| *** coolsvap has joined #openstack-requirements | 03:00 | |
| *** markmcclain has quit IRC | 03:08 | |
| *** markmcclain has joined #openstack-requirements | 03:09 | |
| *** derekjhyang has joined #openstack-requirements | 03:15 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements: Updated from generate-constraints https://review.openstack.org/346615 | 06:49 |
|---|---|---|
| coolsvap | prometheanfire: around? | 13:36 |
| dirk | do we have a script for running the requirements meeting that we follow? | 14:03 |
| coolsvap | dirk: no script but we have detailed steps i | 14:06 |
| coolsvap | url is in topic | 14:06 |
| dirk | ah, topic! | 14:07 |
| dirk | that was too easy.. thanks :) | 14:08 |
| coolsvap | welcome :) | 14:08 |
| prometheanfire | coolsvap: ? | 14:35 |
| coolsvap | prometheanfire: good morning | 14:36 |
| prometheanfire | moin | 14:36 |
| coolsvap | so there were requests for python-ceilometerclient and oslo.context | 14:37 |
| coolsvap | merge | 14:37 |
| coolsvap | python-ceilometerclient looks good with generated u-c | 14:37 |
| prometheanfire | was going to hold off because of the xstatic changes, but it was just ordering | 14:45 |
| coolsvap | i was also looking for the horizon u-c patch | 14:50 |
| coolsvap | if there are any issues | 14:50 |
| *** number80 has quit IRC | 14:55 | |
| *** coolsvap is now known as coolsvap_ | 15:01 | |
| openstackgerrit | Michael Gugino proposed openstack/requirements: Restore pylxd https://review.openstack.org/347453 | 15:27 |
| *** number80 has joined #openstack-requirements | 15:45 | |
| tonyb | number80: why do you say pylxd shoudl be in 'other-requirements.txt' ? | 16:27 |
| number80 | tonyb: optional, LXD is not really necessary for nova | 16:28 |
| tonyb | number80: but it *is* needed for nova-lxd | 16:29 |
| number80 | yeah, but LXD is really an ubuntu-only thing | 16:29 |
| tonyb | number80: the nova solution for that is to add it to test-requirements | 16:29 |
| number80 | I'm not -1 as there's no consensus on that | 16:29 |
| tonyb | number80: Well they may be nehind it but it's a stackforge project and is subscribed to the requirements process so I don't think that's a valid reason to drop it | 16:30 |
| number80 | yeah, they should be tracked too | 16:30 |
| number80 | optional-requirements should feel a gap there, but it's not approved yet | 16:30 |
| number80 | s/feel/fill/ | 16:31 |
| tonyb | My opinionm is that we add pylxd back to requirements and ask them to add it to nova-lxd's test-requirements al python-libvirt | 16:31 |
| number80 | wfm | 16:31 |
| number80 | that raises an interesting topic for RPM packaging, how to blacklist some platform requirements | 16:32 |
| number80 | I don't see anyone maintaining LXD in RPM based currently, it requires some investment to make it work | 16:33 |
| tonyb | number80: What is the trigger for it to be added to RPM Packages? just the line in g-r? | 16:34 |
| number80 | tonyb: currently, we rely on requirements.txt and test-requirements.txt to generate dependencies | 16:34 |
| number80 | this should be handled at RPM packaging level | 16:34 |
| tonyb | number80: okay so if you don't process openstack/nova-lxd (which as you say is a canonical thing mostly) you wont see the requirement for pylxd. or did I miss something? | 16:35 |
| number80 | oh, it will be in nova-lxd? that works then | 16:35 |
| number80 | I though it would be in nova requirements for tests | 16:36 |
| tonyb | number80: Yeah not in openstack/nova. | 16:36 |
| number80 | then, go ahead for merging it :) | 16:36 |
| tonyb | number80: to be totally frank it'll probably *never* land in nova but that's no reason to make life hard for them | 16:36 |
| number80 | tonyb: I agree on that point | 16:37 |
| tonyb | number80: needs +2+W ;P | 16:37 |
