*** jmasud has quit IRC | 00:04 | |
*** jmasud has joined #oooq | 00:06 | |
*** jmasud has quit IRC | 00:10 | |
*** tosky has quit IRC | 00:14 | |
*** holser has joined #oooq | 00:14 | |
*** jmasud has joined #oooq | 00:18 | |
*** holser has quit IRC | 00:40 | |
*** jmasud has quit IRC | 00:43 | |
*** jmasud has joined #oooq | 00:45 | |
*** jmasud has quit IRC | 01:03 | |
*** jmasud has joined #oooq | 01:04 | |
*** jmasud has quit IRC | 01:08 | |
*** jmasud has joined #oooq | 01:10 | |
*** jmasud has quit IRC | 01:25 | |
*** sshnaidm is now known as sshnaidm|afk | 01:31 | |
*** jmasud has joined #oooq | 01:43 | |
*** jmasud has quit IRC | 01:50 | |
*** jmasud has joined #oooq | 01:53 | |
*** jmasud has quit IRC | 01:57 | |
*** jmasud has joined #oooq | 02:08 | |
*** jmasud has quit IRC | 02:10 | |
*** apetrich has quit IRC | 03:09 | |
*** ysandeep|afk is now known as ysandeep | 03:14 | |
*** ykarel|away is now known as ykarel | 03:37 | |
*** jmasud has joined #oooq | 03:44 | |
*** udesale has joined #oooq | 04:25 | |
*** ratailor has joined #oooq | 04:40 | |
*** jmasud has quit IRC | 04:54 | |
*** jmasud has joined #oooq | 04:55 | |
*** skramaja has joined #oooq | 05:19 | |
*** pojadhav|afk is now known as pojadhav | 05:26 | |
*** sanjayu_ has joined #oooq | 05:42 | |
*** holser has joined #oooq | 06:09 | |
*** irclogbot_0 has quit IRC | 06:29 | |
*** holser__ has joined #oooq | 06:31 | |
*** holser has quit IRC | 06:33 | |
*** dpawlik has joined #oooq | 06:58 | |
*** jfrancoa has joined #oooq | 07:01 | |
*** marios has joined #oooq | 07:05 | |
*** ratailor has quit IRC | 07:05 | |
*** ratailor has joined #oooq | 07:08 | |
chandankumar | marios, morning, fs035, fs020, fs01 failed with pcs issue rechecked via https://review.rdoproject.org/r/25909 and baremetal fs01 cirros failure recheck via https://review.rdoproject.org/r/#/c/25865/ rest is in control | 07:15 |
---|---|---|
marios | chandankumar: o/ looking | 07:23 |
marios | chandankumar: ack so we don't have confirmed issue yet for fs035/1/20 yet i.e. waiting on the test run | 07:24 |
marios | chandankumar: but if all 3 failed in same way then its likely a legit issue | 07:24 |
*** irclogbot_0 has joined #oooq | 07:29 | |
*** bogdando has joined #oooq | 07:39 | |
*** bogdando_ has joined #oooq | 07:42 | |
*** bogdando has quit IRC | 07:44 | |
*** bogdando_ is now known as bogdando | 07:44 | |
*** holser__ has quit IRC | 07:54 | |
*** holser has joined #oooq | 07:54 | |
*** matbu has joined #oooq | 08:01 | |
*** dpawlik has quit IRC | 08:07 | |
*** dpawlik has joined #oooq | 08:07 | |
*** tesseract has joined #oooq | 08:12 | |
*** dpawlik has quit IRC | 08:18 | |
*** amoralej|off is now known as amoralej | 08:20 | |
marios | panda: rebased the https://review.rdoproject.org/r/#/c/24774/ but i saw the molecule was still red on those is that a thing or maybe cleared now ? | 08:23 |
marios | rebasing mine ontop again | 08:24 |
marios | are we gonna merge this today then | 08:24 |
marios | panda: ? | 08:24 |
*** dpawlik has joined #oooq | 08:28 | |
*** apetrich has joined #oooq | 08:31 | |
chandankumar | marios, I am adding the layout for scenario 12 in periodic pipeline | 08:33 |
*** rascasoft has joined #oooq | 08:35 | |
marios | chandankumar: ack the comment was more about if the centos7 version was currently being used in periodics but it isn't thanks | 08:38 |
*** ykarel is now known as ykarel|lunch | 08:39 | |
*** jpena|off is now known as jpena | 08:50 | |
*** jmasud has quit IRC | 08:54 | |
*** jmasud has joined #oooq | 08:56 | |
*** ccamacho has joined #oooq | 08:57 | |
chandankumar | marios, s12 entry https://review.rdoproject.org/r/#/q/topic:centos-8-fs12+(status:open+OR+status:merged) | 09:00 |
*** jaosorior has joined #oooq | 09:01 | |
*** tosky has joined #oooq | 09:07 | |
dpawlik | sshnaidm|afk, marios could you confirm that all of your jobs are using at least fedora 30? | 09:08 |
marios | dpawlik: yes and if they aren't (e.g. f28 we *used* to use) I dont think they even matter any more... don't thnk we have any f28 jobs left... checking | 09:09 |
dpawlik | if this one will be merged, https://review.opendev.org/#/c/713177 jobs will fail if they are running on f29/f28 | 09:09 |
marios | dpawlik: chandankumar: https://review.rdoproject.org/zuul/builds?job_name=periodic-tripleo-ci-fedora-28-ovb-1ctlr_1comp-featureset002-master-upload (2019-03-11) | 09:10 |
marios | dpawlik: perhaps also send an email to pcci/rdo about that? | 09:11 |
marios | weshay|ruck: fyi 11:09 < dpawlik> if this one will be merged, https://review.opendev.org/#/c/713177 jobs will fail if they are running on f29/f28 | 09:12 |
marios | thanks dpawlik don't worry about the mail i thought that was your review ^ | 09:12 |
marios | anyone have link to the current ruck|rover pad/hackmd | 09:13 |
chandankumar | marios, https://hackmd.io/7MBqFHurTA2e5H8kYRwgag | 09:13 |
marios | chandankumar: thanks, they didn't update grafana link | 09:13 |
dpawlik | marios, I will send email after wesh confirm that those jobs will be updated soon, ok? | 09:14 |
marios | dpawlik: ack added to the https://hackmd.io/7MBqFHurTA2e5H8kYRwgag for now | 09:15 |
dpawlik | marios, but wait. Those jobs are failing | 09:15 |
panda | > | 09:16 |
marios | < | 09:16 |
panda | marios: always the pessimist. | 09:17 |
marios | ;) avoids disappointment | 09:17 |
*** sanjayu_ has quit IRC | 09:22 | |
panda | marios: we got 6 promotions in the weekend, I think we are very close to merge | 09:30 |
marios | panda: nice | 09:31 |
*** arxcruz has joined #oooq | 09:32 | |
arxcruz | jesus, my bouncer was crazy :/ | 09:32 |
chandankumar | arxcruz, fyi https://bugs.launchpad.net/tripleo/+bug/1867599 | 09:32 |
openstack | Launchpad bug 1867599 in tripleo "overcloud deploy failing on fs030 and fs016 while pulling mariadb container from undercloud registry" [Critical,Confirmed] | 09:32 |
arxcruz | I knew something was wrong when I saw nobody pinging me this morning :D | 09:33 |
*** arxcruz is now known as arxcruz|rover | 09:34 | |
zbr | who can help me with reviews on some upstream zuul-jobs changes? (even if you are not core) | 09:35 |
zbr | if i get reviews before i ping infra, they merge much faser. | 09:35 |
arxcruz|rover | akahat: hey, is it working the mistral tests? I'm checking https://zuul.openstack.org/builds?job_name=tripleo-ci-centos-7-undercloud-containers but can't find a mistral run can you double check so i can close the bug ? | 09:35 |
arxcruz|rover | zbr: sure | 09:36 |
zbr | arxcruz|rover: https://review.opendev.org/#/c/708642/ | 09:37 |
chandankumar | arxcruz|rover, one more https://bugs.launchpad.net/tripleo/+bug/1867602 | 09:59 |
openstack | Launchpad bug 1867602 in tripleo "overcloud deploy failed with Systemd start for pcsd failed " [Critical,Confirmed] | 09:59 |
*** ykarel|lunch is now known as ykarel | 09:59 | |
arxcruz|rover | chandankumar: ack | 09:59 |
*** saneax has joined #oooq | 10:01 | |
marios | chandankumar: so did they all fail like .URLError: <urlopen error [Errno 101] Network is unreachable> | 10:03 |
marios | chandankumar: was looking at fs1 now | 10:03 |
marios | * https://logserver.rdoproject.org/openstack-component-baremetal/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-ovb-3ctlr_1comp-featureset001-baremetal-master/61ab8a3/job-output.txt | 10:04 |
chandankumar | marios, it was a network issue, | 10:04 |
chandankumar | marios, rerunned it here https://review.rdoproject.org/r/#/c/25865/ sorry cirros website issue | 10:05 |
chandankumar | and it passed | 10:05 |
marios | chandankumar: ack cool | 10:06 |
zbr | marios: did you see my email re https://review.rdoproject.org/r/#/c/25904/ ? | 10:07 |
dpawlik | marios, but the periodic jobs last time execution was very long time ago | 10:08 |
dpawlik | probaly sshnaidm|afk has changed that, or? | 10:08 |
*** sshnaidm|afk is now known as sshnaidm | 10:08 | |
sshnaidm | dpawlik, sorry, what are we talking about? | 10:09 |
