*** diablo_rojo has quit IRC | 00:05 | |
*** diablo_rojo has joined #openstack-tc | 00:06 | |
*** diablo_rojo has quit IRC | 01:13 | |
*** gagehugo has quit IRC | 01:58 | |
*** jamesmcarthur has joined #openstack-tc | 02:44 | |
*** jamesmcarthur has quit IRC | 03:31 | |
*** jamesmcarthur has joined #openstack-tc | 03:32 | |
*** jamesmcarthur has quit IRC | 03:59 | |
*** jamesmcarthur has joined #openstack-tc | 04:00 | |
*** jamesmcarthur has quit IRC | 04:04 | |
*** tetsuro has quit IRC | 04:13 | |
*** tetsuro has joined #openstack-tc | 04:13 | |
*** jamesmcarthur has joined #openstack-tc | 05:01 | |
*** jamesmcarthur has quit IRC | 05:06 | |
*** gmann has quit IRC | 05:27 | |
*** evrardjp has quit IRC | 05:34 | |
*** evrardjp has joined #openstack-tc | 05:34 | |
*** jamesmcarthur has joined #openstack-tc | 06:52 | |
*** jamesmcarthur has quit IRC | 06:54 | |
*** ricolin has quit IRC | 07:11 | |
*** lpetrut has joined #openstack-tc | 07:29 | |
*** lpetrut has quit IRC | 07:30 | |
*** lpetrut has joined #openstack-tc | 07:31 | |
*** witek has joined #openstack-tc | 07:55 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/governance-website master: Introducing ideas on the governance website https://review.opendev.org/706439 | 07:56 |
---|---|---|
openstackgerrit | Jean-Philippe Evrard proposed openstack/governance-website master: Introducing ideas on the governance website https://review.opendev.org/706439 | 07:58 |
*** slaweq__ has joined #openstack-tc | 08:00 | |
*** slaweq__ has quit IRC | 08:05 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/governance master: Add Ideas into the governance sidebar https://review.opendev.org/706440 | 08:08 |
*** slaweq__ has joined #openstack-tc | 08:08 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/governance-sigs master: Add Ideas into the governance sidebar https://review.opendev.org/706441 | 08:10 |
*** rpittau|afk is now known as rpittau | 08:10 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/governance-sigs master: Add Ideas into the governance sidebar https://review.opendev.org/706441 | 08:10 |
*** slaweq has joined #openstack-tc | 08:17 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/election master: Add Ideas into the top sidebar https://review.opendev.org/706444 | 08:18 |
*** iurygregory has joined #openstack-tc | 08:19 | |
*** slaweq__ has quit IRC | 08:19 | |
*** tetsuro has quit IRC | 08:46 | |
*** tosky has joined #openstack-tc | 08:48 | |
*** david-lyle has quit IRC | 09:16 | |
*** david-lyle has joined #openstack-tc | 09:16 | |
frickler | hi, I would like to know how the TC feels about continuing to claim RHEL8 support for Ussuri given that there is still no working devstack support (seems stalled on waiting for RDO to release things https://review.opendev.org/688614) | 10:03 |
*** e0ne has joined #openstack-tc | 10:08 | |
*** slaweq has quit IRC | 10:13 | |
*** slaweq has joined #openstack-tc | 10:15 | |
*** purplerbot has quit IRC | 10:50 | |
*** slaweq_ has joined #openstack-tc | 10:51 | |
*** purplerbot has joined #openstack-tc | 10:51 | |
*** slaweq has quit IRC | 10:53 | |
*** bauzas has quit IRC | 10:57 | |
*** bauzas has joined #openstack-tc | 10:57 | |
*** rpittau is now known as rpittau|bbl | 11:18 | |
evrardjp | frickler: good question. I think the idea behind the support of the distro is two ways: the infra & projects side, and the users side. On the users side, it's good to say that OpenStack has been tested for RHEL8. However, you are right, one can doubt this claim, if it's not even tested in devstack. | 11:56 |
