| *** Goneri has joined #oooq | 00:05 | |
| *** jmasud has quit IRC | 00:11 | |
| *** jmasud has joined #oooq | 00:14 | |
| *** ysandeep|away is now known as ysandeep | 00:17 | |
| *** dsneddon has joined #oooq | 01:35 | |
| *** dsneddon has quit IRC | 01:41 | |
| *** Goneri has quit IRC | 01:47 | |
| *** rfolco|bbl has quit IRC | 02:40 | |
| *** rlandy has quit IRC | 03:03 | |
| *** apetrich has quit IRC | 03:12 | |
| *** ykarel|away is now known as ykarel | 03:30 | |
| ykarel | weshay|ruck, wrt <weshay|ruck> the logs are broken again, wrong setting are applied there as we rely on provider, https://review.opendev.org/#/c/705379/ should fix it | 03:33 |
|---|---|---|
| ykarel | can bring that setting in seperate patch though if combined one takes time | 03:35 |
| weshay|ruck | ykarel, isn't this crazy :) | 03:38 |
| ykarel | too much :) | 03:38 |
| weshay|ruck | hrm.. https://review.opendev.org/#/c/705378/ | 03:41 |
| weshay|ruck | I'd make a joke.. but | 03:42 |
| ykarel | hmm ^^ is blocked as zbr wanted to do dynamic generation of logs | 03:42 |
| ykarel | readme | 03:42 |
| weshay|ruck | ya.. while we bleed | 03:42 |
| weshay|ruck | ykarel, who hooked up the footer in 3rd party? it's awesome | 03:46 |
| ykarel | me no idea | 03:55 |
| ykarel | weshay|ruck, ^ | 03:56 |
| weshay|ruck | ykarel, well.. high five that person if we find out :) | 03:56 |
| weshay|ruck | ykarel, and thanks to you as well for the logging patches :) | 03:56 |
| weshay|ruck | ykarel, didn't think I had to coordinate "fix the logging" but apparently we do | 03:57 |
| weshay|ruck | I'll round folks up this week | 03:57 |
| ykarel | weshay|ruck, ack | 03:57 |
| *** jmasud has quit IRC | 04:13 | |
| *** jmasud has joined #oooq | 04:13 | |
| *** skramaja has joined #oooq | 04:15 | |
| *** raukadah is now known as chkumar|rover | 04:22 | |
| ykarel | weshay|ruck, so i understand u only fixed the ReadME to footer, thats nice Thanks | 04:34 |
| ykarel | with https://review.opendev.org/#/c/702286/2/tasks/publish.yml | 04:35 |
| *** udesale has joined #oooq | 04:53 | |
| *** soniya29 has joined #oooq | 05:29 | |
| *** ratailor has joined #oooq | 05:40 | |
| *** marios has joined #oooq | 06:02 | |
| *** jbadiapa has joined #oooq | 06:51 | |
| *** jfrancoa has joined #oooq | 06:52 | |
| * chkumar|rover headed to wework | 07:18 | |
| *** dtantsur has joined #oooq | 07:25 | |
| *** jaosorior has joined #oooq | 07:38 | |
| *** ykarel is now known as ykarel|lunch | 07:42 | |
| *** ysandeep is now known as ysandeep|lunch | 07:46 | |
| marios | panda: o/ hey see email when you get a chance re "ppc issues again" ... afaics it isn't fatal but maybe I missed something. I am hoping its just the (expected) error pulling ppc containers and then it continue, but weshay|ruck saw the error and assumed it was fatal. not sure but i can't see something more in logs | 07:46 |
| *** jaosorior has quit IRC | 07:53 | |
| *** amoralej|off is now known as amoralej | 07:58 | |
| *** jaosorior has joined #oooq | 08:06 | |
| *** holser has joined #oooq | 08:08 | |
| *** jmasud has quit IRC | 08:12 | |
| *** jtomasek has quit IRC | 08:12 | |
| *** jmasud has joined #oooq | 08:14 | |
| *** tesseract has joined #oooq | 08:16 | |
| zbr | ykarel|lunch: re readme, i am afraid that the progress is blocked waiting for your review on https://review.opendev.org/#/c/705454/ | 08:17 |
| chkumar|rover | ykarel|lunch, http://zuul.openstack.org/builds?job_name=tripleo-ci-centos-7-standalone&branch=stable%2Ftrain&pipeline=gate&branch=stable%2Ftrain it appears to be healthy | 08:20 |
