Monday, 2021-03-15

*** jamesmcarthur has quit IRC00:05
*** ianw has joined #openstack-infra00:11
*** tosky has quit IRC01:02
*** ociuhandu has joined #openstack-infra01:12
*** ociuhandu has quit IRC01:26
*** prometheanfire has quit IRC01:28
*** prometheanfire has joined #openstack-infra01:32
*** redrobot has quit IRC01:56
*** redrobot has joined #openstack-infra01:59
*** zzzeek has quit IRC02:09
*** zzzeek has joined #openstack-infra02:11
*** zzzeek has quit IRC02:36
*** zzzeek has joined #openstack-infra02:39
*** rcernin has quit IRC02:50
*** rcernin has joined #openstack-infra02:55
*** rcernin has quit IRC03:12
*** ricolin has quit IRC03:26
*** rcernin has joined #openstack-infra03:31
*** psachin has joined #openstack-infra03:33
*** rcernin has quit IRC03:34
*** rcernin has joined #openstack-infra03:34
*** dhill has quit IRC03:34
*** ricolin has joined #openstack-infra04:12
*** ykarel has joined #openstack-infra04:25
*** vishalmanchanda has joined #openstack-infra04:31
*** ricolin has quit IRC04:33
*** dasm has quit IRC04:45
*** dasm has joined #openstack-infra04:57
*** ykarel_ has joined #openstack-infra05:20
*** ykarel has quit IRC05:22
*** ykarel_ is now known as ykarel05:28
*** jamesdenton has quit IRC05:38
*** jamesdenton has joined #openstack-infra05:39
*** ramishra has joined #openstack-infra06:18
*** ociuhandu has joined #openstack-infra06:56
*** lpetrut has joined #openstack-infra07:03
*** ralonsoh has joined #openstack-infra07:13
*** rcernin has quit IRC07:17
*** eolivare has joined #openstack-infra07:41
*** ociuhandu has quit IRC07:44
*** ociuhandu has joined #openstack-infra07:46
*** ociuhandu has quit IRC07:51
*** dchen has quit IRC08:00
*** sboyron has joined #openstack-infra08:01
*** ociuhandu has joined #openstack-infra08:04
*** hashar has joined #openstack-infra08:08
*** jamesdenton has quit IRC08:13
*** jamesdenton has joined #openstack-infra08:15
*** andrewbonney has joined #openstack-infra08:16
*** ociuhandu has quit IRC08:17
*** priteau has joined #openstack-infra08:17
*** ociuhandu has joined #openstack-infra08:18
*** rpittau|afk is now known as rpittau08:22
*** ociuhandu has quit IRC08:23
*** jcapitao has joined #openstack-infra08:30
*** piotrowskim has joined #openstack-infra08:44
*** ociuhandu has joined #openstack-infra08:48
*** tosky has joined #openstack-infra08:53
*** lucasagomes has joined #openstack-infra08:57
*** jpena|off is now known as jpena08:58
*** ociuhandu has quit IRC08:59
*** ociuhandu has joined #openstack-infra08:59
*** derekh has joined #openstack-infra09:06
*** rcernin has joined #openstack-infra09:36
*** rcernin has quit IRC09:40
*** derekh has quit IRC09:57
*** derekh has joined #openstack-infra09:57
*** dciabrin_ is now known as dciabrin10:12
*** sboyron has quit IRC10:17
openstackgerritAurelien Lourot proposed openstack/project-config master: Add Magnum charms to OpenStack charms  https://review.opendev.org/c/openstack/project-config/+/78021110:17
*** sboyron has joined #openstack-infra10:18
*** derekh has quit IRC10:29
*** derekh has joined #openstack-infra10:29
*** dtantsur|afk is now known as dtantsur10:31
*** openstackgerrit has quit IRC10:43
*** jcap has joined #openstack-infra10:51
*** jcapitao has quit IRC10:51
*** jcap is now known as jcapitao10:51
*** hashar has quit IRC11:29
*** rlandy has joined #openstack-infra11:34
*** dhill has joined #openstack-infra11:58
*** jtomasek has joined #openstack-infra12:03
*** jcapitao is now known as jcapitao_lunch12:12
*** gshippey has joined #openstack-infra12:25
*** ykarel has quit IRC12:26
*** ykarel has joined #openstack-infra12:27
*** jtomasek has quit IRC12:38
*** jtomasek has joined #openstack-infra12:40
*** derekh has quit IRC12:50
*** ricolin_ has joined #openstack-infra12:50
*** ricolin_ has quit IRC12:50
*** ricolin has joined #openstack-infra12:51
*** derekh has joined #openstack-infra12:51
stephenfinclarkb: Looks like we've lost openstackbot from #openstack-oslo :( Could we get it restarted, please?12:52
