Tuesday, 2020-11-03

*** tosky has quit IRC00:26
*** LinPeiWen has joined #openstack-sdks00:30
*** lbragstad has joined #openstack-sdks01:37
*** lbragstad has quit IRC02:13
*** lbragstad has joined #openstack-sdks02:40
*** brinzhang has joined #openstack-sdks03:23
*** lbragstad has quit IRC03:27
*** lbragstad has joined #openstack-sdks03:33
*** brinzhang has quit IRC03:44
*** lbragstad has quit IRC04:44
*** evrardjp has quit IRC05:33
*** evrardjp has joined #openstack-sdks05:33
openstackgerritzhufl proposed openstack/openstacksdk master: Fix invalid assertIsNotNone statement  https://review.opendev.org/76105305:47
*** brinzhang has joined #openstack-sdks05:58
openstackgerritwu.shiming proposed openstack/openstacksdk master: Update TOX_CONSTRAINTS_FILE  https://review.opendev.org/76105505:59
*** nikparasyr has joined #openstack-sdks06:31
*** gtema has joined #openstack-sdks07:35
*** brinzhang_ has joined #openstack-sdks07:37
*** brinzhang has quit IRC07:40
*** ralonsoh has joined #openstack-sdks07:41
*** Luzi has joined #openstack-sdks07:45
*** slaweq has joined #openstack-sdks08:01
*** gtema has quit IRC08:08
*** gtema has joined #openstack-sdks08:10
*** rpittau|afk is now known as rpittau08:19
*** tosky has joined #openstack-sdks08:40
*** jpich has joined #openstack-sdks09:01
*** dtantsur|afk is now known as dtantsur09:35
*** nightmare_unreal has joined #openstack-sdks09:41
nightmare_unrealbrinzhang_:  stephenfin  can you review the code : https://review.opendev.org/#/c/741181/09:41
nightmare_unrealThanks . Hopefully it's correct this time09:41
*** brinzhang_ has quit IRC09:49
*** brinzhang_ has joined #openstack-sdks09:49
brinzhang_nightmare_unreal: ack, added in my list09:50
*** Hybrid512 has joined #openstack-sdks09:52
*** sshnaidm|afk is now known as sshnaidm|rover09:56
openstackgerritMerged openstack/python-openstackclient master: Fix formatting of the flavor properties  https://review.opendev.org/76090610:07
dtantsurhey folks! we're working on ansible 2.10 support in bifrost, does this ring any bells:10:11
dtantsursshnaidm|rover: ^^?10:11
dtantsurthe same code used to work a few days ago10:11
gtemamaybe https://review.opendev.org/#/c/746883/?10:12
dtantsurhave we released that&10:13
gtemaI guess not10:13
gtemait's merged, but not released in galaxy10:13
sshnaidm|roverright, not released10:14
dtantsurwe're using galaxy, so it's not it10:14
dtantsurhttps://github.com/ansible/ansible/blob/e05c62547b3d547ffb433d1cf66e82b353f24262/lib/ansible/utils/collection_loader/_collection_finder.py#L398 is not new at all, I wonder why we only hit it now10:14
sshnaidm|roverdtantsur, I think I got these errors while using old ansible-lint10:15
sshnaidm|roverlooks very familair10:15
dtantsurmmm, that's a hint! it's the linter job that fails10:15
dtantsurhow old was that, do you recall?10:15
dtantsurI think we're using the latest one though..10:16
sshnaidm|roverdtantsur, https://review.opendev.org/#/c/753760/10:16
sshnaidm|roverseems like related to py3.810:16
sshnaidm|roverand this https://review.opendev.org/#/c/754074/10:19
sshnaidm|rovermoved to bionic since it doesn't have 3.8 as default10:19
dtantsuravoiding 2.10 is an option, but we cannot do it forever10:19
dtantsur(especially since the actual code works)10:19
sshnaidm|roverdtantsur, where is your ansible-lint config?10:20
dtantsursshnaidm|rover: not much interesting there https://opendev.org/openstack/bifrost/src/branch/master/.ansible-lint10:21
sshnaidm|roverstill 1.0.0? https://opendev.org/openstack/bifrost/src/branch/master/ansible-collections-requirements.yml10:24
dtantsursshnaidm|rover: is it a hard cap or a lower cap?10:25
sshnaidm|roverI think hard10:25
sshnaidm|roverdtantsur, just for an experiment, I'd try to run it on ubuntu-bionic10:25
sshnaidm|roverto know if it's py3.8 issue10:25
sshnaidm|roverbut most probably not..10:26
dtantsurI can change basepython locally10:26
