| *** rfolco|rover has joined #openstack-qa | 00:05 | |
| *** rfolco|rover has quit IRC | 00:47 | |
| *** brinzhang has joined #openstack-qa | 01:02 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Work around to make multinode job as n-v https://review.opendev.org/736411 | 01:08 |
|---|---|---|
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Always source lib/apache when lib/tls is used https://review.opendev.org/699515 | 01:10 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Add GRENADE_USE_EXTERNAL_DEVSTACK to skip part of the setup https://review.opendev.org/736283 | 01:11 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 01:17 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 01:20 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Update grenade setting for stable/train https://review.opendev.org/736285 | 01:21 |
| *** Liang__ has joined #openstack-qa | 01:21 | |
| *** Liang__ is now known as LiangFang | 01:21 | |
| *** rfolco|rover has joined #openstack-qa | 01:23 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Always source lib/apache when lib/tls is used https://review.opendev.org/699515 | 01:27 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 01:27 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Add GRENADE_USE_EXTERNAL_DEVSTACK to skip part of the setup https://review.opendev.org/736283 | 01:27 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 01:27 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Update grenade setting for stable/train https://review.opendev.org/736285 | 01:28 |
| *** paras333 has joined #openstack-qa | 01:40 | |
| *** brinzhang_ has joined #openstack-qa | 01:59 | |
| *** yamamoto has joined #openstack-qa | 02:01 | |
| *** brinzhang has quit IRC | 02:02 | |
| *** paras333 has joined #openstack-qa | 02:11 | |
| *** brinzhang0 has joined #openstack-qa | 02:26 | |
| *** yamamoto has quit IRC | 02:26 | |
| *** yamamoto has joined #openstack-qa | 02:27 | |
| *** yamamoto has quit IRC | 02:27 | |
| *** yamamoto has joined #openstack-qa | 02:27 | |
| *** brinzhang_ has quit IRC | 02:29 | |
| *** paras333 has quit IRC | 02:42 | |
| *** xinranwang_ has joined #openstack-qa | 02:43 | |
| *** brinzhang_ has joined #openstack-qa | 02:47 | |
| *** rcernin has quit IRC | 02:48 | |
| *** brinzhang0 has quit IRC | 02:50 | |
| *** paras333 has joined #openstack-qa | 02:52 | |
| *** paras333 has quit IRC | 02:56 | |
| *** ysirndjuro has quit IRC | 02:58 | |
| *** rcernin has joined #openstack-qa | 02:59 | |
| *** brinzhang_ is now known as brinzhang | 03:01 | |
| *** paras333 has joined #openstack-qa | 03:03 | |
| *** rcernin has quit IRC | 03:05 | |
| *** rcernin has joined #openstack-qa | 03:20 | |
| *** rcernin has quit IRC | 03:22 | |
| *** rcernin has joined #openstack-qa | 03:22 | |
| *** yamamoto has quit IRC | 03:41 | |
| *** yamamoto has joined #openstack-qa | 03:49 | |
| *** markvoelker has quit IRC | 03:49 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-qa | 04:33 | |
| *** gcheresh has joined #openstack-qa | 04:49 | |
| *** abdysn has joined #openstack-qa | 05:11 | |
| *** udesale has joined #openstack-qa | 05:42 | |
| *** markvoelker has joined #openstack-qa | 05:51 | |
| *** markvoelker has quit IRC | 05:55 | |
| openstackgerrit | Pratik proposed openstack/bashate master: optimize size and time using "--no-cache-dir" https://review.opendev.org/736442 | 06:10 |
| *** markvoelker has joined #openstack-qa | 06:19 | |
| *** markvoelker has quit IRC | 06:24 | |
| *** rpittau|afk is now known as rpittau | 06:43 | |
| *** gfidente|afk is now known as gfidente | 06:44 | |
| *** slaweq has joined #openstack-qa | 07:07 | |
| openstackgerrit | Jens Harbott (frickler) proposed openstack/devstack master: Prepare for dropping keystone admin endpoint https://review.opendev.org/651492 | 07:11 |
| openstackgerrit | Jens Harbott (frickler) proposed openstack/devstack master: Stop creating a keystone admin endpoint https://review.opendev.org/735472 | 07:11 |
| *** ralonsoh has joined #openstack-qa | 07:21 | |
| *** tosky has joined #openstack-qa | 07:25 | |
| *** elod has quit IRC | 07:26 | |
| *** LiangFang has quit IRC | 07:31 | |
| *** elod has joined #openstack-qa | 07:33 | |
| *** Liang__ has joined #openstack-qa | 07:34 | |
| *** ccamposr__ has joined #openstack-qa | 07:42 | |
| *** Liang__ has quit IRC | 07:44 | |
| openstackgerrit | Luigi Toscano proposed openstack/devstack stable/stein: sync-devstack-data: new argument devstack_data_base_dir https://review.opendev.org/736475 | 07:45 |