dpawlik | sshnaidm|afk, http://paste.openstack.org/show/790728/ | 10:09 |
dpawlik | sshnaidm, tl;dr does all of tripleo job running on at least f30? | 10:10 |
sshnaidm | dpawlik, afaik yes | 10:10 |
dpawlik | sshnaidm, because if it will be merged https://review.opendev.org/#/c/713177 jobs with f29 will fail | 10:10 |
sshnaidm | dpawlik, there is some trash here: https://github.com/rdo-infra/rdo-jobs/blob/master/zuul.d/deprecated-jobs.yaml#L10-L29 | 10:11 |
sshnaidm | dpawlik, ok, if it will fail we'll know that we still have such jobs :) | 10:11 |
dpawlik | sshnaidm, ah! Thats why marios found info about that job | 10:11 |
dpawlik | lol | 10:11 |
*** ratailor has quit IRC | 10:11 | |
dpawlik | so lets vote https://review.opendev.org/#/c/713177 | 10:12 |
*** ratailor has joined #oooq | 10:13 | |
dpawlik | and sshnaidm pls vote here https://review.opendev.org/#/c/713169/. Thanks! | 10:15 |
marios | 11:10 < marios> dpawlik: chandankumar: | 10:15 |
marios | https://review.rdoproject.org/zuul/builds?job_name=periodic-tripleo-ci-fedora-28-ovb-1ctlr_1comp-featureset002-master-upload (2019-03-11) | 10:15 |
marios | dpawlik: ack yeah i said they ran long time ago ^^ | 10:15 |
chandankumar | marios, https://review.rdoproject.org/r/#/c/25915/ | 10:20 |
chandankumar | please +w it | 10:20 |
marios | also change the upstream version? chandankumar | 10:21 |
marios | chandankumar: (ack done) | 10:21 |
chandankumar | marios, changing it | 10:23 |
chandankumar | marios, https://review.opendev.org/713184 | 10:26 |
ykarel | chandankumar, but in upstream it was already running | 10:28 |
ykarel | so likely those parameters can just be dropped in periodic | 10:29 |
ykarel | instead of changing upstream also | 10:29 |
marios | ykarel: was it? i thought it was only 10-ovn that ran tempest | 10:30 |
chandankumar | ykarel, marios i got the issue https://github.com/openstack/tripleo-quickstart/blob/master/config/general_config/featureset062.yml#L35 | 10:30 |
chandankumar | it is set to true no need to enable there, /me abandon upstream patch | 10:31 |
marios | chandankumar: ack on fs62 | 10:31 |
ykarel | chandankumar, ack | 10:33 |
*** ykarel is now known as ykarel|afk | 10:40 | |
*** jbadiapa has joined #oooq | 10:55 | |
marios | chandankumar: did you see that tempest one * https://logserver.rdoproject.org/openstack-component-compute/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-standalone-full-tempest-compute-master/afd840d/logs/undercloud/var/log/tempest/stestr_results.html.gz | 10:58 |
marios | * test_stamp_pattern[compute,id-10fd234a-515c-41e5-b092-8323060598c5 | 10:58 |
marios | chandankumar: ran saturday looks like | 10:58 |
weshay|ruck | ysandeep, setting up.. few min | 10:59 |
ysandeep | weshay|ruck, sure | 11:00 |
*** dtantsur has joined #oooq | 11:02 | |
weshay|ruck | arxcruz|rover, 0 | 11:03 |
weshay|ruck | 0. | 11:03 |
weshay|ruck | o/ | 11:03 |
arxcruz|rover | weshay|ruck: o/ | 11:03 |
weshay|ruck | how about that.. arxcruz|rover need anything? | 11:04 |
arxcruz|rover | too earlier for you no? | 11:04 |
arxcruz|rover | weshay|ruck: we have two blockers | 11:04 |
arxcruz|rover | weshay|ruck: https://hackmd.io/7MBqFHurTA2e5H8kYRwgag?both#TripleO-ISSUES | 11:04 |
arxcruz|rover | chandankumar had open the bugs | 11:04 |
weshay|ruck | looking | 11:04 |
weshay|ruck | k.. think I saw them | 11:04 |
arxcruz|rover | brb 10 min | 11:05 |
weshay|ruck | arxcruz|rover, hey.. afaict the baremetal component promoted today.. 3/16.. it should not have, http://dashboard-ci.tripleo.org/d/UDA4H3aZk/component-pipeline?orgId=1&from=now-7d&to=now | 11:06 |
weshay|ruck | https://logserver.rdoproject.org/openstack-promote-component/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-centos-8-master-component-baremetal-promote-to-promoted-components/2feb388/job-output.txt | 11:06 |
weshay|ruck | can you please check that out | 11:06 |
*** ratailor_ has joined #oooq | 11:07 | |
chandankumar | weshay|ruck, baremetal fs01 failure was a transient issue | 11:09 |
*** ratailor has quit IRC | 11:10 | |
weshay|ruck | chandankumar, ya. but it promoted | 11:12 |
weshay|ruck | ysandeep, join #openstack-infra | 11:14 |
chandankumar | marios, you are working on adding fs030 to periodic pipeline if not taking it over then? | 11:15 |
chandankumar | weshay|ruck, with this patch goes in https://review.opendev.org/712940 fs012 is working without tempest | 11:28 |
chandankumar | *sc12 | 11:28 |
weshay|ruck | NICE | 11:29 |
chandankumar | weshay|ruck, please https://review.opendev.org/#/c/712294/ +w it | 11:29 |
*** ykarel|afk is now known as ykarel | 11:29 | |
marios | chandankumar: no didn't dig there yet go ahead | 11:29 |
*** rlandy has joined #oooq | 11:29 | |
chandankumar | marios, ok then | 11:30 |
arxcruz|rover | should not have what ? | 11:31 |
chandankumar | rlandy, morning | 11:39 |
rlandy | chandankumar: hey | 11:39 |
chandankumar | rlandy, I opened a bug related to fs030 https://bugs.launchpad.net/tripleo/+bug/1867599 registry issue you are working on that? | 11:40 |
openstack | Launchpad bug 1867599 in tripleo "overcloud deploy failing on fs030 and fs016 while pulling mariadb container from undercloud registry" [Critical,Confirmed] | 11:40 |
chandankumar | I have added it to cix, it is also blocking slawq work | 11:40 |
rlandy | chandankumar: ok | 11:40 |
chandankumar | rlandy, https://review.rdoproject.org/r/#/q/topic:centos-8-fs12+(status:open+OR+status:merged) good to go | 11:41 |
rlandy | chandankumar: https://hackmd.io/HrQd03c9SxOMtFPFrq50tg?both is out of date | 11:42 |
rlandy | we should pull out what we still need there | 11:42 |
chandankumar | rlandy, yes sure | 11:42 |
chandankumar | rlandy, may be we can sit after scrum, get it cleaned | 11:43 |
sshnaidm | chandankumar, does scenario012 pass? | 11:43 |
rlandy | chandankumar: looks like we have some tempest success http://tripleo-cockpit.usersys.redhat.com/d/2tivP9BWz/component-pipeline | 11:43 |
chandankumar | sshnaidm, expect tempest, all is working | 11:43 |
chandankumar | sshnaidm, https://review.opendev.org/#/c/712940/ will fix current issue | 11:43 |
rlandy | either way - we might as well move scenario012 | 11:43 |
rlandy | no point in keeping it centos7 passing or not | 11:44 |
rlandy | chandankumar: who would investigate this? https://bugs.launchpad.net/tripleo/+bug/1867599 | 11:45 |
openstack | Launchpad bug 1867599 in tripleo "overcloud deploy failing on fs030 and fs016 while pulling mariadb container from undercloud registry" [Critical,Confirmed] | 11:45 |
rlandy | is it CIX'ed for CI team or for a DF? | 11:46 |
chandankumar | rlandy, insecure registry missing on subnode1 | 11:46 |
chandankumar | https://logserver.rdoproject.org/50/25550/8/check/periodic-tripleo-ci-centos-8-multinode-1ctlr-featureset030-master/f75e2e4/logs/subnode-1/etc/containers/registries.conf.txt.gz | 11:46 |
chandankumar | rlandy, may be DF one | 11:46 |
rlandy | chandankumar: yep I know the failure - ok I see you have put it with PCCI atm | 11:47 |
chandankumar | rlandy, ok | 11:48 |
rlandy | weshay|ruck: chandankumar: looking at escalation board - are well still keeping all the OVB cards alive? | 11:49 |
rlandy | also https://trello.com/c/6agowoQH/1388-cixlp1867177tripleociproa-cannot-download-noarch-python3-cssselect-092-13el8noarchrpm | 11:50 |
chandankumar | rlandy, weshay|ruck since fs01 is passing and fs020 is known, so i think we can move it done? | 11:50 |
rlandy | can we close this out? | 11:50 |
rlandy | chandankumar: ^^ python3-cssselect-092-13el8noarchrpm - looks ok for some time? | 11:51 |
chandankumar | rlandy, yes, I have not seen this issue from last week till now | 11:51 |
rlandy | gates seem to be moving. | 11:51 |
rlandy | arxcruz|rover: ^^? | 11:51 |
arxcruz|rover | rlandy: yup | 11:52 |
weshay|ruck | arxcruz|rover, hey.. may have a blocking red job in stein fyi | 11:53 |