dtantsur(I see it on my fedora just as well, but it also uses 3.810:26
dtantsurfails with 3.6 just as well10:28
sshnaidm|roverdtantsur, where do you install collection in the job?10:30
dtantsurokay, I think I got it. lemme show you in a minute10:30
dtantsursshnaidm|rover: https://review.opendev.org/#/c/760127/5/tools/ansible-lint.sh10:34
sshnaidm|roverdtantsur, oh, so it didn't find collection10:35
dtantsursshnaidm|rover: it DID find the collection, that's the problem10:36
dtantsurit found it via some implicit logic and choked on it10:36
openstackgerritCédric Jeanneret (Tengu) proposed openstack/keystoneauth stable/train: Cleanup session on delete  https://review.opendev.org/76091110:37
sshnaidm|rovertbh hard to understand where it's installed.. why not to build from ~/src/openstack/openstack-ansible-collections or whatever it is10:38
sshnaidm|roverthat way you can use the newest from repo, not 1.0.0 release..10:39
dtantsursshnaidm|rover: you mean, install it from source? maybe that's what we should do, dunno10:40
sshnaidm|roverdtantsur, yep10:40
dtantsur(the linters job always uses a pre-created source installation)10:40
*** Luzi has quit IRC10:55
stephenfingtema: Any chance you could stick https://review.opendev.org/#/c/757874/ on your review queue, please? :)11:03
gtemayes. BTW you make me disappointed in my changes ;-)11:04
stephenfingtema: Ha, sorry. I hope I wasn't too harsh /o\ Let me know if you disagree with anything, of course11:05
gtemano problem11:05
stephenfin(The main problem was the performance issue of the fetch-update proxy calls, btw. The rest were nits and could be ignored)11:05
gtemabtw, regarding your comment on delete_flavor_extra_specs_property calling get_Flavor 2 times11:05
gtemaare you sure? Cause it actually should not be the case11:06
stephenfinI'm not sure, no. I was just reading the the openstacksdk code and it seemed like it would do a fetch each time11:06
stephenfinPerhaps I'm misreading it?11:06
gtemanope, it should not11:06
gtemathis is the get_resource, which is in the end being invoked to decide what we pass to it: id or instance of the object.11:07
gtemain reality it doesn't invoke the fetch again11:07
stephenfinYou're calling it with flavor.id though11:08
gtemayes, and it is enough for it11:09
stephenfinwhich means you trigger this path, I think? https://opendev.org/openstack/openstacksdk/src/branch/master/openstack/proxy.py#L327-L33111:09
stephenfinthough i didn't check if that actually fetches things11:09
gtemaif I pass complete object it has no advantage, but disadvantage of passing more data to other function (especially ugly in tests)11:09
* stephenfin checks11:09
gtemaeven here no request is being send11:10
stephenfinoh, okay, I wasn't expecting that11:10
* stephenfin is relatively new to openstacksdk11:10
gtemawithout commit this is not doing anything11:11
gtemano problem with that11:11
stephenfinin that case, ignore that comment :) sorry for the confusion11:11
gtemano problem11:11
gtemaI am actually since perhaps a year confused on pretty much that issue - passing more than require is not good (especially for huge objects), but passing less cause some extra functions11:12
stephenfinyeah, the Resource model and general lazy loading concept it uses does take a bit of time to wrap your head around11:13
gtemayes, it's woodoo11:13
*** dtantsur is now known as dtantsur|bbl11:16
*** dulek has joined #openstack-sdks11:30
dulekHi! We're using openstacksdk as our main tool to connect to OpenStack APIs. When there's high load we start too see "urllib3.connectionpool [-] Connection pool is full, discarding connection" warnings.11:31
dulekWhile it seems those are harmless, is there a way to increase pool size? How deep should I look?11:31
gtemadulek, what exactly are you doing?11:32
dulekgtema: A ton of calls to Neutron and Octavia. Kuryr is reacting to events in K8s, so the number of events being handled might be massive.11:32
