*** prometheanfire has quit IRC | 01:11 | |
*** spsurya has joined #openstack-requirements | 01:18 | |
*** prometheanfire has joined #openstack-requirements | 02:21 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements master: Updated from generate-constraints https://review.opendev.org/693653 | 06:23 |
---|---|---|
*** ccamacho has joined #openstack-requirements | 07:27 | |
*** ccamacho has quit IRC | 07:39 | |
*** tosky has joined #openstack-requirements | 08:20 | |
*** ccamacho has joined #openstack-requirements | 08:33 | |
*** ralonsoh has joined #openstack-requirements | 08:49 | |
*** brinzhang has joined #openstack-requirements | 09:15 | |
*** dtantsur|afk is now known as dtantsur | 09:33 | |
*** brinzhang has quit IRC | 10:01 | |
*** brinzhang has joined #openstack-requirements | 10:02 | |
*** brinzhang_ has joined #openstack-requirements | 10:07 | |
*** tosky_ has joined #openstack-requirements | 10:08 | |
*** tosky is now known as Guest94751 | 10:08 | |
*** tosky_ is now known as tosky | 10:08 | |
*** Guest94751 has quit IRC | 10:09 | |
*** brinzhang has quit IRC | 10:11 | |
*** hberaud|gone has quit IRC | 10:15 | |
*** brinzhang has joined #openstack-requirements | 10:55 | |
*** brinzhang has quit IRC | 10:55 | |
*** brinzhang_ has quit IRC | 10:58 | |
*** brinzhang has joined #openstack-requirements | 12:44 | |
*** brinzhang_ has joined #openstack-requirements | 14:04 | |
*** brinzhang has quit IRC | 14:07 | |
*** bnemec has joined #openstack-requirements | 14:29 | |
*** bnemec has quit IRC | 14:29 | |
*** e0ne has joined #openstack-requirements | 14:34 | |
openstackgerrit | Sean McGinnis proposed openstack/requirements master: Add new dependency for akamai API library https://review.opendev.org/692819 | 14:43 |
*** bnemec has joined #openstack-requirements | 14:49 | |
*** brinzhang has joined #openstack-requirements | 14:53 | |
*** brinzhang has quit IRC | 14:55 | |
*** brinzhang has joined #openstack-requirements | 14:55 | |
*** brinzhang_ has quit IRC | 14:56 | |
*** brinzhang has quit IRC | 14:56 | |
*** brinzhang has joined #openstack-requirements | 14:57 | |
*** brinzhang has quit IRC | 14:58 | |
*** brinzhang has joined #openstack-requirements | 14:58 | |
*** brinzhang has quit IRC | 14:59 | |
*** brinzhang has joined #openstack-requirements | 15:00 | |
*** brinzhang has quit IRC | 15:01 | |
*** brinzhang has joined #openstack-requirements | 15:01 | |
*** brinzhang has quit IRC | 15:02 | |
*** brinzhang has joined #openstack-requirements | 15:03 | |
*** brinzhang has quit IRC | 15:04 | |
*** brinzhang has joined #openstack-requirements | 15:05 | |
*** e0ne has quit IRC | 15:06 | |
*** e0ne has joined #openstack-requirements | 15:57 | |
*** e0ne has quit IRC | 16:01 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/requirements master: update constraint for oslo.messaging to new release 10.3.0 https://review.opendev.org/693723 | 16:14 |
*** bnemec has quit IRC | 16:20 | |
*** bnemec has joined #openstack-requirements | 16:47 | |
smcginnis | tosky: You're already doing a lot, so I hesitate to ask, but we could really use your help on some legacy requirements jobs. | 17:41 |
smcginnis | tosky: Particularly requirements-integration. I think that's going to become problematic very shortly. | 17:42 |
tosky | smcginnis: uh, what kind of help? | 17:42 |
smcginnis | I tried taking a stab at it, but was really just stumbling blindly. | 17:42 |
smcginnis | tosky: It has two major issues that I'm aware of - it's a legacy job, and it uses the deprecated devstack-gate. | 17:43 |
tosky | that's basically the same issue :) | 17:43 |
smcginnis | I'm not sure how to get it to do what we want as a v3 job. | 17:43 |
tosky | let me see - is it defined inside requirements.git or inside openstack-zuul-jobs? | 17:43 |