rlandy | https://trello.com/c/mpFqJeuO/1377-cixlp1866202tripleociproa-ovb-on-centos8-fails-because-of-networking-failures | 11:53 |
weshay|ruck | looking at the cockpit.. pass rate is 53% | 11:53 |
arxcruz|rover | weshay|ruck: checking | 11:53 |
arxcruz|rover | weshay|ruck: which one? there are scen000 scen009 and container-update | 11:55 |
arxcruz|rover | upgrades is failing with this: | 11:55 |
arxcruz|rover | 2020-03-15 00:26:58 | "msg": "Error: Package: ceph-ansible-4.0.14-1.el7.noarch (quickstart-centos-ceph-nautilus)\n Requires: ansible >= 2.8\n Installed: ansible-2.6.19-1.el7.ans.noarch (@delorean-rocky-deps)\n ansible = 2.6.19-1.el7.ans\n", | 11:55 |
marios | rlandy: thanks for review can you please check my comment @ https://review.opendev.org/#/c/711507/3/zuul.d/standalone-jobs.yaml when you next have a minute thanks... just wrt the duplication not sure it's necessary | 11:56 |
rlandy | marios: ack | 11:57 |
arxcruz|rover | ykarel: upgrade from rocky to stein is failing because ansible package... need your help here :) | 11:57 |
rlandy | how did baremetal promote when it failed OVB? | 11:58 |
chandankumar | rlandy, rerun the job | 11:58 |
rlandy | chandankumar: ah | 11:58 |
chandankumar | *rerunned | 11:58 |
rlandy | chandankumar: ok - so we're not entirely out of the woods with baremetal jobs | 11:58 |
chandankumar | rlandy, yes | 11:59 |
chandankumar | things seems to be fine | 11:59 |
ykarel | arxcruz|rover, but ansible is not changed for long in rocky | 11:59 |
ykarel | arxcruz|rover, more context please | 11:59 |
rlandy | chandankumar: ok - so I'll leave the networking card open for a one more meeting | 11:59 |
chandankumar | rlandy, aye | 11:59 |
rlandy | look slike we only started to see success today | 11:59 |
chandankumar | rlandy, brb for an hour | 11:59 |
rlandy | chandankumar: ack | 11:59 |
ykarel | in rocky we have 2.6.9 | 12:00 |
ykarel | ansible | 12:00 |
ykarel | so likely ceph-ansible updated recently? | 12:00 |
arxcruz|rover | ykarel: https://b8ad378ded78e3ab6d47-59bc25a858d3cf4494b5a72aa2369a4f.ssl.cf5.rackcdn.com/713096/2/check/tripleo-ci-centos-7-containerized-undercloud-upgrades/6bc39ab/logs/undercloud/home/zuul/undercloud_upgrade.log | 12:00 |
ykarel | looking | 12:01 |
ykarel | arxcruz|rover, looks like last success for that job is in june 2019 http://zuul.openstack.org/builds?job_name=tripleo-ci-centos-7-containerized-undercloud-upgrades&branch=stable%2Fstein&result=SUCCESS | 12:03 |
ykarel | in stein | 12:03 |
arxcruz|rover | :O | 12:03 |
arxcruz|rover | weshay|ruck: what's the blocking red job in stein ? | 12:03 |
weshay|ruck | arxcruz|rover, there are only like 2-3 patches in stein over the last 2 days.. | 12:04 |
weshay|ruck | so.. may not be something actionable yet | 12:05 |
arxcruz|rover | ok | 12:05 |
ykarel | arxcruz|rover, so it's happening due to https://opendev.org/openstack/tripleo-heat-templates/src/branch/master/deployment/undercloud/undercloud-upgrade.yaml#L197-L202 | 12:09 |
ykarel | even if ansible 2.8 exists in stein | 12:09 |
*** ysandeep is now known as ysandeep|afk | 12:11 | |
weshay|ruck | panda, morning.. how'd the promotions over the weekend go? | 12:12 |
marios | weshay|ruck: fyi 11:30 < panda> marios: we got 6 promotions in the weekend, I think we are very close to merge | 12:14 |
weshay|ruck | marios, /me going through the logs | 12:17 |
rlandy | marios: responded in https://review.opendev.org/#/c/711507/ | 12:22 |
rlandy | we can discuss further at the meeting | 12:22 |
rlandy | imho, it makes it cleaner going forward to keep the main var set on the centos-8 job | 12:22 |
rlandy | and override the centos-7 job | 12:23 |
rlandy | but in any case ... | 12:23 |
rlandy | we made the decision while discussing parenting to duplicate | 12:23 |
panda | lunch | 12:25 |
marios | rlandy: ok replied again for clarification but will duplicate them anyway | 12:25 |
rlandy | marios: we can clarify at meeting | 12:25 |
rlandy | it should be clear to everyone | 12:25 |
*** udesale_ has joined #oooq | 12:26 | |
rlandy | this is a pretty funndamental decision we made | 12:26 |
marios | rlandy: ack | 12:26 |
*** ratailor__ has joined #oooq | 12:27 | |
*** udesale has quit IRC | 12:28 | |
*** jmasud has quit IRC | 12:28 | |
*** ratailor_ has quit IRC | 12:30 | |
*** ratailor__ has quit IRC | 12:39 | |
zbr | weshay|ruck: should tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades become 8? | 12:45 |
zbr | AFAIK, on upgrades are incompatible with os switching | 12:45 |
zbr | because there is no such thing as upgrading centos-7 to 8. | 12:45 |
weshay|ruck | zbr, look at the hackmd | 12:46 |
weshay|ruck | for centos-8 | 12:46 |
rlandy | arxcruz|rover: want to join me on the escalation meeting? | 12:46 |
rlandy | nv - we're done :) | 12:47 |
zbr | yep, not supported but this does not explain what we do with existing jobs | 12:47 |
zbr | or this means that tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades must be removed | 12:47 |
arxcruz|rover | rlandy: sorry, I don't have this cix on my calendar, is it a new meeting, it's not in rhos team meetings calendar | 12:47 |
zbr | afaik, all jobs with centos-8 in their name are supposed to be replaced by centos-8, or removed. | 12:48 |
rlandy | arxcruz|rover: no worries - it was mainly centos-8 so I didn't stumble too much:) | 12:48 |
weshay|ruck | ugh | 12:48 |
weshay|ruck | arxcruz|rover, working w/ jbuchta to fix that.. and get it added back to rhos-team-meetings | 12:51 |
arxcruz|rover | weshay|ruck: ack | 12:51 |
arxcruz|rover | weshay|ruck: I got invitation, for today, but it should be on rhos-team-meetings | 12:52 |
rlandy | https://hackmd.io/HrQd03c9SxOMtFPFrq50tg is getting out of date | 12:52 |
weshay|ruck | arxcruz|rover, they are working that out now | 12:54 |
weshay|ruck | rlandy, ya.. I did not rescrub that | 12:54 |
weshay|ruck | I can do that while I listen to this stuff | 12:54 |
rlandy | weshay|ruck: worth cleaning it up at today's scrum? | 12:55 |
rlandy | we can take care of it | 12:55 |
rlandy | weshay|ruck: right now, zuul and cockpit are better trackers | 12:55 |
weshay|ruck | rlandy, so.. if you guys have time.. def.. dive into the promotion server | 12:56 |
rlandy | weshay|ruck: I think the rest is fairly sorted | 12:56 |
rlandy | a couple of failing test to cover | 12:56 |
weshay|ruck | marios, panda I don't see a full successful run on centos-8 panda-rulz etc.. but I may be out of context.. I see containers work.. but it stops at overcloud-images.. which may be intentional | 12:56 |
marios | weshay|ruck: ack call in 3 mins | 12:57 |
weshay|ruck | rlandy, I would have that the centos-7 upgrade jobs would have started to fail by now in master, but are not.. but can be killed | 12:57 |
rlandy | weshay|ruck: ok - will get some reviews up on that | 12:58 |
rlandy | after scrum | 12:58 |
*** jpena is now known as jpena|lunch | 12:59 | |
rlandy | chandankumar: scrum | 13:01 |
marios | * https://logserver.rdoproject.org/openstack-component-compute/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-standalone-full-tempest-compute-master/afd840d/logs/undercloud/var/log/tempest/stestr_results.html.gz | 13:04 |
marios | rlandy: ^ | 13:04 |
*** amoralej is now known as amoralej|lunch | 13:07 | |
*** apetrich has quit IRC | 13:25 | |
weshay|ruck | chandankumar, fyi https://review.rdoproject.org/zuul/builds?job_name=tripleo-podman-integration-rhel-8-standalone | 13:28 |
weshay|ruck | we have centos-8 going there yet? | 13:28 |
chandankumar | weshay|ruck, yes, https://review.rdoproject.org/r/#/c/25916/ | 13:28 |
weshay|ruck | chandankumar, thanks.. | 13:28 |
weshay|ruck | ykarel, fyi https://review.rdoproject.org/r/#/c/25916/ | 13:28 |
*** dpawlik has quit IRC | 13:29 | |
*** apetrich has joined #oooq | 13:30 | |
ykarel | weshay|ruck, ack noted | 13:30 |
ykarel | we will ignore rhel8 master failures | 13:30 |