| *** ccamposr has quit IRC | 07:45 | |
| *** Liang__ has joined #openstack-qa | 07:46 | |
| *** rcernin has quit IRC | 07:54 | |
| *** rcernin_ has joined #openstack-qa | 07:54 | |
| AJaeger | frickler, ianw, openSUSE is failing on older releases with py27 not found - why is this working for other distros and not openSUSE? See https://review.opendev.org/#/c/735641/ | 07:55 |
| *** jpena|off is now known as jpena | 07:57 | |
| AJaeger | (this is for devstack) | 07:58 |
| openstackgerrit | Merged openstack/grenade stable/train: Work around to make multinode job as n-v https://review.opendev.org/736411 | 08:04 |
| AJaeger | frickler, ianw, seems Xenial installs pip via tools/install_pip, see https://zuul.opendev.org/t/openstack/build/32bc241cafb24150aefb122d36666177/log/job-output.txt#3807 | 08:05 |
| AJaeger | mmh, pip is installed for openSUSE as well - and then install_pip uninstalls and fails ;( | 08:07 |
| *** lucasagomes has joined #openstack-qa | 08:08 | |
| openstackgerrit | Ignacy Osetek proposed openstack/devstack stable/queens: Fix for downloading uwsgi on centos in lib/apache https://review.opendev.org/736189 | 08:18 |
| *** udesale_ has joined #openstack-qa | 08:19 | |
| *** elod has quit IRC | 08:19 | |
| *** rcernin_ has quit IRC | 08:20 | |
| frickler | AJaeger: seems this has USE_PYTHON3=False, maybe you want to change that to true? otherwise I think we'd need to hold a node to debug further | 08:20 |
| * frickler sets up a hold anyway, you can choose between recheck and changing to py3 | 08:21 | |
| *** udesale has quit IRC | 08:21 | |
| openstackgerrit | Luigi Toscano proposed openstack/grenade stable/train: Always source lib/apache when lib/tls is used https://review.opendev.org/699515 | 08:22 |
| openstackgerrit | Luigi Toscano proposed openstack/grenade stable/train: Add GRENADE_USE_EXTERNAL_DEVSTACK to skip part of the setup https://review.opendev.org/736283 | 08:22 |
| openstackgerrit | Luigi Toscano proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 08:23 |
| openstackgerrit | Luigi Toscano proposed openstack/grenade stable/train: Update grenade setting for stable/train https://review.opendev.org/736285 | 08:23 |
| *** elod has joined #openstack-qa | 08:26 | |
| openstackgerrit | Ignacy Osetek proposed openstack/devstack master: Fix for downloading uwsgi on centos in lib/apache https://review.opendev.org/736194 | 08:27 |
| *** dtantsur|afk is now known as dtantsur | 08:29 | |
| *** yamamoto has quit IRC | 08:42 | |
| *** yamamoto has joined #openstack-qa | 08:42 | |
| ianw | AJaeger: probably just an ensure-pip call in devstack/playbooks/pre.yaml would do it, if it's not being installed otherwise by scripts? | 08:43 |
| openstackgerrit | zhufl proposed openstack/tempest master: Fix test_get_service_by_invalid_params for microversion 2.75 https://review.opendev.org/736503 | 08:43 |
| frickler | ianw: IMO devstack should work outside the gate, too, so we shouldn't rely on any pre stuff there | 08:47 |
| *** elod has quit IRC | 09:00 | |
| openstackgerrit | Xinliang Liu proposed openstack/devstack-plugin-ceph master: support focal https://review.opendev.org/728362 | 09:05 |
| openstackgerrit | Xinliang Liu proposed openstack/devstack-plugin-ceph master: Ceph-iscsi: Use pip to install targetcli-fb https://review.opendev.org/733751 | 09:05 |
| openstackgerrit | Xinliang Liu proposed openstack/devstack-plugin-ceph master: Make ceph iscsi gateway work again https://review.opendev.org/736508 | 09:05 |
| *** yamamoto has quit IRC | 09:12 | |
| *** yamamoto has joined #openstack-qa | 09:13 | |
| *** yamamoto has quit IRC | 09:13 | |
| *** vishalmanchanda has joined #openstack-qa | 09:17 | |
| *** elod has joined #openstack-qa | 09:17 | |
| openstackgerrit | Xinliang Liu proposed openstack/devstack-plugin-ceph master: Make ceph iscsi gateway work again https://review.opendev.org/736508 | 09:18 |
| *** elod has quit IRC | 09:19 | |
| *** yamamoto has joined #openstack-qa | 09:23 | |
| *** elod has joined #openstack-qa | 09:25 | |
| openstackgerrit | Thomas Morin proposed openstack/tempest master: object storage: fix and cleanup header checks https://review.opendev.org/736292 | 09:35 |
| *** yamamoto has quit IRC | 09:40 | |
| ianw | frickler: fair point, that was kind of the idea that this uncovers things that were hidden in the infra images :) | 09:51 |
| frickler | ianw: yeah, seems to work as intended | 09:55 |