smcginnis | Heh, well that's good at least. See, you already fixed 50% of my issue. :D | 17:43 |
smcginnis | I believe it's all (or mostly) inside openstack/requirements. | 17:44 |
smcginnis | playbooks/requirements-integration/* | 17:44 |
smcginnis | fungi: If gunicorn is py3 only, should they be publishing a py2.py3 wheel? | 17:46 |
smcginnis | https://pypi.org/project/gunicorn/#files | 17:46 |
tosky | from a quick glance, it's basically deploying devstack and running a script | 17:46 |
smcginnis | Not sure if having that py2 in there is causing an issue. | 17:46 |
smcginnis | tosky: Yeah, I think it tries to spin up devstack with every possible thing to make sure requirements are coinstallable for everything. | 17:46 |
smcginnis | Though I'll admit my actual knowledge on it is low. It was there before I got involved in requirements. | 17:47 |
tosky | the script was first committed on Sep 2, 2014 | 17:48 |
smcginnis | Hah, probably about due for an update. | 17:51 |
openstackgerrit | Sean McGinnis proposed openstack/requirements master: Cap gunicorn for py27 https://review.opendev.org/693742 | 18:01 |
*** dtantsur is now known as dtantsur|afk | 18:20 | |
fungi | yeah, if it's py3-only then it should produce py3-only wheels, not universal | 18:46 |
smcginnis | fungi: Is that something that would be worth contacting the publisher to let them know. | 18:51 |
smcginnis | Not sure if it is the source of our recent problems with requirements, but a py2 job tries to pull in the py3 only version of that lib. | 18:51 |
fungi | i expect if our job doesn't differentiate then it would try to build a wheel from the sdist anyway | 18:53 |
smcginnis | I'll see if it works out to have the cap in the global-requirements file instead of just upper-constraints. | 18:55 |
smcginnis | prometheanfire: It looks like we may need https://review.opendev.org/693742, but I would like to better understand why. | 19:16 |
prometheanfire | not surprising, but knowing why would be nice | 19:19 |
smcginnis | prometheanfire: In the past, have these issues usually been from a transient dependency that doesn't get upper-constraints enforced? | 19:38 |
smcginnis | prometheanfire: Like, we have a dependency that has a dependency on gunicorn without any constraints defined in its requirements, and since we are not installing it directly, it doesn't take our upper-constraints into consideration. | 19:39 |
prometheanfire | | openstack/octavia | requirements.txt | 60 | gunicorn>=19.9.0 # MIT | | 20:38 |
prometheanfire | a few of the x/projects have it too | 20:39 |
smcginnis | That should still work with https://opendev.org/openstack/requirements/src/branch/master/upper-constraints.txt#L220 though. | 20:47 |
prometheanfire | yep, I'm not sure what's up | 20:53 |
smcginnis | Well, the cap in global-requirements passes and should hopefully handle it for anything that isn't being constrained by upper-constraints I guess. | 20:55 |
smcginnis | Won't be an issue soon since stable/train is capped at 19.9.0 and master should be dropping py2. | 20:55 |
smcginnis | But we need to get our requirements-integration job updated to not run under py27 too. | 20:56 |
prometheanfire | ya, I'm guessing something is not using constraints right | 20:57 |
prometheanfire | tonyb: you mind reviewing the last few reviews? | 20:57 |
prometheanfire | or dirk ? | 20:57 |
prometheanfire | mainly smcginnis's reviews | 20:58 |
prometheanfire | need a second | 20:58 |
smcginnis | Would be great to have a second set of eyes on all that. | 20:58 |
*** spsurya has quit IRC | 21:07 | |
*** ralonsoh has quit IRC | 23:44 | |
*** tosky has quit IRC | 23:44 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!