gtemahmm, not that I am aware of any explicit connection pools we have11:33
gtemacould be it is coming from keystonauth11:33
dulekgtema: Yeah, or even from requests. Okay, I'll try to dig deeper.11:34
gtemaI hope there might be some logs pointing where it is really coming from11:34
gtemadulek - I haven't found any reference in SDK/Keystone for using connectionpools at all11:39
dulekgtema: There's this: https://github.com/openstack/keystoneauth/blob/1f0412a042103009c5a95230e058d08e5bd91fa1/keystoneauth1/session.py#L142311:40
gtemaseems you might then face the defaults11:40
gtemaaah, from adapters11:40
openstackgerritRodolfo Alonso Hernandez proposed openstack/python-openstackclient master: Add "fields" parameter to ListPort query  https://review.opendev.org/75411711:45
dulekIt doesn't seem trivial to pass bigger pool there.11:46
openstackgerritCédric Jeanneret (Tengu) proposed openstack/keystoneauth stable/train: Cleanup session on delete  https://review.opendev.org/76091112:10
openstackgerritArtem Goncharov proposed openstack/openstacksdk master: Switch flavor ops in the cloud layer to proxy  https://review.opendev.org/76112412:21
*** Hybrid512 has quit IRC12:45
*** lbragstad has joined #openstack-sdks12:53
*** dtantsur|bbl is now known as dtantsur12:54
*** Luzi has joined #openstack-sdks12:56
openstackgerritMerged openstack/python-openstackclient master: Add 'openstack server migration list --type' option  https://review.opendev.org/75787412:58
*** lbragstad has quit IRC13:06
*** lbragstad has joined #openstack-sdks13:09
openstackgerritHervé Beraud proposed openstack/microversion-parse master: Use Python 3.8 as the default runtime for tox  https://review.opendev.org/76114513:32
*** jpich has quit IRC13:36
*** jpich has joined #openstack-sdks13:37
*** slaweq has quit IRC13:42
*** slaweq has joined #openstack-sdks13:47
*** jpich has quit IRC13:50
*** jpich has joined #openstack-sdks13:51
*** Hybrid512 has joined #openstack-sdks13:58
openstackgerritMerged openstack/microversion-parse master: Use Python 3.8 as the default runtime for tox  https://review.opendev.org/76114514:08
*** purplerbot has quit IRC14:52
*** mordred has quit IRC15:00
*** Luzi has quit IRC15:10
lbragstadfollow up stable review if folks are interested in taking a look https://review.opendev.org/#/c/746968/15:37
*** mordred has joined #openstack-sdks15:41
*** mordred has quit IRC15:47
*** mordred has joined #openstack-sdks15:57
openstackgerritCédric Jeanneret (Tengu) proposed openstack/keystoneauth stable/train: Cleanup session on delete  https://review.opendev.org/76091115:59
*** LinPeiWen has quit IRC16:00
*** dave-mccowan has joined #openstack-sdks16:35
*** rpittau is now known as rpittau|afk16:46
*** dtantsur is now known as dtantsur|afk16:47
*** purplerbot has joined #openstack-sdks17:03
*** jpich has quit IRC17:04
*** dave-mccowan has quit IRC17:12
*** dave-mccowan has joined #openstack-sdks17:17
*** sshnaidm|rover is now known as sshnaidm|afk17:22
openstackgerritStephen Finucane proposed openstack/python-openstackclient master: compute: Format list of tags for 'server list'  https://review.opendev.org/76120517:22
*** tosky has quit IRC17:57
*** gtema has quit IRC18:07
*** gtema has joined #openstack-sdks18:29
*** ralonsoh has quit IRC18:49
*** gtema has quit IRC18:55
*** ianw_pto is now known as ianw18:59
*** Hybrid512 has quit IRC19:06
*** camelCaser has quit IRC19:17
*** nikparasyr has left #openstack-sdks19:21
*** camelCaser has joined #openstack-sdks19:23
*** tosky has joined #openstack-sdks20:27
*** slaweq has quit IRC20:33
*** slaweq has joined #openstack-sdks20:44
*** slaweq has quit IRC22:02
*** holser has quit IRC22:10
*** holser has joined #openstack-sdks22:11
*** tkajinam has joined #openstack-sdks22:59
*** tosky has quit IRC23:01

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