| *** tkajinam has quit IRC | 10:05 | |
| *** rpittau is now known as rpittau|bbl | 10:06 | |
| openstackgerrit | Thomas Morin proposed openstack/tempest master: object storage: fix and cleanup header checks https://review.opendev.org/736292 | 10:07 |
| *** yamamoto has joined #openstack-qa | 10:14 | |
| *** Liang__ has quit IRC | 10:18 | |
| openstackgerrit | Federico Ressi proposed openstack/devstack master: Handle uwsgi install for CentOS/RHEL https://review.opendev.org/735558 | 10:18 |
| *** chaconpiza has joined #openstack-qa | 10:19 | |
| *** yamamoto has quit IRC | 10:20 | |
| *** yamamoto has joined #openstack-qa | 10:20 | |
| *** yamamoto has quit IRC | 10:43 | |
| *** yamamoto has joined #openstack-qa | 10:44 | |
| *** yamamoto has quit IRC | 10:44 | |
| *** rcernin_ has joined #openstack-qa | 10:48 | |
| *** yamamoto has joined #openstack-qa | 10:48 | |
| *** yamamoto has quit IRC | 10:53 | |
| *** yamamoto has joined #openstack-qa | 10:54 | |
| *** yamamoto has quit IRC | 10:59 | |
| AJaeger | frickler: in train, it should be USE_PYTHON3=False, shouldn't it? | 11:02 |
| *** udesale_ has quit IRC | 11:12 | |
| *** yamamoto has joined #openstack-qa | 11:16 | |
| *** yamamoto has quit IRC | 11:20 | |
| *** markvoelker has joined #openstack-qa | 11:24 | |
| *** markvoelker has quit IRC | 11:29 | |
| openstackgerrit | Ignacy Osetek proposed openstack/devstack master: Fix for downloading uwsgi on centos in lib/apache https://review.opendev.org/736194 | 11:31 |
| openstackgerrit | Ignacy Osetek proposed openstack/devstack stable/queens: Fix for downloading uwsgi on centos in lib/apache https://review.opendev.org/736189 | 11:32 |
| *** yamamoto has joined #openstack-qa | 11:34 | |
| *** TomStappaerts has quit IRC | 11:37 | |
| *** TomStappaerts has joined #openstack-qa | 11:38 | |
| *** jpena is now known as jpena|lunch | 11:38 | |
| frickler | AJaeger: that's the default probably, but at this point in time I'm not sure we need to keep it that way, py3 should be fully supported as well | 11:39 |
| frickler | except for curiosities like swift | 11:39 |
| AJaeger | frickler: I'm talking about *Train* here - but let me try... | 11:40 |
| AJaeger | also, on train the xenial and centos devstack-platform jobs are failing ;8 | 11:41 |
| frickler | AJaeger: yes, that seemed to be some tempest/reqs issue, maybe someone will fix that, otherwise we'll just drop them | 11:42 |
| openstackgerrit | Andreas Jaeger proposed openstack/devstack stable/train: Set USE_PYTHON3 for openSUSE https://review.opendev.org/736647 | 11:42 |
| AJaeger | frickler: let's try this ^ | 11:43 |
| openstackgerrit | Pratik Raj proposed openstack/coverage2sql master: optimize size and time using --no-cache-dir https://review.opendev.org/736650 | 11:47 |
| *** nweinber has joined #openstack-qa | 11:48 | |
| frickler | AJaeger: hmm, I'm just looking at the node I held and it does seem to have USE_PYTHON3=True. it also does have pip2.7. looking at the error again, the search path seems wrong, it doesn't include /usr/local/bin: "which: no pip2.7 in (/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/sbin:/usr/sbin:/sbin:/usr/sbin/)" | 11:48 |
| frickler | and that's where pip2.7 resides | 11:48 |
| *** raildo has joined #openstack-qa | 11:50 | |
| openstackgerrit | Pratik Raj proposed openstack/devstack master: optimize size and time using --no-cache-dir https://review.opendev.org/736659 | 11:52 |
| openstackgerrit | Pratik Raj proposed openstack/devstack-plugin-ceph master: optimize size and time using --no-cache-dir https://review.opendev.org/736662 | 11:53 |
| frickler | does anyone know that --no-cache-dir spammer? looks like automatically generated patches editing comments to utter nonsense | 11:56 |
| openstackgerrit | Pratik Raj proposed openstack/devstack-tools master: optimize size and time using --no-cache-dir https://review.opendev.org/736668 | 11:56 |
| AJaeger | no, first time spammer ;( | 11:56 |
| kopecmartin | hi, has anyone heard about some changes regarding tox and virtualenv? I noticed some jobs started failing on 'virtualenv: command not found' or 'tox: No such file or directory' .. so i'm tracking now how these dependencies was installed before .. | 12:04 |
| openstackgerrit | Pratik Raj proposed openstack/grenade master: optimize size and time using --no-cache-dir https://review.opendev.org/736680 | 12:05 |
| AJaeger | kopecmartin: did you read the mailing list? ianw send details there... | 12:05 |