evrardjp | However (double combo!) devstack is not the only way to claim its tested though, therefore it might be not enough to remove it. Also, removing after the fact might be triggering discussions about _which python versions_ needed to be supported at all during the Ussuri cycle. | 11:58 |
evrardjp | On the infra & projects, I suppose that it's up to infra to decide to remove the image, and up to the the projects to not test with rhel/centos (AFAIK not many projects do, maybe swift does, but my memory could be wrong). However, I think we'll cross that bridge when we'll get to it, don't you think? | 11:59 |
evrardjp | So IMHO, we should think about helping them getting this achieved more than dropping them. It's probably better for OpenStack in general | 12:00 |
evrardjp | Ofc I can change my mind with new arguments | 12:01 |
*** gmann has joined #openstack-tc | 12:08 | |
*** slaweq__ has joined #openstack-tc | 12:09 | |
*** slaweq_ has quit IRC | 12:11 | |
gmann | o/ | 12:12 |
fungi | "Each project should run some functional tests on these platforms so we make sure OpenStack works with distros used in production. The scope of these functional tests are discussed for every project, and may adjust their coverage depending of resources and support investment. These tests are run by using existing tooling, which comes with a reasonable expectation that it’s viable on the indicated | 12:17 |
fungi | distributions." | 12:17 |
fungi | so it doesn't specifically say that the functional tests have to be devstack/tempest | 12:17 |
frickler | so you are saying each project should invent their own way to test on centos instead of being able to rely on devstack as we did up to now? | 12:23 |
fungi | nope, just saying that the policy doesn't specifically require devstack (and some projects, for example deployment projects, already don't use it) | 12:28 |
evrardjp | that's exactly what I meant, thanks for clarifying fungi :) | 12:29 |
evrardjp | frickler: do you know what's going on, so that devstack is not progressing? | 12:29 |
frickler | evrardjp: I'm not sure, it's a mix of "lack of developer investment" and "waiting on RDO" I guess. | 12:31 |
frickler | so maybe also just misinterpreting the "should" in the first sentence of fungi's quote. | 12:31 |
frickler | like would it be o.k. for a project (like designate) to say: we don't test anything on centos until devstack supports it, and still be in compliance with the PTI? | 12:32 |
frickler | we also don't test on opensuse afaict, but that's a different issue | 12:32 |
* njohnston sees if he can find someone from RDO to make a comment | 12:40 | |
njohnston | ^^ ianw | 12:48 |
gmann | frickler project should not need devstack for functional testing which is required criteria ti declare distro is tested and supported. | 12:50 |
gmann | evrardjp helping them and support testing as much as possible is generally good and we should do that BUT QA team is shortage of man power. We are somehow managing the load till now but i am afraid we need to start cutting the things-to-support in near future. | 12:54 |
evrardjp | frickler: yeah a project can decide afaik | 12:54 |
evrardjp | gmann: I think it's fair | 12:55 |
evrardjp | I am all creating the idea framework to make sure we can rethink from the ground up everything if necessary. | 12:55 |
evrardjp | gmann: I am looking forward to an idea of yours on how we could refactor this. I might have one on your direction :p | 12:56 |
gmann | other thing to not here - nobody verify the all supported distro testing on project side , i am 100% sure majority of projects do not test centos or opensuse. so claiming their support in openstack need more mandatory functional jobs on all those distro. | 12:57 |
evrardjp | gmann: I thought of that, when I brought the addition of the opensuse support in fact. | 12:59 |