| *** jtomasek has joined #oooq | 08:34 | |
| *** panda has quit IRC | 08:34 | |
| *** panda has joined #oooq | 08:35 | |
| *** bogdando has joined #oooq | 08:41 | |
| *** apetrich has joined #oooq | 08:48 | |
| zbr | marios: i am back on WIP on collect-cmds change, as sagi pointed critical bug (redirection not workig), i already had alternative uploaded but i need to see it producing desired results. | 08:49 |
| zbr | local testing passed but is not enough ;) | 08:49 |
| *** ykarel|lunch is now known as ykarel | 08:52 | |
| ykarel | chkumar|rover, yes that issue is random and hit only some time | 08:52 |
| *** jpena|off is now known as jpena | 08:53 | |
| ykarel | zbr, sure i will check, but i would say get feedback from others as well as it's changing the management of that docs | 08:54 |
| ykarel | for me main thing is it should be easily managable/editable | 08:54 |
| marios | zbr: ack | 08:54 |
| *** rascasoft has joined #oooq | 08:54 | |
| ykarel | and should render fine in logs footer | 08:54 |
| zbr | ykarel: first look at the outcome, you can see both, new one is lowercase. | 08:54 |
| ykarel | zbr, yup i saw that | 08:55 |
| zbr | ykarel: i need initial feedback before investing more time in it. | 08:55 |
| zbr | maybe even better if you want to takeover the challenge, we could make it a reality faster. | 08:56 |
| zbr | the idea is that a static file is a dead end, because it never knows what you really collected, or if the urls are correct. | 08:56 |
| zbr | dynamic is the key, run after you already collected all files, the POC is using a simple ansible module to produce that, one using jinja2 inside, so it should easy to understand. | 08:57 |
| zbr | not much magic involved. | 08:57 |
| zbr | my initial attempt was to do it in ansible but soon i realised it would not work, ansible is too slow to do this. 10 lines of python did the trick, | 08:58 |
| ykarel | zbr, as first look, LGTM, would need additionally:- log list categorized, template outside of python file | 09:00 |
| zbr | ykarel: i like that we think alike. | 09:00 |
| ykarel | zbr, so posted my +1 on the review with remarks, will remove the static changes from my review | 09:17 |
| ykarel | just keep vexxhost related changes | 09:18 |
| *** rascasoft has quit IRC | 09:20 | |
| *** rascasoft has joined #oooq | 09:20 | |
| *** derekh has joined #oooq | 09:40 | |
| chkumar|rover | sshnaidm|afk, zbr https://review.rdoproject.org/r/#/c/24803/ | 09:45 |
| chkumar|rover | please have a look when free | 09:45 |
| zbr | chkumar|rover: e may not be the only users of this nodeset, you should not combine nodeswitching with its removal. | 09:47 |
| chkumar|rover | zbr, fixed | 09:47 |
| zbr | put removal separated and you have my full support. | 09:47 |
| chkumar|rover | zbr, yes sure | 09:48 |
| chkumar|rover | arxcruz, as we discussed yesterday, do we want to bump concurrency of standalone full tempest? | 09:49 |
| arxcruz | chkumar|rover: let's test | 09:49 |
| zbr | chkumar|rover: lets switch to rdo about that subject | 09:51 |
| chkumar|rover | zbr, aye | 09:51 |
| panda | marios: I think you are right, but even last time I could not find the fatal log. And centos7_master logs for that time are missing. | 09:55 |
| marios | panda: k lets wait for weshay|ruck maybe there is something else we're missing | 09:58 |
| marios | panda: thanks for sanity check | 09:58 |
| *** rfolco|bbl has joined #oooq | 10:08 | |
| *** jmasud has quit IRC | 10:11 | |
| *** jmasud has joined #oooq | 10:12 | |