| kopecmartin | oh, i didn't lately .. let me check | 12:06 |
| openstackgerrit | Pratik Raj proposed openstack/hacking master: optimize size and time using --no-cache-dir https://review.opendev.org/736681 | 12:06 |
| AJaeger | frickler: "New contributor" - the adjutant change of the series has "Welcome, New Contributor" ;( | 12:06 |
| kopecmartin | AJaeger: i see now, thanks | 12:06 |
| frickler | AJaeger: working alphabetically, bashate came early, yes | 12:08 |
| *** gcheresh has quit IRC | 12:08 | |
| AJaeger | https://review.opendev.org/#/c/736447/ - read the commetns there ;( | 12:09 |
| *** yamamoto has quit IRC | 12:12 | |
| *** gcheresh has joined #openstack-qa | 12:14 | |
| *** rpittau|bbl is now known as rpittau | 12:20 | |
| *** rh-jelabarre has joined #openstack-qa | 12:28 | |
| openstackgerrit | Martin Kopec proposed openstack/tempest master: Ensure tox is installed in run-tempest role https://review.opendev.org/736701 | 12:30 |
| AJaeger | frickler: the changes are all abandoned, read #opendev for details | 12:33 |
| frickler | AJaeger: yeah I saw that, thanks for the fast action to all involved | 12:34 |
| *** jpena|lunch is now known as jpena | 12:38 | |
| *** paras333 has joined #openstack-qa | 13:05 | |
| openstackgerrit | Thomas Morin proposed openstack/tempest master: object storage: fix and cleanup header checks https://review.opendev.org/736292 | 13:07 |
| openstackgerrit | Thomas Morin proposed openstack/tempest master: object storage: fix and cleanup header checks https://review.opendev.org/736292 | 13:19 |
| *** yamamoto has joined #openstack-qa | 13:28 | |
| *** yamamoto has quit IRC | 13:34 | |
| *** abdysn has quit IRC | 13:49 | |
| gmann | ianw: frickler for 'virtualenv: command not found' happening on devstack based job, do we need to include 'ensure-virtualenv' role on all subnodes? or that is done in base job. or that you made part of 'ensure-pip' but i think that you said for xenial case only? | 13:50 |
| openstackgerrit | Dan Smith proposed openstack/devstack master: WIP: Try a more unified glance upload helper https://review.opendev.org/736231 | 13:53 |
| tosky | gmann: is it about the failure of the grenade/py2 job in stein (which made you disable it)? | 13:55 |
| gmann | tosky: its in multiple places of multinode job. I thought it is for legacy job only which need d-g fix. | 13:56 |
| tosky | at least the candidate grenade job on train (so from stein) is failing with that error (missing virtualenv) | 13:59 |
| tosky | so maybe something py2/xenial? | 13:59 |
| gmann | there are other project job also | 14:00 |
| *** ysirndjuro has joined #openstack-qa | 14:00 | |
| *** rfolco|rover is now known as rfolco | 14:01 | |
| *** rcernin_ has quit IRC | 14:14 | |
| clarkb | gmann: I think ensure-virtualenv should happen on all nodes and ensure-pip should be used first if running on xenial | 14:32 |
| gmann | clarkb: yeah, i was searching if zuul job does that or we need to do in devstack base job. | 14:32 |
| clarkb | the idea is jobs will do it becausethe problems arose when we assumed everyone wanted it | 14:33 |
| clarkb | the change we made was forcing an opt in | 14:33 |
| gmann | ok. | 14:37 |
| gmann | clarkb: do you think doing from devstack master can create issue? i am seeing error only on stable branches for now. | 14:43 |
| clarkb | gmann: hrm I wouldve expected it to fail consistently regardless of branch unless newer devstack installs that stuff on itsown | 14:45 |
| clarkb | does latest devstack still use d-g roles? | 14:45 |
| clarkb | maybe we should identify why current devstack works and backport that fix | 14:45 |
| gmann | i think yes, let me confirm | 14:45 |
| clarkb | gmann: do you have links to failed and successful builds? I'll try to help compare over breakfast | 14:49 |
| gmann | clarkb: this backport is failing on neutron-grenade-dvr-multinode - https://review.opendev.org/#/c/735948/2 where master is all good. | 14:50 |
| gmann | clarkb: we just prepare the d-g role but test-matrix is removed so we can remove it from here too- https://opendev.org/openstack/devstack/src/branch/master/.zuul.yaml#L213 | 14:55 |
| clarkb | I think that may be part of ehy | 14:56 |
| clarkb | test-matrix uses a venv right? | 14:56 |
| clarkb | but if we dont use it in master we'd avoid the problem? | 14:56 |
| clarkb | can we avoid the use of test matrix on stable branches or should we make a fix to install virtualenv? (also still need to confirm that is the issue) | 14:57 |