chandankumar | weshay|ruck, for ceph-ansible is it ok to drop rhel-8 jobs and move to centos-8? | 13:36 |
weshay|ruck | chandankumar, yes.. all rhel8 goes bye bye | 13:40 |
marios | https://review.rdoproject.org/r/#/c/25782/ | 13:40 |
marios | rlandy: ^ | 13:40 |
chandankumar | bhagyashris, ^^ | 13:41 |
weshay|ruck | panda++ | 13:41 |
*** ChanServ sets mode: +o panda | 13:43 | |
*** panda changes topic to ""Docs: https://docs.openstack.org/tripleo-quickstart/latest/ || 16th March promoter patch to review https://review.rdoproject.org/r/24774 || next promoter patch in the chain https://review.rdoproject.org/r/25782" | 13:43 | |
*** panda sets mode: -o panda | 13:43 | |
*** skramaja has quit IRC | 13:43 | |
*** skramaja has joined #oooq | 13:44 | |
*** ysandeep|afk is now known as ysandeep | 13:45 | |
*** TrevorV has joined #oooq | 13:46 | |
rlandy | marios: sorry - we didn't discuss the duplication of vars | 13:49 |
marios | rlandy: nm i updated anyway | 13:50 |
rlandy | want to chat about that? | 13:50 |
marios | rlandy: discussion in gerrit if any more is needed np | 13:50 |
rlandy | marios: it's what we did above | 13:50 |
rlandy | marios: ack | 13:50 |
*** jpena|lunch is now known as jpena | 13:50 | |
chandankumar | rlandy, marios please +w it https://review.opendev.org/#/c/712294/ | 13:50 |
chandankumar | https://review.opendev.org/#/c/712013/ needs that | 13:51 |
sshnaidm | panda, is the promoter code py3 ready? | 13:52 |
sshnaidm | panda, I wonder if we can move promoted to centos8 in some point | 13:52 |
marios | sshnaidm: yes see py36 unit tests | 13:52 |
sshnaidm | marios, ack | 13:52 |
sshnaidm | maybe worth a try | 13:53 |
marios | chandankumar: rlandy: beat me to it | 13:53 |
bhagyashris | chandankumar, weshay|ruck yup working on it chandankumar | 13:53 |
marios | sshnaidm: yeah good idea... for our 'main' promoter server once we move to 'new' | 13:54 |
marios | panda: did you already consider this? ^ | 13:54 |
marios | panda: 'this' use centos8 for promoter server | 13:55 |
panda | marios: sshnaidm we already discussed this with wes, move to centos8 will be a task in next sprint. The code is tested in both py27 and py36, so that can be ported without problems. The only part taht sill needs change is the provision playbook. | 13:57 |
marios | panda: cool story bro | 13:57 |
sshnaidm | marios, :D | 13:58 |
marios | ;) | 13:58 |
rlandy | chandankumar: marios: https://hackmd.io/7MBqFHurTA2e5H8kYRwgag - added section on Reviews still in play to add/move jobs | 14:00 |
marios | rlandy: ack | 14:00 |
marios | rlandy: will do in bit | 14:00 |
rlandy | ykarel: hi ... wrt the email on md5 and mirrors, "using md5sum command to detect aggregate hash which is wrong as repo files can be modified in jobs" | 14:03 |
rlandy | we only do this on the container build jobs | 14:03 |
rlandy | and it was bu design | 14:03 |
rlandy | by | 14:03 |
ykarel | rlandy, but checksum changes after modificatin of repo files | 14:04 |
ykarel | and that's wrong imo to use md5sum command on modified repo files | 14:04 |
rlandy | ykarel: the repo files should not change in container build jobs | 14:04 |
ykarel | rlandy, but those are changing | 14:04 |
ykarel | due to mirrors | 14:05 |
rlandy | if we had not done that, we would never had known they were changing | 14:05 |
ykarel | yes issue exist for long but was not noticed | 14:05 |
rlandy | it's an easy fix to pull the md5 | 14:05 |
ykarel | rlandy, yes i filed bug https://bugs.launchpad.net/tripleo/+bug/1867580 and proposed patch too | 14:06 |
openstack | Launchpad bug 1867580 in tripleo "tripleo-buildcontainers jobs build containers with wrong tag" [High,In progress] - Assigned to yatin (yatinkarel) | 14:06 |
rlandy | ykarel: ok - so if we do go with this change, we will lose our check | 14:09 |
rlandy | ie: the mirrors are out of date | 14:09 |
*** Goneri has joined #oooq | 14:09 | |
rlandy | the mirror we reference updates about 1.5 hours after the job is run | 14:10 |
rlandy | and we build images with out of date repos | 14:10 |
ykarel | rlandy, u saying mirror updates after 1.5 hours? | 14:11 |
ykarel | sorry didn't get that part | 14:11 |
rlandy | ykarel: yep - the mirrors lag | 14:11 |
ykarel | :( i doubt 1.5 is too much | 14:12 |
ykarel | it should sync in seconds or minutes | 14:12 |
rlandy | even if it is minutes, it's too late for the pipeline | 14:12 |
rlandy | so I am not saying your patch is wrong, but using mirror is not good either | 14:13 |
ykarel | i don't have numbers, but i think jpena can put some light on how much it takes for mirrors to sync | 14:14 |
ykarel | i think here we mainly looking at promotion jobs, so rdo mirror, right? | 14:14 |
*** amoralej|lunch is now known as amoralej | 14:14 | |
rlandy | ykarel: ack | 14:15 |
jpena | rlandy, ykarel: is that the AFS mirrors? I can check the expiration time, give me some mins | 14:16 |
*** Goneri has quit IRC | 14:16 | |
rlandy | jpena: thanks - maybe we just hit a particular lag on friday | 14:16 |
ykarel | jpena, http://mirror.regionone.rdo-cloud-tripleo.rdoproject.org:8080/rdo one | 14:16 |
ykarel | ack take ur time | 14:16 |
*** chem has quit IRC | 14:17 | |
*** Goneri has joined #oooq | 14:19 | |
*** chem has joined #oooq | 14:19 | |
*** udesale_ has quit IRC | 14:23 | |
jpena | ykarel, rlandy: the default cache expiry time is 1 hour | 14:26 |
rlandy | jpena: so iiuc, the mirrors update once per hour | 14:29 |
jpena | rlandy: they expire content every hour the latest. They can expire content before that, but it depends on the web server sending certain data (I can't remember off the top of my head, there is some HTTP expiry time header iirc) | 14:30 |
ykarel | jpena, so isn't cache expiry and mirror sync different? | 14:30 |
jpena | ykarel: they are two different things. In the AFS mirrors, we have two separate entities: | 14:31 |
ykarel | hmm that what i understood, so here question was for mirror sync | 14:31 |
jpena | 1- the mirror side, which is synced periodically (that's the centos, fedora, debian, etc mirrors) | 14:31 |
jpena | 2- the caching proxy side (used for RDO Trunk data iirc) | 14:31 |
rlandy | we need to reference the RDO Trunk data | 14:32 |
rlandy | either way, "periodically" could mean a miss for the pipelines | 14:33 |
rlandy | which need to run immediately after the tripleo-ci-testing pin is updated | 14:33 |
rlandy | https://review.rdoproject.org/zuul/builds?pipeline=openstack-periodic-master | 14:33 |
jpena | right, so if you need to reference the delorean.repo file, I would advise to get it from trunk.rdo | 14:36 |
rlandy | which is why I think we should avoid mirrors for all pipeline jobs | 14:36 |
rlandy | ykarel: ^^ we can keep your patch or not | 14:36 |
ykarel | rlandy, patch is using trunk.rdo | 14:37 |
rlandy | I'm ok with the change if we are now sure we are getting trunk.rdo | 14:37 |
rlandy | the whole point in doing the md5 check before was to catch that | 14:37 |
ykarel | that patch is more on topic for using delorean.repo.md5 or md5sum command | 14:37 |
rlandy | ykarel: so would you agree that we skip the mirror-info-fork for all jobs running in a periodic pipeline? | 14:38 |
ykarel | if md5sum was kept to check consistency, then i think it should be handled in the job itsel | 14:39 |
rlandy | ykarel: I'm ok with your change | 14:39 |
ykarel | rlandy, mirror-info fork runs already in deployment jobs | 14:39 |
ykarel | it was missing in other jobs | 14:39 |
rlandy | https://review.rdoproject.org/r/#/c/25900/3/playbooks/base/pre.yaml | 14:40 |
rlandy | ^^ that pre runs on non-deployment jobs | 14:40 |
chandankumar | rlandy, marios all reviews added here https://hackmd.io/7MBqFHurTA2e5H8kYRwgag?view#Reviews-still-in-play-to-addmove-jobs from my side | 14:46 |
rlandy | chandankumar: thanks ... pls see discussion above | 14:47 |
rlandy | jpena: if we made all periodic jobs hit trunk.rdo directly ( and not the mirrors) could that cause any issues? | 14:48 |