gmann | evrardjp: +1. "refactoring" need someone to distribute load which is difficult now a days. I still feel, this is what BoD should help at least on horizontal team resources. | 12:59 |
evrardjp | and you're right, only a few test centos or opensuse | 12:59 |
fungi | we also went for years without devstack working on centos but still claimed openstack ran on it | 12:59 |
evrardjp | but for me if a deployment project test centos/opensuse, it's good enough for the users, as its what the users are doing | 12:59 |
njohnston | gmann: Perhaps we should add an investment opportunity for QA team members? | 12:59 |
evrardjp | njohnston: I think it already is there | 13:00 |
evrardjp | but sure, I don't mind ensuring that for 2020, this is proposed. | 13:00 |
njohnston | evrardjp: I checked https://governance.openstack.org/tc/reference/upstream-investment-opportunities/ and did not see it | 13:00 |
gmann | njohnston yeah, that is good idea. | 13:00 |
gmann | evrardjp not till now. | 13:00 |
evrardjp | I thought it was, my bad. | 13:00 |
gmann | evrardjp but PTI says every project need fucntional jobs for each distro | 13:00 |
evrardjp | njohnston: do you want to propose it? | 13:01 |
gmann | not only deployment projects | 13:01 |
evrardjp | was it every project? | 13:01 |
njohnston | evrardjp: Definitely. | 13:01 |
evrardjp | damn my memory is bad | 13:01 |
fungi | "Each project should run some functional tests on these platforms..." | 13:01 |
fungi | https://governance.openstack.org/tc/reference/project-testing-interface.html#pti-linux-distros | 13:01 |
gmann | yes | 13:01 |
fungi | (one could argue that "should" is not the same as "must") | 13:02 |
gmann | hum, that is good point. | 13:02 |
evrardjp | "Each project should run some functional tests on these platforms so we make sure OpenStack works with distros used in production. " -> I think it was on purpose made vague | 13:02 |
evrardjp | well thanks fungi again :D | 13:03 |
evrardjp | With the explicit: | 13:03 |
evrardjp | "The scope of these functional tests are discussed for every project, and may adjust their coverage depending of resources and support investment" | 13:03 |
fungi | it was made vague, because we have a lot of different sorts of projects and not all of them can be tested in the same ways | 13:03 |
evrardjp | yeah exactly | 13:03 |
evrardjp | and as long as one could test _multiple_ projects for the distro, we were happy. We didn't want to multiply tests for no reason | 13:03 |
fungi | in the past, when devstack wasn't working on centos yet, we argued that projects like tripleo exercising openstack on that platform was a means of providing coverage for the projects they integrate, even if not every project runs tripleo jobs on every proposed change | 13:05 |
evrardjp | But I think this was contentious in the past, and I would look forward a different wording and simplifications, assuming there are resources provided to qa to make things happen | 13:05 |
evrardjp | fungi: yeah that's it! | 13:05 |
evrardjp | fungi: thanks for being the smart living hard drive that you are :p | 13:06 |
*** slaweq has joined #openstack-tc | 13:13 | |
gmann | coming back to frickler question, do we have functional tox jobs/template for centos/opensuse ? so that we can say project can still test their code on centos7/opensuse even devstack not supporting centos. | 13:14 |
*** rpittau|bbl is now known as rpittau | 13:14 | |
*** slaweq__ has quit IRC | 13:15 | |
*** Luzi has joined #openstack-tc | 13:16 | |
evrardjp | I suppose that OSA has for centos opensuse. | 13:17 |
evrardjp | it's worth checking if it's still the case | 13:17 |