| gmann | clarkb: ohk, i think that was needed till queens only. let me backport that | 14:57 |
| gmann | clarkb: but ussuri also failing where no test-matrix. | 14:59 |
| *** Luzi has joined #openstack-qa | 15:01 | |
| clarkb | gmann: https://zuul.opendev.org/t/openstack/build/9418dbd339724ad98582c3411cbe50fb/log/job-output.txt#13474 its failing in old on ussuri which is actually train | 15:03 |
| clarkb | gmann: that same code seems to exist on master in lib/infra so I'm trying to figure out why master is ok | 15:04 |
| gmann | yeah, but i was confused with why other neutron-grenade-multinode job pass.. | 15:05 |
| clarkb | ya I'm looking at that aspect now | 15:06 |
| clarkb | gmann: I understand it now. If we are running with python2 we set VIRTUALENV_CMD to virtualenv and if we are running with python3 we set it to python3 -m venv | 15:08 |
| clarkb | gmann: it is only failing on python2 because virtualenv is not there | 15:08 |
| clarkb | I'm guessing we don't want to change the python version on those jobs so we should probably add something like if xenial ensure-pip fi ensure-virtualenv | 15:09 |
| clarkb | gmann: do grenade jobs parent to devstack jobs or are they separate? /me is looking now too | 15:12 |
| clarkb | grenade-base is its own job base | 15:12 |
| gmann | yeah from multinode job not devstack | 15:13 |
| gmann | but passing one is also py2 -https://zuul.opendev.org/t/openstack/build/a1dedf64ac304c4ba3e17429f6a3d39c/log/job-output.txt#2205 | 15:13 |
| openstackgerrit | Vasyl Saienko proposed openstack/tempest master: Update skip check for ServersTestBootFromVolume https://review.opendev.org/736742 | 15:14 |
| clarkb | gmann: https://zuul.opendev.org/t/openstack/build/9ae4c5380deb436692cce72ee3292901/log/job-output.txt#2488 that is a passing one I looked at | 15:14 |
| *** ccamposr has joined #openstack-qa | 15:14 | |
| clarkb | weird | 15:15 |
| clarkb | I wonder if we have a stale image | 15:16 |
| *** ccamposr__ has quit IRC | 15:17 | |
| gmann | the one you saw is master which is py3 always as ussuri-?master. but Ussuri job which install train devstack end up with py2 | 15:18 |
| gmann | but another confusion here is even ususri grenade base job use USE_PYTHON3=ture so not sure how they endup on py2 | 15:19 |
| clarkb | gmann: I think that is it | 15:21 |
| clarkb | the python2 passing job ran in rax-iad | 15:21 |
| AJaeger | frickler: setting USE_PYTHON3 did not help the train openSUSE job in https://review.opendev.org/736647 . So, if anybody hasan idea why openSUSE fails on train (see also backscroll for what frickler found), would be great | 15:21 |
| clarkb | the cloud's most recent bionic image is from about June 15 00:00 UTC | 15:21 |
| clarkb | but the removal of pip and virtualenv merged at about June 15 00:37 UTC | 15:22 |
| clarkb | I've not yet confirmed this for sure as the timestamps are all a bit odd but working on it | 15:22 |
| gmann | AJaeger: does it end up running on py2? | 15:22 |
| clarkb | gmann: I think we should assume the python2 failures can be widespread and we need to add ensure-virtualenv and ensure-pip on grenade and devstack | 15:23 |
| gmann | yeah, train gate is all mixed of py2 and py3. | 15:23 |
| gmann | so ensure-pip and then ensure-virtualenv without condition should work fine for py3 also right? | 15:24 |
| clarkb | and I've just foudn our builders aren't building new images because we've run out of disk I think because the old plain images are still around | 15:24 |
| clarkb | gmann: yes that should be safe for both versions | 15:24 |
| gmann | ok. | 15:24 |
| clarkb | I'm going to work on cleaning up our builders this morning if you can work on the job side | 15:24 |
| gmann | sure, thanks | 15:24 |
| gmann | and d-g side, what to do on that ? | 15:25 |
| AJaeger | gmann: https://zuul.opendev.org/t/openstack/build/fe9a3469360d45a39e2c2e20a394940d/log/job-output.txt#3044 - install_pip does some crazy stuff and then pip27 is in /usr/local/bin but devstack does not find it | 15:25 |
| gmann | ask projects to move to zuulv3? | 15:25 |
| gmann | clarkb: found that devstack is all good, ianw did that in train and ussuri onwards is all py3. I will add on greande side - https://review.opendev.org/#/c/727000/ | 15:29 |
| clarkb | for d-g yes I think the goal should continue to be deleting it from as many places as possible | 15:30 |
| clarkb | gmann: cool thanks | 15:30 |