| number80 | well, I don't have +2 but I'll change it into +1 to encourage someone else to do :) | 16:37 |
| tonyb | number80: :) | 16:38 |
| openstackgerrit | Merged openstack/requirements: Updated from generate-constraints https://review.openstack.org/346615 | 16:38 |
| prometheanfire | I've updated this a little bit for tomorrow's meeting https://etherpad.openstack.org/p/requirements-team-mascot | 18:27 |
| openstackgerrit | Tony Breeds proposed openstack/requirements: Check for missing run-time requirements https://review.openstack.org/333650 | 19:43 |
| prometheanfire | was the rebase needed?? | 19:45 |
| *** coolsvap_ has quit IRC | 20:01 | |
| tonyb | prometheanfire: probably not | 20:09 |
| tonyb | prometheanfire: I've added a 2nd +2 to https://review.openstack.org/#/c/335469 but leaned the +/-W to you | 21:46 |
| patchbot | tonyb: patch 335469 - requirements - Add scikit-learn, a library for machine learning. | 21:46 |
| prometheanfire | ok | 21:49 |
| sigmavirus | tonyb: prometheanfire re 335469, I think it's kind of silly to be concerned with an upstream vendoring given the myriad of existing openstack requirements that do that | 21:58 |
| tonyb | sigmavirus: I tend to agree but one of the things the requirements team shoudl do is help non openstack libraries improve. | 21:59 |
| sigmavirus | tonyb: libraries rarely vendor unless they have very good reasons, vendoring six is surprising but I suspect they didn't do it willy-nilly | 22:00 |
| sigmavirus | Respecting upstream authors and the idea that they're distributing something not only for OpenStack is something the OpenStack community rarely understands or handles well | 22:01 |
| tonyb | sigmavirus: Sure, which is why it's going in but it doesn't hurt us to kee a list of discussion we should have | 22:01 |
| tonyb | sigmavirus: it's possible that the reason is no longer valid | 22:01 |
| sigmavirus | tonyb: certainly is possible | 22:02 |
| sigmavirus | The above is more of a caution that the OpenStack development community has far more upstream enemies than friends because of how disrespectful they often are in discussions like these | 22:02 |
| tonyb | sigmavirus: Sure I'm not suggesting we go in there and say "we are openstack chnage or else", more like "we noticed that you're vendoring six, can you help us understand why?" and based on that discussion decide what the next steps are | 22:02 |
| sigmavirus | tonyb: in my experience it's always the latter followed by "Well that's stupid. We work on OpenStack and know better" | 22:03 |
| *** openstackgerrit has quit IRC | 22:03 | |
| *** openstackgerrit has joined #openstack-requirements | 22:04 | |
| prometheanfire | sigmavirus: it's just something I wanted to note | 22:04 |
| tonyb | sigmavirus: :( We have an obligation to stop / correct that where we see it | 22:05 |
| sigmavirus | tonyb: everytime I've tried I've been cursed out in private | 22:05 |
| prometheanfire | sigmavirus: knowing why would have been nice | 22:05 |
| prometheanfire | sigmavirus: that person should be shamed | 22:05 |
| sigmavirus | And this community's code of conduct is not enforced as I've seen multiple times over the last 2+ years | 22:05 |
| tonyb | sigmavirus: :( | 22:05 |
| tonyb | I think we've gone off topic, sigmavirus I hope you see that we're not going to do that thing. | 22:09 |
| openstackgerrit | Merged openstack/requirements: Restore pylxd https://review.openstack.org/347453 | 22:16 |
| *** openstackgerrit has quit IRC | 22:48 | |
| *** openstackgerrit has joined #openstack-requirements | 22:48 | |
| openstackgerrit | Merged openstack/requirements: Add scikit-learn, a library for machine learning. https://review.openstack.org/335469 | 23:11 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!