*** hashar has joined #openstack-infra13:00
*** jpena is now known as jpena|lunch13:03
*** nweinber has joined #openstack-infra13:06
*** noonedeadpunk has joined #openstack-infra13:09
noonedeadpunkI guess that gerritbot is dead for a while `openstackgerrit has quit IRC` http://eavesdrop.openstack.org/irclogs/%23openstack-ansible/latest.log.html#t2021-03-15T10:43:3213:10
*** jamesdenton has quit IRC13:14
*** jamesdenton has joined #openstack-infra13:15
*** ricolin has quit IRC13:15
*** jcapitao_lunch is now known as jcapitao13:26
*** ociuhandu has quit IRC13:28
*** ociuhandu has joined #openstack-infra13:29
*** ociuhandu has quit IRC13:39
*** ociuhandu has joined #openstack-infra13:42
*** jcapitao has quit IRC13:43
*** ricolin has joined #openstack-infra13:45
*** smcginnis has joined #openstack-infra13:50
*** ociuhandu has quit IRC13:50
*** ociuhandu has joined #openstack-infra13:57
fungistephenfin: noonedeadpunk: thanks for the heads up, it gets confused in netslipts and the like. restarting it now13:59
*** jpena|lunch is now known as jpena14:01
*** ociuhandu has quit IRC14:02
*** ociuhandu has joined #openstack-infra14:03
fungiit's announced something in the glance channel a few minutes ago, so seems back on track: http://eavesdrop.openstack.org/irclogs/%23openstack-glance/latest.log.html#t2021-03-15T14:00:5814:03
*** jcapitao has joined #openstack-infra14:04
*** jcapitao has quit IRC14:06
*** jcapitao has joined #openstack-infra14:07
*** ociuhandu has quit IRC14:08
*** ociuhandu has joined #openstack-infra14:08
noonedeadpunkyep working14:11
*** openstackgerrit has joined #openstack-infra14:20
openstackgerritMerged openstack/project-config master: Add Magnum charms to OpenStack charms  https://review.opendev.org/c/openstack/project-config/+/78021114:20
stephenfinfungi: Thanks. In other news, you might be interested in https://review.opendev.org/c/openstack/pbr/+/780561 It's trivial but it's been annoying me for a few weeks :)14:51
fungistephenfin: oh, yep! that's a great one15:06
fungii completely agree that's the fix there15:06
stephenfinsweet, thanks :)15:06
*** jtomasek has quit IRC15:07
*** lpetrut has quit IRC15:16
clarkbstephenfin: fungi: note that there are both python_requires and requires_python and they are aliases but one is deprecated iirc15:17
clarkbI can never remember which is the preferred option but https://setuptools.readthedocs.io/en/latest/references/keywords.html seems to imply python_requires should be preferred15:18
stephenfinclarkb: yup, mordred added a fix last year to handle the deprecated one15:18
clarkbmaybe we should keep the warning in place for requires_python to get people to switch?15:18
stephenfind5da2121f875109a94809eeb8aa271fd5a0b3edd15:18
stephenfin(change Id49cf641f2eddac17f1b21b186aa3a150bd863f9)15:18
clarkbya so requires_python is the deprecated value. Why disable the warning in that case?15:19
fungioh, i see what you're saying, 780561 should have python_requires in it (pbr will never put requires_python in the dist anyway)15:20
clarkbright the change is fixing a thing we we don't want to fix. We want people to switch to python_requires15:21
clarkb(taht warning is actually useful if emitted)15:21
stephenfinoh, damn, I completely glazed over that15:23
stephenfinbut nova and python-novaclient (where I see this) both use the newer 'python_requires' in setup.cfg15:24
stephenfin$ tox -e py36 -- -n path/to/test.py15:25
stephenfin /home/stephenfin/openstack/nova/.tox/py36/lib/python3.6/site-packages/pbr/core.py:131: UserWarning: Unknown distribution option: 'requires_python'15:25
stephenfin$ ag requires_python15:25
stephenfinnada15:25
stephenfinso pbr must be doing something funky under the hood15:26
stephenfinor it's one of the projects I see that do seem to have an entry15:27
stephenfin$ ag requires_python .tox/py36/lib/python3.6/site-packages/ -l15:27
stephenfinI see oslo.log, oslo.context, debtcollector, osprofiler, oslotest and pycadf15:27