| openstackgerrit | Merged openstack/grenade stable/ussuri: Use ironic-grenade https://review.opendev.org/731842 | 15:32 |
| *** slaweq has quit IRC | 15:33 | |
| *** gcheresh has quit IRC | 15:38 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/ussuri: Add ensure-virtualenv role https://review.opendev.org/736750 | 15:39 |
| gmann | clarkb: ^^ for testing , added depends-on in this https://review.opendev.org/#/c/735948/ | 15:41 |
| gmann | thanks for help. | 15:41 |
| gmann | clarkb: frickler can you review these backports which is needed bymain backport of grenade zuulv3 - https://review.opendev.org/#/c/699515/5 https://review.opendev.org/#/c/736283/4 | 15:43 |
| gmann | merging them first will be easy to backport the venv fix also | 15:44 |
| clarkb | done | 15:46 |
| gmann | thanks | 15:47 |
| clarkb | gmann: for the image cleanup we are landing a change ot upadte nodepool configs which i'll monitor then do any additional cleanup by hand if that becomes necessary | 15:47 |
| clarkb | and that should hopefully get us to a more consistent place which mgiht hurt for a minute but help us find any remaining problems quickly | 15:47 |
| gmann | +1. | 15:48 |
| *** bhagyashris is now known as bhagyashris|away | 15:49 | |
| *** Luzi has quit IRC | 16:01 | |
| *** lucasagomes has quit IRC | 16:03 | |
| tosky | gmann: so if we have this https://review.opendev.org/#/c/727000/ in grenade too, does it mean we can reenable the py2 grenade job in https://review.opendev.org/#/c/736284/ ? | 16:03 |
| gmann | tosky: testing it on ussuri first to see how it behaves | 16:04 |
| tosky | ack | 16:04 |
| *** ccamposr__ has joined #openstack-qa | 16:09 | |
| *** ccamposr has quit IRC | 16:11 | |
| *** gfidente is now known as gfidente|afk | 16:49 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/tempest master: Fix request body for compute unshelve action API https://review.opendev.org/670287 | 16:51 |
| *** rpittau is now known as rpittau|afk | 17:00 | |
| openstackgerrit | Merged openstack/grenade stable/train: Always source lib/apache when lib/tls is used https://review.opendev.org/699515 | 17:06 |
| openstackgerrit | Merged openstack/grenade stable/train: Add GRENADE_USE_EXTERNAL_DEVSTACK to skip part of the setup https://review.opendev.org/736283 | 17:06 |
| *** TomStappaerts has quit IRC | 17:07 | |
| *** TomStappaerts has joined #openstack-qa | 17:07 | |
| *** TomStappaerts has quit IRC | 17:08 | |
| *** dtantsur is now known as dtantsur|afk | 17:10 | |
| *** jpena is now known as jpena|off | 17:21 | |
| *** ccamposr has joined #openstack-qa | 17:29 | |
| *** ccamposr__ has quit IRC | 17:32 | |
| openstackgerrit | Merged openstack/devstack-plugin-ceph master: support focal https://review.opendev.org/728362 | 17:44 |
| *** gcheresh has joined #openstack-qa | 17:55 | |
| *** ralonsoh has quit IRC | 17:57 | |
| openstackgerrit | Victoria Martinez de la Cruz proposed openstack/devstack-plugin-ceph master: Updates script to use latest version of Ceph https://review.opendev.org/676722 | 18:20 |
| *** gcheresh has quit IRC | 18:26 | |
| kopecmartin | gmann: hi, how do we handle installation of tox in tempest? python-tempestconf stein job is failing on not having tox installed, however, train and master jobs are fine | 18:28 |
| clarkb | kopecmartin: I believe that the devstack base job is using the ensure-tox role in devstack/playbooks/pre.yaml | 18:32 |
| clarkb | er thats not the right path | 18:33 |
| kopecmartin | clarkb: ah, i see, playbooks/tox/pre.yaml ? | 18:34 |
| clarkb | kopecmartin: https://opendev.org/openstack/devstack/src/branch/stable/stein/playbooks/tox/pre.yaml#L8 that path now to double check which jobs use that | 18:34 |
| clarkb | yup | 18:34 |
| clarkb | that is used by devstack-tox-functional-consumer | 18:34 |
| clarkb | ah there it is in devstack/lib/tempest there is a tox install in devstack itself | 18:35 |
| gmann | tox playbooks only for functional jobs | 18:35 |
| clarkb | kopecmartin: gmann https://opendev.org/openstack/devstack/src/branch/stable/stein/lib/tempest#L691-L708 that should install it for any devstack job that enables the tempest devstack service | 18:36 |
| gmann | yeah, kopecmartin you have link for stein patch failing for tox not there? | 18:37 |
| kopecmartin | gmann: f.e this patch https://review.opendev.org/#/c/736527/2 | 18:39 |
| kopecmartin | clarkb: gmann: the jobs in python-tempestconf use a devstack base job but i think they don't use devstack to install tempest, which could explain it | 18:40 |
