07:29 | |
| jpena | nick jpena | 07:29 |
| openstackgerrit | Julien Danjou proposed openstack/requirements: Remove tooz from project list https://review.openstack.org/320824 | 08:13 |
| *** SarathMekala has joined #openstack-requirements | 08:24 | |
| *** SarathMekala has left #openstack-requirements | 08:24 | |
| *** SarathMekala has joined #openstack-requirements | 08:24 | |
| *** SarathMekala has left #openstack-requirements | 08:24 | |
| *** coolsvap is now known as coolsvap_ | 08:30 | |
| *** coolsvap_ is now known as coolsvap | 09:50 | |
| *** _amrith_ is now known as amrith | 10:40 | |
| *** openstackgerrit has quit IRC | 11:18 | |
| *** openstackgerrit has joined #openstack-requirements | 11:18 | |
| *** tonyb has left #openstack-requirements | 11:41 | |
| *** tonyb has joined #openstack-requirements | 11:41 | |
| * tonyb doesn't computer good :( | 11:42 | |
| dims | We are meeting in 3 mins in #openstack-meeting - Courtesy ping for : amrith coolsvap coreycb david-lyle dhellmann dirk dolphm GheRivero IgorYozhikov jpena jroll markmcclain masayukig mugsie notmorgan number80 prometheanfire sigmavirus24 toabctl tonyb tsufiev ttx | 11:58 |
| number80 | o/ | 11:58 |
| amrith | hello dims | 11:58 |
| tonyb | dims: Thanks | 11:58 |
| coolsvap | dims, ack. Thanks! | 11:59 |
| tonyb | dims: I'm on the "town hall" so I'll be distracted but present | 11:59 |
| dims | tonyb : will you be able to run it? :) | 11:59 |
| dims | ah ok | 11:59 |
| dims | no worries :) | 11:59 |
| tonyb | dims: I should be fine to run next week's if that helps you. | 12:00 |
| jroll | \o | 12:07 |
| * jroll moves to the right room | 12:07 | |
| jroll | is the courtesy ping just for everyone in this channel? :) | 12:08 |
| *** amrith is now known as _amrith_ | 12:30 | |
| *** jpena is now known as jpena|lunch | 12:42 | |
| dims | jroll : over some time, we can tune the list :) at the moment just pinging everyone here as the list is small | 12:56 |
| jroll | dims: yeah, just curious :) | 12:57 |
| dims | jroll : "if you are here, you are interested, if you are interested, we would like you to help us, so we'll ping you" :) | 12:57 |
| jroll | hehe | 12:58 |
| * coolsvap will be back later in evening | 12:59 | |
| sigmavirus24 | Sorry for missing the meeting this morning :( | 13:38 |
| * sigmavirus24 goes to find the logs | 13:38 | |
| *** _amrith_ is now known as amrith | 13:41 | |
| *** amrith is now known as _amrith_ | 13:43 | |
| *** _amrith_ is now known as amrith | 13:44 | |
| *** amrith is now known as _amrith_ | 13:45 | |
| *** _amrith_ is now known as amrith | 13:49 | |
| *** amrith is now known as _amrith_ | 13:49 | |
| *** _amrith_ is now known as amrith | 13:50 | |
| *** jpena|lunch is now known as jpena | 13:52 | |
| openstackgerrit | Merged openstack/requirements: update constraint for stevedore to new release 1.14.0 https://review.openstack.org/320704 | 14:55 |
| *** amrith is now known as _amrith_ | 15:01 | |
| *** _amrith_ is now known as amrith | 15:04 | |
| openstackgerrit | Merged openstack/requirements: update constraint for tooz to new release 1.37.0 https://review.openstack.org/320706 | 15:04 |
| *** amrith is now known as _amrith_ | 15:05 | |
| openstackgerrit | Merged openstack/requirements: update constraint for taskflow to new release 2.0.0 https://review.openstack.org/320705 | 15:06 |
| *** _amrith_ is now known as amrith | 15:09 | |
| dhellmann | dims, tonyb : I'm seeing a lot of conflicts in the constraint updates for the oslo releases. I'm trying to decide whether to squash those into 1 patch, or make it a series. Thoughts? | 16:02 |
| dims | dhellmann : either one is good | 16:04 |
| dhellmann | squashing will lead to the changes merging faster | 16:04 |
| dhellmann | I would just approve the bot update, except there's a comment about something in horizon breaking | 16:05 |
| dhellmann | https://review.openstack.org/#/c/320792/ | 16:05 |
| openstackgerrit | Merged openstack/requirements: update constraint for pbr to new release 1.10.0 https://review.openstack.org/320109 | 16:11 |
| dims | dhellmann : unrelated error, please go ahead | 16:11 |
| openstackgerrit | Doug Hellmann proposed openstack/requirements: update constraints for oslo releases 24 May 2016 https://review.openstack.org/320703 | 16:25 |
| dhellmann | dims, tonyb : squashed ^^ | 16:25 |
| dims | dhellmann : +2 will wait for at least one more +1 before +A'ing it | 16:42 |
| openstackgerrit | Merged openstack/requirements: update constraint for oslo.vmware to new release 2.7.0 https://review.openstack.org/320702 | 17:04 |
| *** jpena is now known as jpena|pto | 17:22 | |
| openstackgerrit | Doug Hellmann proposed openstack/requirements: update constraints for oslo releases 24 May 2016 https://review.openstack.org/320703 | 17:31 |
| dhellmann | dims : rebased ^^ | 17:31 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/requirements: Remove unused dcos dependency https://review.openstack.org/321203 | 19:46 |
| *** amrith is now known as _amrith_ | 20:05 | |
| openstackgerrit | Kyle Mestery proposed openstack/requirements: Update neutron-lib to 0.2.0 https://review.openstack.org/321228 | 20:40 |