stephenfinplus jsonpointer but I don't know who owns that15:27
* stephenfin will fix those15:27
*** hashar is now known as hasharAway15:31
stephenfinNope, none of those are specifying that so maybe it's pbr alone messing things up. Sigh15:31
clarkbthat certainly makes this more interesting15:33
clarkbIs it providing both maybe?15:33
clarkbthe only place requires_python shows up in pbr's code base is the line mordred added to map it to python_requiers15:34
stephenfinyeah, I'm not entire sure what generates metadata.json but that's where I see the entries15:35
fungithe entries added to metadata.json are for requires_python though?15:36
funginot python_requires?15:36
fungimetadata.json is not shipped in the dist, fwiw15:38
fungistephenfin: how new is the environment where you're seeing that warning? could it be an interaction with old setuptools writing out the deprecated version of the metadata or something?15:39
stephenfinvery new. Latest version of Fedora (33) with the latest version of python3-virtualenv packaged there15:40
stephenfinwhich I think bundles setuptools and pip15:40
*** hberaud has quit IRC15:42
fungiyeah, in a local environment (debian/sid with basically latest everything) i'm finding requires_python in metadata.json files for lots of things installed by pip/setuptools/wheel15:43
fungipbr's own metadata.json expecially15:44
clarkbmetadata.json is the setuptools produced file (maybe more accurately pkg_resources which is part of setuptools) ?15:45
clarkbperhaps they didn't fully undeprecate it themselves and we're hitting a corner case becase distutils or something along those lines?15:46
fungithe tox venv i see this in includes setuptools 53.0.015:47
*** hberaud has joined #openstack-infra15:50
fungipython_requires is the field setuptools expects, but Requires-Python is the metadata field it writes with that value (as required by pep 345)?15:52
fungiclarkb: stephenfin: https://github.com/pypa/setuptools/blob/main/setuptools/dist.py#L190-L19215:53
clarkbwhat is feeding that back into setuptools wrong then?15:54
clarkbis that pbr assuming all content of metadata.json goes back in 1:1?15:55
fungiwell, according to stephenfin it's pbr complaining about it, not setuptools15:55
clarkbthe use of pbr anyway, but I couldn't find anywhere else in pbr that touched this other than to map it to python_requires for setuptools15:55
*** dklyle has joined #openstack-infra15:56
fungiwell, yeah, the exception stephenfin pasted above suggests it's getting passed into pbr.core.pbr() via the attr parameter, so this is called by setuptools16:01
fungii think setuptools is supplying requires_python there?16:02
*** another_larsks is now known as larsks16:07
stephenfinsorry, had meeting16:08
stephenfin*s16:08
stephenfinlemme grab a stacktrace to see what's triggering this16:08
fungiseems like setuptools maps python_requires to requires_python when writing out the metadata, then later it's feeding that metadata verbatim to pbr, and pbr is complaining it's unrecognized... maybe we just didn't map requires_python to python_requires in enough places in pbr16:09
fungithough i really wish setuptools would maintain bidirectional parity with these fields16:10
stephenfinyup, that's what it seems like16:10
stephenfinhttp://paste.openstack.org/show/803572/16:10
noonedeadpunkfolks, can you kindly point why in the world zuul complains on `Job is not permitted to shadow` https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/78057916:10
funginoonedeadpunk: usually that's because you're trying to parent to or vary a job marked as "final" but i'll take a look16:11
noonedeadpunkso while any branch would have this job set zuul will complain?16:11
clarkbstephenfin: fungi: I think the key here is that pbr is operating on dist.metadata which is the translated form16:11
*** ajitha has joined #openstack-infra16:11
clarkband in that case I think stephenfin's fix is actually correct, though would mask issues with people setting it directly?16:12