jpena | rlandy: we've never tested the load... My suggestion would be to use trunk.rdo to fetch the .repos, then locally modify them to use the mirrors if needed | 14:49 |
rlandy | ok - that's what the release files do anyways - so it's just the jobs that don't use release files (container and image build that need to adjust) | 14:51 |
chandankumar | rlandy, so basically we need to make sure the delorean.md5 should be same with the nodepool proxy? | 15:00 |
rlandy | chandankumar: that's half the problem | 15:02 |
rlandy | the code to recalculate the md5 sum was intentional | 15:02 |
rlandy | to check this | 15:02 |
rlandy | ykarel has a patch to change the code to download the md5 rather than recalculate it | 15:03 |
rlandy | which is ok | 15:03 |
rlandy | but lose our check | 15:03 |
rlandy | in such a case, we need to make sure that mirrors are not used for the image and container builds | 15:03 |
rlandy | iiuc - for any releases | 15:03 |
rlandy | essentially the proxy can be out of date so we need to hit trunk.rdo directly in the pipeline | 15:04 |
rlandy | iow ... https://review.rdoproject.org/r/#/c/25900/3/playbooks/base/pre.yaml for all jobs where we do not deploy | 15:05 |
chandankumar | rlandy, https://review.rdoproject.org/r/#/c/25900/3/playbooks/base/pre.yaml but it is not going to handle image build na? | 15:07 |
rlandy | chandankumar: correct - see the email suggestion | 15:07 |
rlandy | chandankumar: but I think it needs to be done not only for centos8 | 15:08 |
rlandy | putting up patch - sec | 15:08 |
chandankumar | rlandy, ok | 15:08 |
chandankumar | arxcruz|rover, weshay|ruck rlandy https://lists.rdoproject.org/pipermail/dev/2020-March/009333.html | 15:17 |
rlandy | chandankumar: https://review.rdoproject.org/r/25926 | 15:17 |
rlandy | ^^ so I am not sure how wide spread this should be | 15:17 |
chandankumar | rlandy, build_override_repos will work I think | 15:19 |
rlandy | chandankumar: what about kolla_base_tag? | 15:20 |
rlandy | will that only get periodic | 15:20 |
rlandy | I think 8 pr 7 doesn't matter | 15:20 |
rlandy | I removed the original 8 only | 15:20 |
arxcruz|rover | curl https://corona-stats.online/USA | 15:21 |
arxcruz|rover | now that's cool | 15:21 |
chandankumar | rlandy, kolla_base_tag is used only for rhel-8 or c-8 not seeing for c7 | 15:21 |
rlandy | chandankumar: so I think rhel 8 should also be included | 15:22 |
rlandy | arxcruz|rover: /o\ ... this whole city is a ghost town | 15:22 |
*** Goneri has quit IRC | 15:22 | |
rlandy | other than the grocery store - which is a mob scene | 15:23 |
rlandy | ykarel: pls review https://review.rdoproject.org/r/#/c/25926/ - to check if we've got the right conditions | 15:24 |
chandankumar | arxcruz|rover, https://www.covidindia.com/ | 15:24 |
ykarel | rlandy, looking | 15:24 |
arxcruz|rover | chandankumar: the curl command says 113, the website says 43 | 15:26 |
chandankumar | arxcruz|rover, https://covidindia.org/ | 15:26 |
chandankumar | arxcruz|rover, people are crazy they made so many sites | 15:26 |
*** sshnaidm is now known as sshnaidm|afk | 15:28 | |
arxcruz|rover | rlandy: I went to the market today, nothing there, but I notice people are walking on the streets normally, although, the kindergarten and schools are now officially closed | 15:29 |
arxcruz|rover | and yes, I was able to buy toilet paper | 15:29 |
arxcruz|rover | lol | 15:29 |
*** Goneri has joined #oooq | 15:29 | |
*** dpawlik has joined #oooq | 15:30 | |
panda | zbr: what can I do to make rdo-tox-molecule work with pytst-html ? | 15:30 |
zbr | panda: probably to put a <2.0 condition on it | 15:31 |
panda | zbr: trying | 15:31 |
rlandy | arxcruz|rover: this whole social distancing thing is stupid - let's all stay away from each other - except in the grocery store - where we can all tackle each other for the last box of clorox | 15:31 |
arxcruz|rover | rlandy: well, at least here, people were actually keep away from each other on the grocery store | 15:32 |
arxcruz|rover | in some very small groceries they are allowing only a few people to enter | 15:33 |
*** dsneddon_ has joined #oooq | 15:56 | |
panda | zbr: it is already pinned ... | 15:58 |
zbr | paste link to error | 15:58 |
panda | zbr: https://logserver.rdoproject.org/74/24774/73/check/rdo-tox-molecule/5556a82/job-output.txt | 15:58 |
panda | zbr: appaerently the pinning doesn't work 2020-03-16 09:41:42.370839 | cloud-centos-8 | plugins: metadata-1.8.0, html-2.1.0, cov-2.8.1, molecule-1.2.4 | 15:58 |
panda | zbr: molecule-delegated inherits from testenv:molecule | 15:59 |
panda | zbr: and testenv:molecule has the version constraing | 15:59 |
panda | constraint* | 15:59 |
panda | zbr: I'm stupid | 16:00 |
zbr | :D | 16:00 |
panda | zbr: no I'm not | 16:00 |
zbr | panda: is not you, is pip. | 16:00 |
rlandy | chandankumar: https://review.opendev.org/713277 pls review | 16:01 |
zbr | but "html-2.1.0" is the reason | 16:01 |
zbr | be sure this does not get used. | 16:01 |
zbr | they will do a patch soon, but until that, you we need to avoid it. | 16:01 |
marios | damn.. any one know how i can test that (config repo) https://review.rdoproject.org/r/25932 ... dont think i can test it with testproject depends on https://review.rdoproject.org/r/#/c/25796/ | 16:03 |
panda | zbr: I see, there are cross dependencies to correct all over the place | 16:03 |
*** saneax has quit IRC | 16:04 | |
*** jmasud has joined #oooq | 16:06 | |
marios | sshnaidm|afk: rlandy: when you next have a chance can you check https://review.rdoproject.org/r/25932 I think that is what causes "dnf: command not found" @ https://logserver.rdoproject.org/96/25796/4/check/periodic-tripleo-ci-centos-8-ovb-3ctlr_1comp_1supp-featureset039-master/0522040/job-output.txt | 16:10 |
panda | marios: zbr https://review.rdoproject.org/r/25933 let's see if this works | 16:13 |
*** skramaja has quit IRC | 16:14 | |
panda | marios: zbr definitely not :) | 16:14 |
*** derekh has joined #oooq | 16:17 | |
chandankumar | rlandy, sorry in pyconindia call deciding the fate of conference | 16:28 |
chandankumar | will look into that | 16:29 |
rlandy | no rush | 16:29 |
*** tesseract has quit IRC | 16:33 | |
*** bogdando has quit IRC | 16:35 | |
weshay|ruck | rlandy, 0/ out of meetings :) | 16:37 |
rlandy | weshay|ruck: how was it? | 16:38 |
weshay|ruck | meh | 16:39 |
rlandy | ok - you didn't miss much here ... | 16:39 |
weshay|ruck | marios, is the plan merge the promoter change and try to promote asap after merge? | 16:39 |
panda | zbr: I put version constraints everywhere, I have no idea what's installing the wrong version of pytest-html | 16:40 |
marios | weshay|ruck: panda: was going to run it 'properly' since over the weekend only containers were running. | 16:40 |
panda | zbr: https://logserver.rdoproject.org/33/25933/2/check/rdo-tox-molecule/db7eacb/job-output.txt | 16:40 |
marios | weshay|ruck: panda: and then maybe tomorrow we merge that patch | 16:40 |
panda | weshay|ruck: promoter had a full run this time, with containers overcloud and dlrn | 16:40 |
weshay|ruck | marios, ok.. so let a real one fly . .then merge | 16:40 |
panda | marios: ^ | 16:40 |
marios | weshay|ruck: to be clear we need to merge that https://review.rdoproject.org/r/25782 ontop of panda change it has another minor fix we need | 16:40 |
panda | marios: weshay|ruck but we have to fix rdo-tox-molecule to merge | 16:41 |
marios | panda: cool story bro | 16:41 |
marios | ack so 'it works' (tm) | 16:41 |
rlandy | weshay|ruck: I'd like to add fs001 back into criteria for the intregration pipeline | 16:41 |
marios | panda: nice one panda | 16:41 |
* marios gets ready to run away | 16:41 | |
panda | marios: I'm trying to fix the jobs here https://review.rdoproject.org/r/25933 | 16:41 |
weshay|ruck | panda, marios should we not merge first? | 16:41 |
weshay|ruck | just curious.. | 16:41 |