| *** panda is now known as panda|bbl | 10:33 | |
| *** jmasud has quit IRC | 10:36 | |
| *** ysandeep|lunch is now known as ysandeep | 10:38 | |
| sshnaidm|afk | zbr, I didn't understand your comment here: https://review.opendev.org/#/c/705623/ | 10:39 |
| marios | arxcruz: chkumar|rover: fs20 scenario only 3h35m @ https://review.rdoproject.org/r/#/c/24339/ | 10:41 |
| marios | arxcruz: are you still exploring there? can you please comment on https://review.opendev.org/#/c/701016/ ... this has been going on for a while. i'd rather not stall it... if we are going with an alternative approach then lets do that you can take over https://tree.taiga.io/project/tripleo-ci-board/task/1383 and i will abandon the patches there | 10:42 |
| marios | arxcruz: starting to get sick of it ;) | 10:42 |
| chkumar|rover | zbr, how you solved re2-devel issue for centos8? | 10:45 |
| chkumar|rover | zbr, https://logserver.rdoproject.org/03/24803/3/check/tox-linters/13414c7/job-output.txt | 10:45 |
| chkumar|rover | zbr, _re2.cc:37:10: fatal error: re2/re2.h: No such file or directory | 10:45 |
| *** jmasud has joined #oooq | 10:46 | |
| arxcruz | marios: chkumar|rover this is great, we can have both concurrency 4 and split jobs | 10:49 |
| arxcruz | this will decrease the time a lot | 10:50 |
| arxcruz | marios: yesterday we merge concurrency = 4 on fs020 | 10:50 |
| arxcruz | we should not see timeouts now | 10:50 |
| chkumar|rover | FYI!, if you are sending patches on rdo-jobs, tox jobs is broken, working on fixing it | 10:52 |
| marios | arxcruz: ah maybe this is why the time went to 3.35 then | 10:54 |
| arxcruz | marios: not really, because i see your patch have concurrency 3 | 10:54 |
| *** jmasud has quit IRC | 10:55 | |
| arxcruz | marios: but periodic fs020 today pass ;) | 10:56 |
| arxcruz | and with less time than before, but we are still in the edge, any change we might get timeouts again | 10:56 |
| *** udesale_ has joined #oooq | 11:13 | |
| *** udesale has quit IRC | 11:16 | |
| zbr | sshnaidm|afk: i replied. thanks for pointing to the redirection bug, i already fixed it in patchset no2, rdo already reported results. | 11:17 |
| *** derekh has quit IRC | 12:07 | |
| *** derekh has joined #oooq | 12:08 | |
| *** tosky has joined #oooq | 12:12 | |
| * chkumar|rover headed home, will meet during program call | 12:29 | |
| *** jpena is now known as jpena|lunch | 12:33 | |
| *** soniya29 has quit IRC | 12:38 | |
| *** rfolco|bbl is now known as rfolco | 12:45 | |
| rfolco | weshay|ruck, you know where this is called ? 2020-02-05 11:57:28.632750 | primary | + tripleo-repos --mirror http://mirror.centos.org --rdo-mirror https://trunk.rdoproject.org highavailability opstools | 12:46 |
| rfolco | weshay|ruck, its calling w/ highavailability and opstools args | 12:46 |
| marios | rfolco: yeah here | 12:46 |
| marios | rfolco: i was literally just commenting on it https://review.opendev.org/#/c/701937/12/roles/tripleo-repos/tasks/main.yml | 12:46 |
| weshay|ruck | rfolco, marios I have it working :) | 12:47 |
| marios | rfolco: i am updating my patch on weshay|ruck latest stuff and commented there https://review.opendev.org/#/c/702993/11/roles/tripleo-repos/tasks/main.yml | 12:47 |
| weshay|ruck | new match up on the tripleo-repo role | 12:47 |
| rfolco | marios, weshay|ruck ha and powertools should come for FREE | 12:47 |
| weshay|ruck | rfolco, they do | 12:47 |
| weshay|ruck | :) | 12:47 |
| weshay|ruck | if ur on the right dlrn | 12:47 |
| rfolco | we should not specify as args then | 12:47 |