clarkbI'm not sure we can get around that with the translation that setuptools is doing16:12
fungishould we not re-map it back to python_requires when interpreting?16:12
noonedeadpunkfungi: I guess I really shadowing in case in takes into considretion all branches, not only master16:12
clarkbfungi: I don't think so because we are operating on dist.metadata and it sounds like the metdata name for the value is requires_python16:12
clarkbit is only the setuptools setup() function argument that is python_requires16:12
clarkbthis is complicated because I think pbr operates in front of and behind that setup() call. The setup.cfg content is in front and gets passed as args to setup()16:13
clarkbthen core.pbr() is called by setup() with the metadata converted values16:13
clarkbfungi: essentially the mapping happens in front of setup() to ensure that setup() converts it back again properly :)16:14
clarkbstephenfin: fungi: I think the best thing here may be to add ac omment explaining why requires_python is expected in that context but not others then merge the fix?16:15
stephenfinthat sounds reasonable16:15
funginoonedeadpunk: oh, yeah i see you're trying to define the job in a different project?16:16
clarkbfungi: noonedeadpunk: yes, I think the issue is the name collision16:16
funginoonedeadpunk: there is a job with the same name already in openstack/ansible16:17
fungier, openstack/openstack-ansible16:17
clarkbyou can make a variant in a pipeline def, but you can't provide a new job declaration16:17
noonedeadpunkSo the thing here is that this job name is dropped from master branch https://codesearch.opendev.org/?q=openstack-ansible-deploy-aio_metal_ovn-ubuntu-focal&i=nope&files=&excludeFiles=&repos=16:17
noonedeadpunkSo I need to rename all underlying branches as well to get job name freed up?16:18
funginoonedeadpunk: got it, yes job names must be unique across projects (but can shadow one another on different branches of the same project)16:18
fungiif that job exists on any branches of openstack/openstack-ansible then a job with the same name can't also be defined in any other repository16:18
noonedeadpunkok, gotcha. I just thought that zuul takes into consideration project branch (dunno why).16:19
fungiat least not within the same tenant16:19
noonedeadpunkthanks fungi16:19
funginot all repos are expected to have more than one branch, or even the same branch names, so zuul can't just assume that a branch name mismatch is a good reason to ignore a duplicate job name16:20
*** hasharAway is now known as hashar16:20
fungiin part because there are a variety of ways you can ask zuul to map differently-named branches between repositories16:20
fungisome explicit, but also implicit fallback16:21
noonedeadpunkyep, that's fair16:23
fungithe trying to parent to or vary a final job situation i was thinking of generates a different error message16:24
*** auristor has quit IRC16:24
*** auristor has joined #openstack-infra16:26
*** auristor has quit IRC16:33
openstackgerritMerged openstack/project-config master: Add Manila-NetApp backend charm to OpenStack charms  https://review.opendev.org/c/openstack/project-config/+/78000516:44
*** jamesdenton has quit IRC16:46
*** jamesdenton has joined #openstack-infra16:47
*** lucasagomes has quit IRC17:02
*** auristor has joined #openstack-infra17:04
*** rpittau is now known as rpittau|afk17:04
*** psachin has quit IRC17:05
*** jonher_ has joined #openstack-infra17:09
*** jonher has quit IRC17:10
*** jonher_ is now known as jonher17:10
*** auristor has quit IRC17:13
*** guillaumec has quit IRC17:14
*** mordred has quit IRC17:15
*** ykarel is now known as ykarel|away17:19
*** auristor has joined #openstack-infra17:20
*** piotrowskim has quit IRC17:24
*** hamalq has joined #openstack-infra17:27
openstackgerritStephen Finucane proposed openstack/pbr master: Revert "Map requires-python to python-requires"  https://review.opendev.org/c/openstack/pbr/+/78063117:30