| kopecmartin | however it's weird, that it works that way for train and master | 18:40 |
| gmann | kopecmartin: ohk, how it install then? | 18:41 |
| kopecmartin | gmann: here is the playbook: https://github.com/openstack/python-tempestconf/blob/master/playbooks/python-tempestconf-tempest-devstack.yaml | 18:41 |
| gmann | kopecmartin: so this is first place you are preparing the tempest venv right - https://github.com/openstack/python-tempestconf/blob/505e20a4dd771536257a4c4db645da707f727668/roles/install-plugins/tasks/main.yaml#L3 | 18:51 |
| kopecmartin | gmann: yes, but that role is executed only in python-tempestconf-tempest-devstack-admin-plugins job which works | 18:52 |
| kopecmartin | gmann: hmm, yeah, that means the tempest must be installed by devstack somehow, not specifically set in our jobs, so default behaviour probably | 18:53 |
| kopecmartin | gmann: stein job differs from train/master one only by setting override-checkout and branch, f.e. https://github.com/openstack/python-tempestconf/blob/master/.zuul.yaml#L166-L168 | 18:54 |
| *** vishalmanchanda has quit IRC | 18:55 | |
| gmann | kopecmartin: i am trying to understand the flow where you are running tox from cloned tempest | 18:55 |
| *** ysirndjuro has left #openstack-qa | 18:55 | |
| *** ysirndjuro has quit IRC | 18:55 | |
| kopecmartin | gmann: based on the playbook i sent above, we run devstack (run-devstack) role, then we prepare tempest by running tempest roles such as setup-tempest-run-dir, setup-tempest-data-dir, acl-devstack-files | 18:56 |
| kopecmartin | gmann: then we generate our own tempest.conf which is put to a directory where devstack installed tempest | 18:57 |
| kopecmartin | gmann: and then we run run-tempest role | 18:57 |
| gmann | i missed this in that role- chdir: "{{devstack_base_dir}}/tempest" | 18:57 |
| gmann | got it | 18:58 |
| kopecmartin | gmann: any chance devstack in stein branch doesn't install tox? that's the only difference i can think of | 18:58 |
| gmann | kopecmartin: may be it depends on image but i think you can use ensure-tox role for sure as you are preparing the tempest vnenv or the way devstack does | 18:59 |
| gmann | v | 19:00 |
| gmann | https://opendev.org/openstack/devstack/src/branch/stable/stein/lib/tempest#L691-L708 | 19:00 |
| kopecmartin | gmann: btw, how should it work? should be roles independent in meaning if the role uses tox it should also check if it's installed? otherwise the role depends on parents which may or may not install the dependency | 19:04 |
| kopecmartin | but on the other side it would led to many checks on different levels i guess | 19:04 |
| gmann | kopecmartin: ensure-tox does verify if it is installed | 19:08 |
| kopecmartin | gmann: yes, the question is if want something like this or not: https://review.opendev.org/#/c/736701/ | 19:09 |
| gmann | kopecmartin: by having tox_envlist default as smoke in run-tempest role, i think it is fine to do that and let users use run-tempest independently | 19:20 |
| gmann | though i am not sure why python-tempestconf job run-tempest twice 1. before plugins install and 2. after plugins install | 19:21 |
| gmann | if some day you remove the first run-tempest then i will suggest to add ensure-tox in your install-plugins role too as you are creating tox env there at very first time | 19:21 |
| gmann | kopecmartin: ttyl, need to go for lunch. | 19:22 |
| *** gmann is now known as gmann_afk | 19:22 | |
| openstackgerrit | Jens Harbott (frickler) proposed openstack/devstack stable/train: DNM: Test with additional path https://review.opendev.org/736812 | 19:29 |
| *** yamamoto has joined #openstack-qa | 19:32 | |
| *** xinranwang_ has quit IRC | 19:33 | |
| artom | What's the etiquette around adding a waiter that'll be used only in a plugin? | 19:36 |
| artom | Specifically, a nova service waiter? | 19:36 |
| artom | Adding code to tempest that would be dead in tempest seems weird | 19:37 |
| *** gmann_afk is now known as gmann | 19:37 | |
| artom | But... it's a generic enough thing that I could see others using | 19:37 |
| *** yamamoto has quit IRC | 19:37 | |
| gmann | artom: what waiter? service up or not? or some resource status? | 19:37 |
| artom | gmann, yeah, talking about specifically Nova's os-services API, waiting for a service to be up/down/whatever | 19:38 |
| artom | Oh. | 19:38 |
| artom | *facepalm* | 19:38 |