marios | weshay|ruck: we can't until molecule stuff fixed red jobs | 16:41 |
* marios checks review if it was fixd | 16:41 | |
panda | weshay|ruck: at this point we can merge without problem ... after the jobs are fixed | 16:42 |
panda | marios: no, it was not, but I'm trying | 16:42 |
marios | weshay|ruck: https://review.rdoproject.org/r/#/c/24774/ rdo-tox-moleculeFAILURE in 21m 09s rdo-tox-molecule-delegated-centos-8FAILURE in 17m 16s | 16:42 |
weshay|ruck | panda, ya.. so unless you strongly are against it.. I would prefer we merge, then go live | 16:42 |
marios | weshay|ruck: we can't merge until those jobs ^^^ | 16:42 |
panda | weshay|ruck: ah, yeah agreed. | 16:42 |
marios | weshay|ruck: but otherwise yes (I am already +2 there) | 16:42 |
weshay|ruck | is there a review to fix molecule? | 16:43 |
marios | weshay|ruck: 18:41 < panda> marios: I'm trying to fix the jobs here https://review.rdoproject.org/r/25933 | 16:43 |
panda | weshay|ruck: marios there's still something in the requirements taht is installing the wrong version of pytest-html, I'm trying to find it. | 16:44 |
weshay|ruck | zbr, does that tie into the centos-8 switch ur trying to get through ^ | 16:44 |
weshay|ruck | ? | 16:44 |
*** dsneddon_ has quit IRC | 16:48 | |
zbr | weshay|ruck: nope, is deps issue. | 16:49 |
zbr | i will look at it after unblocking the rdo base node change | 16:49 |
*** ykarel is now known as ykarel|away | 16:50 | |
panda | zbr: https://opendev.org/openstack/requirements/raw/branch/master/upper-constraints.txt | 16:52 |
panda | zbr: it's in the tox:molecule dependencies | 16:52 |
panda | zbr: and taht seems to to have been fixed | 16:52 |
panda | zbr: [testenv:molecule] | 16:53 |
panda | deps = | 16:53 |
panda | -c{env:UPPER_CONSTRAINTS_FILE:https://releases.openstack.org/constraints/upper/master} | 16:53 |
panda | zbr: and there pytest-html is pinned ==2.10 | 16:53 |
panda | 2.1.0 | 16:53 |
*** jmasud has quit IRC | 16:54 | |
*** jmasud has joined #oooq | 16:55 | |
chandankumar | rlandy, minute comment https://review.opendev.org/#/c/713277/2 rest is ok | 16:59 |
rlandy | looking | 16:59 |
zbr | i have no control over what reqs team is doing there, two options investigate bug and fix pytest-molecule if bug real, or drop use of constraints. | 16:59 |
rlandy | chandankumar: imagebuild repos look ok now in integration pipeline | 17:00 |
chandankumar | rlandy, sweet | 17:00 |
* chandankumar is thinking how to go back to my native home town | 17:00 | |
zbr | panda: read https://github.com/pytest-dev/pytest-html/issues/282 5days old already | 17:02 |
*** marios has quit IRC | 17:02 | |
rlandy | chandankumar: less corona virus there? | 17:03 |
panda | zbr: anything we can do for the upper constraints ? you had a patch there | 17:03 |
*** holser has quit IRC | 17:06 | |
panda | zbr: found it. | 17:11 |
chandankumar | rlandy, yes | 17:12 |
chandankumar | rlandy, in pune it is sky rocketting each day new 2-3 cases | 17:12 |
chandankumar | all public places closed except local shops | 17:12 |
chandankumar | zbr, https://github.blog/2020-03-16-npm-is-joining-github/ | 17:14 |
panda | zbr: weshay|ruck https://review.opendev.org/713293 should fix the jobs | 17:17 |
panda | zbr: weshay|ruck but I have no idea if it will be merged. | 17:18 |
weshay|ruck | k | 17:18 |
weshay|ruck | panda, probably not.. | 17:19 |
weshay|ruck | panda, zbr can we not create an override molecule job that runs on centos-8? | 17:19 |
panda | weshay|ruck: thn we have 2 options, wait for the next pytest-html release, or remove the upper constraints from our requirements | 17:19 |
chandankumar | zbr, https://review.opendev.org/#/c/713277/ good to go | 17:19 |
panda | zbr: ideas ? ^ | 17:19 |
zbr | haha, tbh, i seen it comming, not surprised npm joins github. | 17:20 |
zbr | i wonder if they are really aware of what pile of mess they bough | 17:21 |
weshay|ruck | panda, why is using centos-8 not an option for molecule? | 17:21 |
zbr | on the other hand, npm did few things right compared with pypi. | 17:21 |
weshay|ruck | panda, zbr https://meet.google.com/esb-ikfb-tfq?authuser=1 | 17:21 |
chandankumar | zbr, github is doing one thing very well how they can capture developer mindshare, providing one shop for all | 17:22 |
chandankumar | based on demands | 17:23 |
*** rascasoft has quit IRC | 17:29 | |
chandankumar | rlandy, weshay|ruck anything to keep an eye tomorrow? | 17:39 |
chandankumar | see ya people stay safe! | 17:39 |
*** chandankumar is now known as raukadah | 17:39 | |
panda | weshay|ruck: zbr based on what I heard on the meeting, I'm renaming all the jobs in ci-config taht are pinned to a distro, to include that distro | 17:43 |
panda | zbr: weshay|ruck we have 5 right now to be renamed | 17:43 |
weshay|ruck | k.. thank you | 17:43 |
weshay|ruck | imho.. it would make sense to also include what they run against | 17:44 |
weshay|ruck | but it's just my opinion | 17:44 |
panda | weshay|ruck: you mean something additional to distro ? | 17:45 |
weshay|ruck | rlandy, do you care if we don't have the job that pins consistent -> component-ci-testing in the results in the cockpit query? | 17:45 |
weshay|ruck | panda, ya.. so the molecule job that runs against the promoter is much different than what runs against collect logs | 17:45 |
weshay|ruck | and one has no way to see that w/o digging in | 17:45 |
*** dsneddon_ has joined #oooq | 17:46 | |
weshay|ruck | I guess the same job can run against multiple different things in our ci-config | 17:46 |
weshay|ruck | not something we need to solve atm | 17:46 |
weshay|ruck | but it's kind of fuzzy | 17:46 |
zbr | panda: put a comment/review on https://github.com/pytest-dev/pytest-html/pull/283 | 17:49 |
panda | zbr: saying what ? How do I check if it solved the problem ? | 17:52 |
zbr | trust me, tested it, fixes the problem. | 17:52 |
zbr | i installed the patch and pytest-molecule is back happy. | 17:52 |
rlandy | raukadah: should be fine tomorrow | 17:53 |
zbr | you can see anything like: "having a hotfix with this would really be handy as we were forced to pin-down pytest-html due to it. | 17:53 |
rlandy | weshay|ruck: that's fine | 17:53 |
rlandy | ie: no consistent -> component-ci-testing in the results in the cockpit | 17:53 |
zbr | do you want me to write a reasoning-generator? it could prove handy :D | 17:54 |
rlandy | oh wow - something hit the pipeline | 17:54 |
weshay|ruck | rlandy, aye.. that should make it more visible when people use testproject to get something through | 17:54 |
weshay|ruck | panda, +2 https://review.rdoproject.org/r/#/c/25933/ | 17:55 |
weshay|ruck | ci came back | 17:55 |
rlandy | 2020-03-16 17:27:51 | msg: 'Container(s) with bad ExitCode: [''container-puppet-neutron''], check logs in /var/log/containers/stdouts/' | 17:56 |
rlandy | weshay|ruck: how did that hit the integration pipeline w/o impacting component ^^? | 17:56 |
panda | \o/ | 17:57 |
panda | merging | 17:57 |
rlandy | https://logserver.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-standalone-full-tempest-master/e67d899/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz | 17:57 |
*** dtantsur is now known as dtantsur|afk | 17:57 | |
*** jpena is now known as jpena|off | 17:58 | |
weshay|ruck | rlandy, that error? | 17:58 |
rlandy | weshay|ruck: ack | 17:58 |
rlandy | shouldn't we have caught that | 17:59 |
rlandy | in the component pipeline | 17:59 |
weshay|ruck | rlandy, some times containers fail to start | 17:59 |
weshay|ruck | not sure if that's this issue.. but I've seen that upstream quite a bit | 17:59 |
* weshay|ruck looks at this issue | 17:59 | |
rlandy | it's taking the whole pipeline down | 18:00 |
*** derekh has quit IRC | 18:00 | |
* rlandy wonders if we shouldn't build containers in component pipeline | 18:00 | |
weshay|ruck | rlandy, ? | 18:01 |
weshay|ruck | oh I see | 18:01 |
weshay|ruck | rlandy, loooks more like a container build issue | 18:05 |