openstackgerritStephen Finucane proposed openstack/pbr master: util: Convert 'D1_D2_SETUP_ARGS' to a list of tuples  https://review.opendev.org/c/openstack/pbr/+/78063217:30
openstackgerritStephen Finucane proposed openstack/pbr master: Map requires-python to python-requires (attempt 2)  https://review.opendev.org/c/openstack/pbr/+/78063317:30
stephenfinfungi: clarkb: ^17:30
*** guillaumec has joined #openstack-infra17:30
stephenfinI think we did a silly thing17:30
*** dtantsur is now known as dtantsur|afk17:37
clarkbha17:38
clarkbstephenfin: nice catch17:38
clarkbI'll give that a proper review when the thing I'm currently looking at is done17:39
stephenfinack17:40
*** ykarel|away has quit IRC17:41
*** jtomasek has joined #openstack-infra17:46
*** jtomasek has quit IRC17:50
clarkbstephenfin: that stack looks good to me. Do we need to be careful to land the whole stack before doing a pbr release? mostly worried that require_python using projects maybe require it17:53
stephenfinRally seems to be the only project using 'require_python', but it would be better to merge all in one fell swoop, yes17:54
clarkbI'm glad I said something now :)17:56
*** jonher has quit IRC17:58
*** derekh has quit IRC18:03
*** wolsen has quit IRC18:03
*** wolsen has joined #openstack-infra18:05
*** jonher has joined #openstack-infra18:06
fungistephenfin: if that's based on a codesearch query, be aware it only indexes master branches so some projects might still have the old form on older branches (though we only release from a few recent branches these days so it's probably fine)18:08
*** wolsen has quit IRC18:08
stephenfinack18:08
stephenfinshouldn't be an issue unless we only merge part of the stack18:08
openstackgerritMerged openstack/project-config master: Startup nl01.opendev.org idle  https://review.opendev.org/c/openstack/project-config/+/77986318:09
clarkbstephenfin: yup only concerned about partial merge of stack then a release18:09
clarkbif we get all three in then release we should be fine18:09
*** jpena is now known as jpena|off18:10
*** andrewbonney has quit IRC18:13
fungisystemd: inline question on 78063318:16
fungithe first two in the stack look reasonable though18:16
fungisystemd? i meant stephenfin ^18:17
fungisorry systemd!18:17
stephenfingood spot18:19
stephenfinwe overwrite whatever we've parsed18:19
* stephenfin fixes up and sees if we can add tests for this somehow18:19
fungiyeah, i thought that's what would happen, just wasn't 100% sure and figured i'd ask because you've been staring at it longer than i have ;)18:19
*** jonher has quit IRC18:19
*** jonher has joined #openstack-infra18:20
clarkbit will only be a problem if requires_python and python_requires are in the same setup.cfg file18:20
clarkb(I left acomment about it)18:20
clarkbmaking python_requires win seems logical, though not entirely necessary as users should just delete requires_python from setup.cfg if they end up in that situation18:21
stephenfinI don't think so. See https://review.opendev.org/c/openstack/pbr/+/780633/1/pbr/util.py#42218:21
stephenfinWe set kwargs[arg] regardless of the value of in_cfg_value18:21
stephenfinunless there's an early continue I've missed18:21
stephenfinwhich there is :)18:21
clarkbya if it isn't in the file we skip18:23
clarkbif it is in the file then the order matters, in this case we can slightly optimize to prefer the non deprecate value but in all cases of an issue with conflicts like this the setup.cfg should be updated to remove the conflict18:24
stephenfinthat's something we're not doing at the moment18:25
stephenfinchecking for conflicts18:25
stephenfinperhaps we should, as a separate follow-up18:25
fungiright, i was thrown off for the same reasons that the original change was wrong putting the post-transformation value first is confusing and seems backwards18:25
stephenfinfungi: Seeing as we're not using a dict anymore, it'd be easy switch them around18:26
stephenfinI mean, we could have done it before but it's easier now18:27