| artom | Nothing in tempest would be starting/stopping these | 19:38 |
| artom | So nothing would need to wait. | 19:38 |
| gmann | artom: yeah, adding in plugin side is needed not in tempest as no service control in tempest. | 19:38 |
| artom | gmann, yep, that seems obvious in retrospect. Sorry for the noise. | 19:39 |
| gmann | np! | 19:40 |
| gmann | clarkb: frickler ensure-virtualenv fix the grenade issue, can you review -https://review.opendev.org/#/c/736750/1 (passed the earlier failing multinode grenade dvr job in https://review.opendev.org/#/c/735948/3) | 19:42 |
| gmann | kopecmartin: i am +2 on this - https://review.opendev.org/#/c/736701/1 | 19:42 |
| kopecmartin | gmann: thanks | 19:45 |
| openstackgerrit | Dan Smith proposed openstack/devstack master: WIP: Try a more unified glance upload helper https://review.opendev.org/736231 | 19:45 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Add ensure-virtualenv role https://review.opendev.org/736819 | 19:51 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 20:00 |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Update grenade setting for stable/train https://review.opendev.org/736285 | 20:01 |
| gmann | we should have grenade jobs on train working now ^^, waiting for results | 20:02 |
| tosky | yay! | 20:04 |
| tosky | thanks | 20:04 |
| *** gmann is now known as gmann_afk | 20:10 | |
| *** ccamposr has quit IRC | 20:12 | |
| *** nweinber has quit IRC | 20:14 | |
| *** ccamposr has joined #openstack-qa | 20:18 | |
| *** ccamposr has quit IRC | 20:19 | |
| clarkb | gmann_afk: done. Maybe ianw can approve it when the day starts on that side of the planet | 20:21 |
| *** slaweq has joined #openstack-qa | 20:23 | |
| *** rfolco has quit IRC | 20:52 | |
| *** gcheresh has joined #openstack-qa | 20:54 | |
| *** gcheresh has quit IRC | 21:00 | |
| *** raildo has quit IRC | 21:10 | |
| *** rcernin_ has joined #openstack-qa | 21:54 | |
| *** markvoelker has joined #openstack-qa | 21:58 | |
| *** markvoelker has quit IRC | 22:02 | |
| *** gmann_afk is now known as gmann | 22:04 | |
| *** tonyb has joined #openstack-qa | 22:05 | |
| gmann | clarkb: thanks | 22:05 |
| tosky | gmann: interesting; grenade is going to fail on https://review.opendev.org/736284, but it passed on https://review.opendev.org/#/c/736285/ | 22:06 |
| gmann | multinode job? checking | 22:07 |
| tosky | no, the base grenade job | 22:11 |
| tosky | and that doesn't make much sense (unless I misread the logs) | 22:12 |
| *** rcernin_ has quit IRC | 22:13 | |
| *** slaweq has quit IRC | 22:13 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade stable/train: Setup native grenade Zuul v3 jobs https://review.opendev.org/736284 | 22:19 |
| gmann | tosky: ^^, squash the branch setting backport also as it was doing train->master upgrade | 22:19 |
| gmann | let's see if tat error continue on py2 job we will make it voting and move forward and later we can debug the py2 job. | 22:20 |
| tosky | oh, right | 22:23 |
| tosky | new was initially checking out as train, but the logic was against master | 22:23 |
| tosky | please note there is devstack stein backport as dependency, which is needed in order to have proper logs | 22:24 |
| tosky | this one : https://review.opendev.org/#/c/736475/ | 22:24 |
| gmann | yeah, i saw that. once things is green i will review that. | 22:24 |
| gmann | thanks for that | 22:24 |
| tosky | let's see tomorrow o/ | 22:25 |
| *** tosky has quit IRC | 22:25 | |
| gmann | i will check after my bike ride if anything extra needed | 22:26 |
| *** gfidente|afk has quit IRC | 22:29 | |
| *** ccamposr has joined #openstack-qa | 22:33 | |
| *** tkajinam has joined #openstack-qa | 23:00 | |
| *** rcernin_ has joined #openstack-qa | 23:15 | |
| *** rcernin_ has quit IRC | 23:15 | |
| *** rcernin has joined #openstack-qa | 23:16 | |
| *** ccamposr__ has joined #openstack-qa | 23:17 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade master: Adding document on Extended Maintainance testing policy https://review.opendev.org/736866 | 23:18 |
| *** ccamposr has quit IRC | 23:20 | |
| *** yamamoto has joined #openstack-qa | 23:34 | |
| *** yamamoto has quit IRC | 23:39 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/grenade master: Adding document on Extended Maintainance testing policy https://review.opendev.org/736866 | 23:48 |
| *** rfolco has joined #openstack-qa | 23:58 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!