rlandy | weshay|ruck: have you seen this happen before? | 18:05 |
rlandy | we rebuild the container? or kolla issue? | 18:06 |
weshay|ruck | 2020-03-16T17:27:38.870604079+00:00 stderr F <13>Mar 16 17:27:38 puppet-user: No such file or directory @ rb_sysopen - /etc/neutron/plugins/networking-ovn/networking-ovn-metadata-agent.ini | 18:06 |
weshay|ruck | which afaik is on the container | 18:07 |
rlandy | weshay|ruck: hmmm ... maybe building containers in the pipeline is required? container update didn't hit this | 18:08 |
weshay|ruck | rlandy, sorry.. it's here undercloud/var/lib/config-data/puppet-generated/neutron/etc/neutron/plugins/networking-ovn | 18:08 |
amoralej | weshay|ruck, i think https://review.opendev.org/#/c/712762/ | 18:09 |
weshay|ruck | rlandy, https://logserver.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-standalone-full-tempest-master/e67d899/logs/undercloud/var/lib/config-data/puppet-generated/ | 18:09 |
* weshay|ruck looks | 18:09 | |
amoralej | now we are forced to unpin neutron | 18:09 |
rlandy | 2020-03-16T17:33:39.356319204+00:00 stderr F <13>Mar 16 17:33:39 puppet-user: Error: Could not set 'present' on ensure: No such file or directory @ rb_sysopen - /etc/neutron/plugins/networking-ovn/networking-ovn-metadata-agent.ini (file: /etc/puppet/modules/neutron/manifests/agents/ovn_metadata.pp, line: 150) | 18:09 |
weshay|ruck | amoralej, unpin it all :) | 18:09 |
weshay|ruck | we're good | 18:09 |
rlandy | amoralej: neutron was not upinned? | 18:09 |
weshay|ruck | thanks for the pointer | 18:09 |
amoralej | we need some fixes | 18:10 |
rlandy | what did I miss? | 18:10 |
amoralej | nop, neutron was pinned to make transition to ovn-in-repo smooth :) | 18:10 |
amoralej | it needs several fixes here and there | 18:10 |
rlandy | I figure container update should have caught this | 18:10 |
rlandy | ie: in the component pipeline | 18:10 |
weshay|ruck | rlandy, it's a kolla change | 18:11 |
amoralej | weshay|ruck, rlandy https://review.rdoproject.org/r/#/c/24462/ | 18:11 |
amoralej | that still has an issue, as it will break octavia with ovn | 18:12 |
amoralej | although at this point this is probably a smaller issue | 18:12 |
rlandy | that is less breakage than what we have now | 18:13 |
amoralej | rlandy, that's only affecting periodic, right? | 18:13 |
weshay|ruck | amoralej, ya | 18:13 |
amoralej | i'm fine with unpinning if that passes ci | 18:14 |
rlandy | yeah - nothing hits tripleo-current yet | 18:14 |
weshay|ruck | amoralej, the rdo-info ci right? | 18:14 |
amoralej | and ask neutron team to create new package for https://github.com/openstack/ovn-octavia-provider | 18:14 |
amoralej | weshay|ruck, first, ci for https://review.rdoproject.org/r/#/c/24462/ itself | 18:15 |
*** sshnaidm|afk is now known as sshnaidm | 18:15 | |
amoralej | i rechecked some time ago | 18:15 |
rlandy | amoralej: is anything else still pinned? | 18:15 |
weshay|ruck | panda, we'll be looking to promote a master c8 job run from ealier today | 18:15 |
amoralej | mistral is also being unpinned today | 18:16 |
panda | weshay|ruck: will promote to panda-man | 18:16 |
panda | weshay|ruck: what's the aggregate-hash ? I see some hashes here taht failed to meet the criteria | 18:17 |
weshay|ruck | panda, /me gets | 18:17 |
*** dsneddon_ is now known as dsneddon | 18:17 | |
panda | weshay|ruck: and criteria is still reduced | 18:17 |
amoralej | rlandy, https://review.rdoproject.org/r/#/c/25919/ | 18:18 |
rlandy | amoralej: thanks LP bug coming up | 18:18 |
weshay|ruck | panda, the last run .. had https://logserver.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-scenario010-standalone-master/4350e26/logs/undercloud/etc/yum.repos.d/delorean.repo.txt.gz | 18:19 |
weshay|ruck | 9e360f046d92bc8cc27921a1e491248c delorean.repo | 18:20 |
weshay|ruck | panda, I'll patch the criteria w/ notes | 18:20 |
panda | weshay|ruck: ok it's not on the list that doesn't meet the criteria | 18:21 |
weshay|ruck | panda, ok.. no changes needed.. everything in criteria for centos-8 master passed in that run | 18:22 |
weshay|ruck | panda, I would expect to be blocked on the ovn issue for at least several days | 18:24 |
amoralej | rlandy, weshay|ruck jobs are failing as that change modified config file location | 18:26 |
weshay|ruck | aye | 18:26 |
weshay|ruck | panda, so we should see centos-8 kick soon and pick up that hash w/ panda_man? | 18:26 |
rlandy | https://review.rdoproject.org/r/#/c/25919/1/tags/ussuri-uc.yml - neutron is still pinned here | 18:26 |
weshay|ruck | panda-man | 18:27 |
rlandy | https://bugs.launchpad.net/tripleo/+bug/1867664 | 18:28 |
openstack | Launchpad bug 1867664 in tripleo "Master periodic jobs are failing overcloud deploy with ''Container(s) with bad ExitCode: [''container-puppet-neutron''], check logs in /var/log/containers/stdouts/'" [Critical,Triaged] | 18:28 |
rlandy | amoralej: weshay|ruck: ^^ | 18:29 |
rlandy | promotion-blocker so it should hit CIX | 18:29 |
panda | weshay|ruck: yes | 18:29 |
amoralej | one option would be to restore the packages to the containers via tripleo-common overrides | 18:33 |
amoralej | is it worthy? | 18:33 |
amoralej | weshay|ruck, rlandy ^ | 18:33 |
weshay|ruck | amoralej, how long do you think it will take to fix neutron.. moving ovn is not small.. | 18:34 |
amoralej | neutron unpin may introduce other issues | 18:34 |
weshay|ruck | amoralej, to buy time.. it's probably worth it | 18:35 |
amoralej | packaging wise, we can fix it and unpin in some hours | 18:35 |
amoralej | but | 18:35 |
amoralej | not sure if config options, etc... are the same | 18:35 |
rlandy | idk - hacking more may hurt us | 18:36 |
amoralej | tomorrow we can probably have support from someone from ovn | 18:36 |
weshay|ruck | panda, it just ran | 18:36 |
rlandy | maybe give neutron team a day? | 18:36 |
weshay|ruck | non candidate found? | 18:36 |
rlandy | amoralej: ack - let's give ovn a chance to respond | 18:36 |
rlandy | if they say it's complicated/long, we can consider overrides | 18:37 |
rlandy | and deal with the consequences | 18:37 |
amoralej | rlandy, weshay|ruck is it fine for you to keep this failing until tomorrow? | 18:38 |
weshay|ruck | ok w/ me | 18:38 |
panda | weshay|ruck: https://trunk.rdoproject.org/centos8-master/tripleo-ci-testing/delorean.repo.md5 I don't see 9e360f046d92bc8cc27921a1e491248c in tripleo-ci-testing | 18:38 |
weshay|ruck | panda, /me checks container tag | 18:39 |
weshay|ruck | for that run | 18:39 |
weshay|ruck | perhaps unzipping / renaming | 18:39 |
rlandy | ack - no emergency | 18:39 |
weshay|ruck | panda, container build says the md5 is 46b71a6620e3372c998db7a694112fd2 | 18:40 |
weshay|ruck | panda, whis *is* in the logs here | 18:40 |
weshay|ruck | 2020-03-16 10:27:02.276029 | primary | { | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276261 | primary | "aggregate_hash": "46b71a6620e3372c998db7a694112fd2", | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276281 | primary | "commit_hash": null, | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276316 | primary | "component": null, | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276330 | primary | "distro_hash": null, | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276337 | primary | "in_progress": false, | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276355 | primary | "job_id": "periodic-tripleo-ci-centos-8-standalone-master", | 18:41 |
weshay|ruck | 2020-03-16 10:27:02.276376 | primary | "notes": "", | 18:42 |
weshay|ruck | 2020-03-16 10:27:02.276398 | primary | "success": true, | 18:42 |
weshay|ruck | 2020-03-16 10:27:02.276411 | primary | "timestamp": 1584354405, | 18:42 |
weshay|ruck | 2020-03-16 10:27:02.276471 | primary | "url": "https://logserver.rdoproject.org/openstack-periodic-master/opendev.org/openstack/tripleo-ci/master/periodic-tripleo-ci-centos-8-standalone-master/b057a46" | 18:42 |