| marios | rfolco: ack that's what i was asking in my comment at https://review.opendev.org/#/c/702993/11/roles/tripleo-repos/tasks/main.yml | 12:47 |
| weshay|ruck | we dont | 12:48 |
| marios | rfolco: i'll remove highavailability | 12:48 |
| rfolco | weshay|ruck, look this https://review.opendev.org/#/c/701937/12/roles/tripleo-repos/tasks/main.yml | 12:48 |
| rfolco | it raises an error here | 12:48 |
| rfolco | https://logserver.rdoproject.org/75/24775/11/check/tripleo-build-containers-centos-8-master/34a8afe/job-output.txt | 12:48 |
| rfolco | marios, cool :) | 12:48 |
| marios | rfolco: see my replies ^^ | 12:48 |
| weshay|ruck | rfolco, please follow my latest changes | 12:48 |
| marios | weshay|ruck: i don't understand why we have 2 patches | 12:49 |
| rfolco | weshay|ruck, ok we need a semaphore to update patches at the same time :) | 12:49 |
| marios | weshay|ruck: should i abandon my one? | 12:49 |
| marios | weshay|ruck: i don't see the point having to sync patches | 12:50 |
| weshay|ruck | marios, not sure.. up to you guys.. I split off a branch to poke at it as well | 12:50 |
| weshay|ruck | marios, you can keep doing what you were doing or jump into my patch | 12:50 |
| weshay|ruck | don't care | 12:50 |
| marios | weshay|ruck: well ^^ 'up to you guys' not really, i mean if i vote to keep my patch then you will just keep updating your one too :D | 12:50 |
| marios | so? | 12:50 |
| rfolco | yeah its one or the other | 12:50 |
| weshay|ruck | want to chat in 5min? | 12:51 |
| marios | weshay|ruck: sure | 12:51 |
| weshay|ruck | k k | 12:51 |
| marios | weshay|ruck: did you see the email re promoter and ppc containers | 12:51 |
| marios | weshay|ruck: i am not sure if there was a problem or i missed something more in logs.. i asked panda|bbl for sanity check too | 12:51 |
| marios | weshay|ruck: as per the email, i think it was working OK (it has error for ppc pull, but we expected those, and then it continues with ignore_errors) | 12:52 |
| weshay|ruck | ok ready marios rfolco | 12:55 |
| rfolco | where | 12:55 |
| weshay|ruck | https://meet.google.com/vui-awyj-qbh?authuser=1 | 12:56 |
| *** rlandy has joined #oooq | 13:01 | |
| chkumar|rover | weshay|ruck, program call? | 13:04 |
| *** nhicher has quit IRC | 13:05 | |
| *** ysandeep is now known as ysandeep|away | 13:07 | |
| chkumar|rover | sshnaidm|afk, zbr rlandy https://review.rdoproject.org/r/#/c/24803/ is good to go | 13:11 |
| chkumar|rover | rdo jobs tox jobs broken currently | 13:11 |
| marios | 2020-02-05 11:11:29.884912 | ubuntu-bionic | ./tripleo_repos/tests/test_main.py:110:43: E502 the backslash is redundant between brackets | 13:11 |
| *** ratailor has quit IRC | 13:12 | |
| zbr | chkumar|rover: you removed an useful comment, # needed to avoid accidental call of /usr/bin/tox | 13:12 |
| chkumar|rover | zbr, you want I can readd back | 13:13 |
| chkumar|rover | zbr, done | 13:14 |
| zbr | added a -2 until we get confirmation about EPEL part. | 13:16 |
| zbr | touching base/pre.yaml is highly sensitive | 13:16 |
| zbr | if we get the ok, wf it. but i just want confirmation from wes and rdo. | 13:17 |
| chkumar|rover | zbr, epel is disabled upstream | 13:17 |
| chkumar|rover | zbr, for re2-devel it comes from epel | 13:18 |
| chkumar|rover | zbr, i can parameterize it based on tox job | 13:19 |
| *** ykarel is now known as ykarel|away | 13:27 | |
| weshay|ruck | chkumar|rover, you need to sync? | 13:31 |
| *** amoralej is now known as amoralej|lunch | 13:31 | |