*** jamesdenton has quit IRC18:31
*** jamesdenton has joined #openstack-infra18:32
*** wolsen has joined #openstack-infra18:37
*** jcapitao has quit IRC18:40
*** hashar has quit IRC18:49
*** ociuhandu has quit IRC18:49
*** ociuhandu has joined #openstack-infra18:50
*** smcginnis has quit IRC18:53
*** ociuhandu has quit IRC18:55
*** janders2 has joined #openstack-infra19:09
openstackgerritStephen Finucane proposed openstack/pbr master: Don't pass empty 'long_description'  https://review.opendev.org/c/openstack/pbr/+/78065719:09
openstackgerritStephen Finucane proposed openstack/pbr master: Add test for cfg -> py transformation  https://review.opendev.org/c/openstack/pbr/+/78065819:09
openstackgerritStephen Finucane proposed openstack/pbr master: Reverse ordering of 'D1_D2_SETUP_ARGS'  https://review.opendev.org/c/openstack/pbr/+/78065919:09
*** owalsh has joined #openstack-infra19:10
*** owalsh_ has quit IRC19:11
*** icey has quit IRC19:11
*** kopecmartin has quit IRC19:11
*** gary_perkins_ has quit IRC19:11
*** Wasaac has quit IRC19:11
*** janders has quit IRC19:11
*** gary_perkins has joined #openstack-infra19:11
*** icey_ has joined #openstack-infra19:11
*** janders2 is now known as janders19:11
*** tbarron has quit IRC19:11
*** Wasaac has joined #openstack-infra19:11
*** ociuhandu has joined #openstack-infra19:22
*** sboyron has quit IRC19:24
*** takamatsu has quit IRC19:32
*** cgoncalves has quit IRC19:32
*** takamatsu has joined #openstack-infra19:33
*** eolivare has quit IRC19:34
*** hamalq has quit IRC19:37
*** hamalq has joined #openstack-infra19:38
*** smcginnis has joined #openstack-infra19:39
*** ociuhandu has quit IRC19:40
*** cgoncalves has joined #openstack-infra19:56
*** nweinber has quit IRC19:59
*** ociuhandu has joined #openstack-infra20:07
*** dciabrin has quit IRC20:17
*** dciabrin has joined #openstack-infra20:17
*** ociuhandu has quit IRC20:27
*** ociuhandu has joined #openstack-infra20:34
openstackgerritKendall Nelson proposed openstack/ptgbot master: Use Opendev Etherpad site  https://review.opendev.org/c/openstack/ptgbot/+/78066820:46
*** ociuhandu has quit IRC20:47
*** mordred has joined #openstack-infra20:52
*** kmarc has quit IRC20:56
*** mtreinish has quit IRC20:56
*** lbragstad has quit IRC20:56
*** bauzas has quit IRC20:56
*** amoralej|off has quit IRC20:56
*** TerianceVi has quit IRC20:56
*** tristanC has quit IRC20:58
*** DinaBelova has quit IRC20:58
*** freerunner has quit IRC20:58
*** amotoki has quit IRC20:58
*** zer0c00l has quit IRC20:58
*** odyssey4me has quit IRC20:58
*** lastmikoi has quit IRC20:58
*** kmarc has joined #openstack-infra20:58
*** mtreinish has joined #openstack-infra20:58
*** lbragstad has joined #openstack-infra20:58
*** bauzas has joined #openstack-infra20:58
*** amoralej|off has joined #openstack-infra20:58
*** TerianceVi has joined #openstack-infra20:58
*** tristanC has joined #openstack-infra20:58
*** DinaBelova has joined #openstack-infra20:58
*** freerunner has joined #openstack-infra20:58
*** amotoki has joined #openstack-infra20:58
*** zer0c00l has joined #openstack-infra20:58
*** odyssey4me has joined #openstack-infra20:58
*** freerunner has quit IRC21:00
*** lastmikoi has joined #openstack-infra21:02
*** freerunner has joined #openstack-infra21:04
*** ociuhandu has joined #openstack-infra21:19
*** lbragstad has quit IRC21:24
*** slaweq has quit IRC21:24
*** lbragstad has joined #openstack-infra21:28
*** vishalmanchanda has quit IRC21:32
*** ociuhandu has quit IRC21:39
*** rlandy has quit IRC21:47
*** rlandy has joined #openstack-infra21:49
*** rcernin has joined #openstack-infra21:57
*** hashar has joined #openstack-infra22:02
*** hashar has quit IRC22:02
*** ociuhandu has joined #openstack-infra22:06
*** ociuhandu has quit IRC22:14
*** rlandy is now known as rlandy|bbl22:17
*** smcginnis has quit IRC22:28
*** ociuhandu has joined #openstack-infra23:00
*** ociuhandu has quit IRC23:05
*** dchen has joined #openstack-infra23:36

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!