weshay|ruck | panda, you may have already promoted that one | 18:42 |
panda | weshay|ruck: yep, that's ancient | 18:43 |
weshay|ruck | hrm | 18:43 |
panda | weshay|ruck: most recent was promoted 2:30 hours ago | 18:43 |
panda | weshay|ruck: but with reduced crteria | 18:43 |
weshay|ruck | panda, can't be ancied | 18:44 |
weshay|ruck | ancient | 18:44 |
rlandy | ok - so back to original question - how come container update never hit this issue? | 18:44 |
weshay|ruck | hopefully | 18:44 |
weshay|ruck | rlandy, container builds | 18:44 |
panda | weshay|ruck: 46b71a6620e3372c998db7a694112fd2 is from 10 hours ago | 18:44 |
weshay|ruck | it was a change in kolla | 18:44 |
weshay|ruck | panda, ok.. that's not ancient | 18:44 |
weshay|ruck | :) | 18:44 |
panda | weshay|ruck: AGES | 18:44 |
panda | weshay|ruck: EONS | 18:44 |
weshay|ruck | depends how closely you watch I guess | 18:44 |
weshay|ruck | panda, so that one looks safe to promote all the way through | 18:44 |
weshay|ruck | but we can hold that until tomorrow | 18:45 |
weshay|ruck | you check how many old attempts? | 18:45 |
weshay|ruck | 10? | 18:45 |
weshay|ruck | rlandy, ideally kolla doesn't change that much :) | 18:45 |
panda | weshay|ruck: 10. not wure what you mean with "that" can be promoted all the way through | 18:45 |
weshay|ruck | let's chat | 18:45 |
panda | weshay|ruck: we already had e03f7a0753e60c3263a20e4e6abf8512 promoted all the way through | 18:46 |
weshay|ruck | https://meet.google.com/poc-dags-wfh?authuser=1 rlandy too | 18:46 |
weshay|ruck | that's more recent? | 18:46 |
panda | yes | 18:46 |
weshay|ruck | I just checked the last build | 18:46 |
weshay|ruck | huh /me looks again | 18:46 |
panda | but with reduced criteria | 18:46 |
weshay|ruck | ya.. | 18:46 |
weshay|ruck | panda, I'd have to see what's passed all the jobs | 18:47 |
rlandy | k | 18:47 |
weshay|ruck | periodic-tripleo-centos-8-master-containers-build-pushopenstack/tripleo-cimasteropenstack-periodic-mastermaster34232020-03-16T08:15:23SUCCESS | 18:50 |
weshay|ruck | rlandy, panda ^ | 18:50 |
panda | rlandy: https://trunk.rdoproject.org/api-centos8-master-uc/api/civotes_agg_detail.html?ref_hash=e03f7a0753e60c3263a20e4e6abf8512 | 18:52 |
panda | https://trunk.rdoproject.org/centos8-master/panda-man/delorean.repo.md5 | 18:56 |
panda | weshay|ruck: rlandy ^ | 18:56 |
rlandy | weshay|ruck: https://review.rdoproject.org/r/#/c/25919/1/tags/ussuri-uc.yml | 19:09 |
rlandy | weshay|ruck: https://review.opendev.org/#/c/713220/ | 19:14 |
*** jbadiapa has quit IRC | 19:18 | |
panda | weshay|ruck: https://review.rdoproject.org/r/25782 | 19:24 |
amoralej | rlandy, weshay|ruck i've sent a new ps for the networking-ovn-metadata-againt.ini, i'm testing it in https://review.rdoproject.org/r/#/c/24462 and https://review.rdoproject.org/r/25937 | 19:29 |
rlandy | amoralej: thanks | 19:29 |
amoralej | i suspect it may miss something, i'll work with neutron guys tomorrow morning | 19:29 |
weshay|ruck | thanks | 19:30 |
*** dpawlik has quit IRC | 19:32 | |
*** dpawlik has joined #oooq | 19:32 | |
*** amoralej is now known as amoralej|off | 19:40 | |
weshay|ruck | hrm.. rlandy still doesn't pickup what ever was run as test project http://dashboard-ci.tripleo.org/d/UDA4H3aZk/component-pipeline?orgId=1&from=now-7d&to=now&fullscreen&panelId=422 | 19:40 |
*** dpawlik has quit IRC | 19:41 | |
weshay|ruck | no passes on fs001 on 3/16 but we promoted | 19:41 |
weshay|ruck | :( | 19:41 |
rlandy | wierd | 19:48 |
rlandy | looking | 19:48 |
rlandy | checking zuul ... | 19:49 |
rlandy | weshay|ruck ... | 19:49 |
rlandy | <rlandy> how did baremetal promote when it failed OVB? | 19:49 |
rlandy | <chandankumar> rlandy, rerun the job | 19:49 |
rlandy | <rlandy> chandankumar: ah | 19:49 |
rlandy | <chandankumar> *rerunned | 19:49 |
*** ccamacho has quit IRC | 19:50 | |
rlandy | https://review.rdoproject.org/zuul/builds?job_name=periodic-tripleo-ci-centos-8-ovb-3ctlr_1comp-featureset001-baremetal-master | 19:50 |
rlandy | ^^ there is the pass | 19:50 |
weshay|ruck | rlandy, hrm.. /me looks | 19:51 |
rlandy | 2020-03-16T07:07:48 | 19:52 |
weshay|ruck | ah maybe because it's check | 19:52 |
* weshay|ruck queries again | 19:52 | |
rlandy | weshay|ruck: not such a big deal with cockpit doesn't pick it up | 19:52 |
rlandy | the general cockpit tracking is the same | 19:52 |
rlandy | you have to search the hash | 19:52 |
rlandy | not a problem | 19:53 |
weshay|ruck | well.. it's just another not very transparent thing | 19:53 |
weshay|ruck | we have soo many of those things | 19:53 |
rlandy | weshay|ruck: ack - if we can but we are not below the current standard | 19:53 |
weshay|ruck | lolz | 19:54 |
weshay|ruck | hrm.. https://trunk.rdoproject.org/api-centos8-master-uc/api/civotes_detail.html?commit_hash=123c6fc147cfdd8607a529799bfa8e800395bd02&distro_hash=40a480aaa83fac115ad82b8917f8280f88ad5cbe | 20:12 |
weshay|ruck | meh | 20:14 |
panda | 46b7 promoted. | 20:20 |
panda | 2020-03-16 20:17:03,054 14822 INFO promoter Candidate hash 'aggregate: 46b71a6620e3372c998db7a694112fd2, commit: ee36da77f5d639ed985421d1a615dc497bb0ec7d, distro: b61bb6e4c4777dc0c28ab7df9b5e99fcfeb2b8fa, component: ui, timestamp: 1584346454': SUCCESSFUL promotion to current-tripleo | 20:21 |
panda | 2020-03-16 20:17:03,056 14822 INFO promoter Summary: Promoted 1 hashes this round | 20:21 |
panda | 2020-03-16 20:17:03,056 14822 INFO promoter ------- -------- Promoter terminated normally | 20:21 |
*** panda is now known as panda|off | 20:25 | |
rlandy | nice | 20:32 |
weshay|ruck | promotion worked.. and I shut it down rlandy panda|off | 20:34 |
weshay|ruck | rlandy, do I cry now or later https://trunk.rdoproject.org/centos8-master/current-tripleo/delorean.repo.md5 | 20:35 |
weshay|ruck | panda|off, def.. have a bug | 20:37 |
rlandy | wrong hash, right? | 20:39 |
weshay|ruck | yup | 20:39 |
weshay|ruck | all jobs will start to fail soon | 20:40 |
rlandy | weshay|ruck: retag to panda_X | 20:40 |
rlandy | and repromote, we can do that right? | 20:40 |
rlandy | oh dear ... | 20:40 |
weshay|ruck | rlandy, if you want to chat... we can start a meet | 20:41 |
weshay|ruck | to fix this myself.. I have to pull all the containers | 20:41 |
rlandy | weshay|ruck: k- let's chat - and see what's reasonable to do | 20:41 |
rlandy | cant we just tag an old hash as current-tripleo? | 20:42 |
rlandy | weshay|ruck: https://meet.google.com/ggm-qzyn-jun | 20:43 |
weshay|ruck | panda|off, retag https://trunk.rdoproject.org/centos8-master/current-tripleo/delorean.repo.md5 e03f7a0753e60c3263a20e4e6abf8512 | 20:47 |
rlandy | 46b71a6620e3372c998db7a694112fd2 | 20:48 |
rlandy | not | 20:48 |
rlandy | e03f7a0753e60c3263a20e4e6abf8512 | 20:48 |
weshay|ruck | retag the containers tagged w/ 46b71a6620e3372c998db7a694112fd2 | 20:49 |
weshay|ruck | e03f7a0753e60c3263a20e4e6abf8512 | 20:49 |
*** jmasud has quit IRC | 20:53 | |
*** jmasud has joined #oooq | 20:56 | |
*** sshnaidm has quit IRC | 21:37 | |
*** jfrancoa has quit IRC | 21:40 | |
*** sshnaidm has joined #oooq | 21:44 | |
rlandy | weshay|ruck: rerun ui component test - think it hit the current-tripleo promotion | 22:04 |
*** TrevorV has quit IRC | 22:04 | |
rlandy | reran | 22:04 |
weshay|ruck | rlandy, oh.. meaning the ui component is using the most recent current-tripleo + ui component? | 22:05 |
rlandy | weshay|ruck: ack | 22:05 |
*** ChanServ sets mode: +o panda|off | 22:46 | |
*** panda|off changes topic to "Docs: https://docs.openstack.org/tripleo-quickstart/latest/ || 17th march promoter critical bugfix https://review.rdoproject.org/r/25938" | 22:47 | |
*** panda|off sets mode: -o panda|off | 22:47 | |
*** holser has joined #oooq | 23:45 | |
*** jmasud has quit IRC | 23:53 | |
*** jmasud has joined #oooq | 23:53 | |
*** jmasud has quit IRC | 23:58 | |
*** holser has quit IRC | 23:58 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!