fungi | it's also debatable how much unit tests for a lot of projects cross the boundary into functional testing territory (nova requiring a running mysql datavbase and libvirt, swift requiring a local xfs filesystem, and so on) | 13:20 |
fungi | the "unit tests" in many projects are not very thoroughly mocked out where system resources are concerned | 13:21 |
njohnston | neutron has a neutron-centos-7-tripleo-standalone job that is currently on hold because of drop-py27 issues | 13:21 |
njohnston | that is on centos 7 | 13:21 |
fungi | yeah, i was going to say, it's probably not testing on centos 8 anyway if it's blocked on needing python 2.7 | 13:22 |
fungi | getting devstack into working order on centos 8 is obviously preferable, but if we can't by ussuri release day for some reason or another then we probably have the ability to say we expect openstack to work on centos 8 based on other indicators | 13:26 |
fungi | and i'm supposing the situation is similar for opensuse | 13:26 |
tosky | afaik the RDO plan (and not only RDO) is to have everything in shape before the release | 13:29 |
*** jamesmcarthur has joined #openstack-tc | 13:31 | |
tosky | so I understand that this is a painful situation, but it should be solved before the release | 13:33 |
gmann | fungi "say we expect openstack to work on centos 8 based on other indicators" - 'other indicators' you mean with current testing on deployment projects or if each project start functional tests ? | 13:33 |
*** gmann_ has joined #openstack-tc | 13:40 | |
*** gmann has quit IRC | 13:40 | |
*** gmann_ is now known as gmann | 13:40 | |
*** gmann has quit IRC | 13:40 | |
*** gmann has joined #openstack-tc | 13:41 | |
*** ijolliffe has joined #openstack-tc | 13:42 | |
*** jamesmcarthur has quit IRC | 13:44 | |
*** jamesmcarthur has joined #openstack-tc | 13:45 | |
fungi | gmann: or by observing tripleo's jobs running successfully on centos-8 nodes (once they do) | 13:47 |
fungi | or if openstack-ansible is testing on centos-8 with a majority of popular services | 13:47 |
slaweq | njohnston: gmann I think that everything in TripleO is run on centos7 now | 13:47 |
fungi | right, i believe they're held up by the same rdo delays as devstack is | 13:48 |
slaweq | also e.g networking-midonet was testing with tempest on centos7 but now those jobs are probably removed as they were broken for different reasons | 13:50 |
gmann | ok, and we need to change that 'should' to 'must in PTI' sometime. may be after ussuri as py2 drop is very noisy to do on testing side in this cycle. | 13:50 |
*** jamesmcarthur has quit IRC | 13:50 | |
fungi | gmann: doing that might also depend on being more specific on what we mean by "project" in that case | 13:51 |
gmann | slaweq: ok. i think midonet is something did not support bionic also ? that is what i remember when migratin to bionic in stein | 13:51 |
gmann | fungi: every code repo ? | 13:51 |
*** jamesmcarthur has joined #openstack-tc | 13:52 | |
fungi | gmann: tripleo must run functional tests on platforms they lack support for? and how about the repos for the i18n team? infrastructure team's repos? ... | 13:52 |
fungi | i have a feeling if we said tripleo must run functional tests on ubuntu and opensuse, they'd have to just cease being part of openstack instead | 13:53 |
fungi | i wonder similarly if openstack-charms would ever work on centos or opensuse | 13:54 |
fungi | requiring functional tests of the packaging-rpm teams deliverables on ubuntu would be similarly weird | 13:55 |
gmann | yeah | 13:55 |
fungi | for that matter teams like qa and release management may have deliverables which make no sense to require testing on multiple platforms | 13:56 |
gmann | some qa tooling need like Tempest, patrole but not all | 13:57 |