| dhellmann | dims, tonyb : this is the compromise for gnocchiclient in liberty: https://review.openstack.org/#/c/321237/1 | 21:08 |
| dims | dhellmann : something else will break and jd__ / telemetry team will be still unhappy. i'd rather they figure out a way to disable constraints for their jobs... | 21:14 |
| dhellmann | dims : ceilometer follows g-r, so it's appropriate for us to constrain their dependencies | 21:14 |
| dhellmann | oh, hmm, maybe they don't | 21:15 |
| dhellmann | I really thought they did | 21:15 |
| dhellmann | I thought only gnocchi wasn't in that list | 21:15 |
| dims | i see only ceilometermiddleware and python-ceilometerclient in https://github.com/openstack/requirements/blob/stable/liberty/global-requirements.txt | 21:16 |
| dhellmann | dims : ceilometer is in projects.txt for stable/liberty | 21:17 |
| dhellmann | so the problem may come back for other branches, but this at least lets us fix liberty | 21:18 |
| dims | dhellmann : that's only used by the bot and the bot is switched off | 21:18 |
| dhellmann | right, but the point is that for liberty they were accepting syncs, and that means they can ask for a constraint, and this constraint fixes the issue | 21:18 |
| dims | dhellmann : they will end up breaking in another way a few days from now | 21:18 |
| dims | https://review.openstack.org/#/q/project:openstack/ceilometer+branch:stable/liberty+owner:%22OpenStack+Proposal+Bot%22 | 21:20 |
| dhellmann | that's possible. | 21:20 |
| dims | they haven't bothered to work the last bot proposed update | 21:20 |
| dims | dhellmann : which jobs break? ceilometer? | 21:22 |
| dhellmann | apparently, yes | 21:22 |
| dhellmann | ceilometer has a devstack plugin that installs gnocchi, and that fails because it's using an unconstrained gnocchiclient | 21:22 |
| dhellmann | the version it chooses wants cliff 2.0, which fails the constraint check | 21:22 |
| dhellmann | that's what led to jd__ raising that constraint. I proposed adding a constraint on gnocchiclient instead. | 21:23 |
| dims | https://review.openstack.org/#/q/project:openstack/ceilometer+branch:stable/liberty | 21:23 |
| dims | the one from may 23 seems fine | 21:23 |
| dhellmann | ok. that's all the information I have, I don't have a link to a log or anything. | 21:23 |
| *** _amrith_ is now known as amrith | 21:26 | |
| dims | ack dhellmann | 21:26 |
| dims | dhellmann : they can pip install whatever they want in their devstack plugin | 21:27 |
| dhellmann | well, they can't, because pip is constrained in devstack | 21:27 |
| dims | dhellmann : then they have to talk to devstack + infra | 21:27 |
| dims | that's what they have to work through for master or liberty | 21:27 |
| dims | they can't have the safety of constraints w/o bothering to be part of the process | 21:28 |
| dhellmann | they *are* part of the process for liberty | 21:28 |
| dims | they faced this problem already in master | 21:28 |
| dims | dhellmann : i can't find a log for ceilometer that fails with this cliff problem | 21:29 |
| dhellmann | ok, I'll have to get jd__ to provide that info, I took that part on faith | 21:29 |
| dims | dhellmann : yep. i understand | 21:29 |
| dims | if it is a project listed in projects.txt for liberty, then let's unblock. | 21:31 |
| dims | dhellmann : we are trying to make lemonade here with the whole g-r/u-c process. if folks don't like it, they should work with infra and whoever else to run jobs that are not subject to constraints. i am really sad to hear the constant stream of complaints | 21:37 |
| dims | w/o active participation or alternate proposals | 21:37 |
| dhellmann | dims : understood. I thought in this case it was an easy fix because that version of ceilometer was participating in g-r. I agree that for master we'd leave it up to them to resolve. | 21:38 |
| tonyb | dhellmann: Thanks. I'll take an action item to work with jd__, you, dims and infra to define a reaonable long term compromise. In otehr words work out how to disable constraints in some jobs and what that really means for the gate / this team. | 22:06 |
| dhellmann | tonyb : alternately we could try to convince the telemetry team to start syncing requirements again, but yeah. | 22:11 |
| dims | tonyb : Thanks! | 22:17 |
| tonyb | dhellmann: I think there's too much 'bad blood' for that to fly anytime soon. | 22:18 |
| dims | tonyb : we are facing the aftermath, but would like to keep sanity :) | 22:18 |
| tonyb | dhellmann: that would be a better solution and one I intend to chat to jd__ about in Barcelona. Given my lack of history I like to think we can find a way forward. | 22:19 |
| tonyb | dims: Yeah. If we make it optional in the gate there would be an explict "buyer be ware" clause. | 22:20 |
| tonyb | Also moving forward in ocata I have a crazy pants tool in my head (where it's perfect but useless) that will allow us to loosen the coupling between projects but still be co-installable ... basically 3.12 | 22:23 |
| tonyb | from the TODO list. | 22:24 |
| tonyb | It seems like it'd be fun to write :) | 22:24 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!