fungi | right, requiring cross-platform testing for openstack-health and coverage2sql, maybe eslint-config-openstack would seem like overkill | 13:58 |
slaweq | gmann: correct | 13:59 |
fungi | so yeah, i think "should" is better than "must" there unless we more tightly scope the term "projects" | 14:00 |
*** lbragstad has joined #openstack-tc | 14:19 | |
*** Luzi has quit IRC | 14:45 | |
cmurphy | ftr keystone tests on opensuse | 14:49 |
*** witek has quit IRC | 15:03 | |
*** rpittau is now known as rpittau|brb | 15:04 | |
*** lpetrut has quit IRC | 15:08 | |
*** rpittau|brb is now known as rpittau | 15:20 | |
fungi | via devstack, or other means? | 15:32 |
cmurphy | fungi: via devstack | 15:35 |
cmurphy | https://opendev.org/openstack/keystone/src/branch/master/.zuul.yaml#L35-L65 | 15:36 |
fungi | cool, so it's at least working there for some projects | 15:37 |
jrosser | openstack-ansible is reasonably close to having something on centos8 - getting some help to iron out non openstack related changes centos7<>8 would be super helpful there | 15:37 |
*** jamesmcarthur has quit IRC | 15:38 | |
jrosser | it's not really clear who to reach out to for that | 15:38 |
*** jamesmcarthur has joined #openstack-tc | 15:38 | |
*** witek has joined #openstack-tc | 15:39 | |
njohnston | jrosser: What kind of changes? | 15:41 |
njohnston | jrosser: You could talk with the tripleo guys, see how they're handling those issues | 15:42 |
jrosser | one trivial example is i can't make repo priorities work | 15:42 |
*** david-lyle is now known as dklyle | 15:53 | |
*** bnemec has quit IRC | 15:55 | |
*** bnemec has joined #openstack-tc | 15:56 | |
*** jamesmcarthur has quit IRC | 16:10 | |
*** jamesmcarthur has joined #openstack-tc | 16:12 | |
*** jamesmcarthur has quit IRC | 16:17 | |
*** jamesmcarthur has joined #openstack-tc | 16:22 | |
*** jamesmcarthur has quit IRC | 16:27 | |
*** slaweq has quit IRC | 16:37 | |
*** rpittau is now known as rpittau|afk | 17:02 | |
*** jamesmcarthur has joined #openstack-tc | 17:07 | |
*** e0ne has quit IRC | 17:08 | |
*** jamesmcarthur has quit IRC | 17:20 | |
*** jamesmcarthur has joined #openstack-tc | 17:21 | |
*** jamesmcarthur has quit IRC | 17:27 | |
*** jamesmcarthur has joined #openstack-tc | 17:29 | |
*** jamesmcarthur has quit IRC | 17:33 | |
*** jamesmcarthur has joined #openstack-tc | 17:33 | |
*** evrardjp has quit IRC | 17:34 | |
*** evrardjp has joined #openstack-tc | 17:34 | |
*** witek has quit IRC | 17:38 | |
*** jamesmcarthur has quit IRC | 17:38 | |
*** tosky has quit IRC | 17:40 | |
*** jamesmcarthur has joined #openstack-tc | 17:41 | |
*** jamesmcarthur has quit IRC | 17:43 | |
*** jamesmcarthur has joined #openstack-tc | 17:44 | |
*** jamesmcarthur has quit IRC | 17:48 | |
*** jamesmcarthur has joined #openstack-tc | 17:53 | |
*** diablo_rojo has joined #openstack-tc | 17:55 | |
*** jamesmcarthur has quit IRC | 18:02 | |
*** jamesmcarthur has joined #openstack-tc | 18:04 | |
*** e0ne has joined #openstack-tc | 18:10 | |
*** e0ne_ has joined #openstack-tc | 18:13 | |
*** e0ne has quit IRC | 18:13 | |
*** jamesmcarthur has quit IRC | 18:27 | |
*** jamesmcarthur has joined #openstack-tc | 18:33 | |
*** jamesmcarthur has quit IRC | 18:44 | |
*** e0ne_ has quit IRC | 19:12 | |
*** lbragstad has quit IRC | 20:48 | |
*** ijolliffe has quit IRC | 20:51 | |
*** iurygregory has quit IRC | 22:29 | |
*** tosky has joined #openstack-tc | 22:31 | |
*** camelCaser has quit IRC | 22:57 | |
*** camelCaser has joined #openstack-tc | 22:58 | |
openstackgerrit | Nate Johnston proposed openstack/governance master: Add QA upstream contribution opportunity https://review.opendev.org/706637 | 23:34 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!