| chkumar|rover | weshay|ruck, yes | 13:31 |
| weshay|ruck | chkumar|rover, https://meet.google.com/vui-awyj-qbh?authuser=1 | 13:32 |
| *** jpena|lunch is now known as jpena | 13:35 | |
| weshay|ruck | rlandy, anyone else using bm env d? | 13:37 |
| weshay|ruck | rlandy, chandan and I are wondering about running rhel8 fs001 on bm | 13:37 |
| rlandy | weshay|ruck: I have been using env E but currently nobody is on env D | 13:38 |
| rlandy | weshay|ruck: we were going to use it for the bootcamp | 13:38 |
| weshay|ruck | rlandy, can you chat w/ us for a hot sec https://meet.google.com/vui-awyj-qbh?authuser=1 | 13:38 |
| rlandy | but I think we can get away with one env | 13:38 |
| arxcruz | zbr: https://review.opendev.org/#/c/704762/ please :) | 13:41 |
| marios | weshay|ruck: rfolco: looks like mark abandoned https://review.opendev.org/#/c/692368/ | 13:46 |
| marios | weshay|ruck: rfolco: so perhaps master is correct as folco is trying already ? we don't need a patch? | 13:46 |
| rfolco | marios, yes, but merged code includes a lot of skips | 13:46 |
| rfolco | marios, docker/base/Dockerfile.j2: # FIXME(mgoddard): Not available for CentOS 8 yet. | 13:47 |
| marios | rfolco: i see | 13:47 |
| rfolco | i am trying to comment out ceph to see what happens | 13:47 |
| rfolco | ceph excludes* | 13:47 |
| *** saneax has quit IRC | 13:49 | |
| chkumar|rover | zbr, weshay|ruck rlandy sshnaidm|afk marios arxcruz panda|bbl rfolco found something interesting https://github.com/cs01/termpair | 13:49 |
| arxcruz | cool | 13:50 |
| arxcruz | it's a tmate for browser | 13:50 |
| weshay|ruck | chkumar|rover, https://review.rdoproject.org/r/#/c/24803/ comments on that | 14:01 |
| *** amoralej|lunch is now known as amoralej | 14:01 | |
| chkumar|rover | weshay|ruck, yes, I will hack based on job type | 14:02 |
| weshay|ruck | chkumar|rover, k.. be really careful here | 14:02 |
| zbr | chkumar|rover: afaik, you need it only for docs for the moment, enable for these only and we shoud be safe, i deal was to have a enable-epel role, as is easier to reuse. | 14:03 |
| chkumar|rover | zbr, it is needed for linters | 14:03 |
| zbr | i think i even have an upstream CR for adding epel option to setup-repos, argument. | 14:03 |
| chkumar|rover | precommit is pulling re2-devel | 14:03 |
| chkumar|rover | may be ansible lint | 14:04 |
| zbr | i remember fixing this issue one or two weeks ago,... bit strange. | 14:04 |
| *** dtantsur is now known as dtantsur|brb | 14:05 | |
| zbr | chkumar|rover: you see https://github.com/rdo-infra/rdo-jobs/blob/master/bindep.txt#L13 | 14:09 |
| zbr | the reson for not conditon is because it comes from epel | 14:10 |
| zbr | even if you put epel-release here it may not help as it is not run as a script | 14:10 |
| zbr | evrika! i remembered where | 14:11 |
| weshay|ruck | chkumar|rover, https://beaker.engineering.redhat.com/view/rdoci-hp-01.v100.rdoci.lab.eng.rdu2.redhat.com#power-settings | 14:12 |
| zbr | chkumar|rover: it was when i moved molecule to ansible zuul, try to sync deps from https://github.com/ansible-community/molecule/blob/master/bindep.txt | 14:14 |
| zbr | i do no think that re2-devel is mandatory, ZAC is already using centos-8 for tox jobs. | 14:15 |
| *** ykarel|away is now known as ykarel | 14:20 | |
| *** TrevorV has joined #oooq | 14:47 | |
| weshay|ruck | rlandy, chkumar|rover https://code.engineering.redhat.com/gerrit/#/c/189436/7/zuul.d/projects.yaml | 14:47 |
| *** dpawlik has joined #oooq | 14:48 | |
| *** panda|bbl is now known as panda | 14:49 | |
| *** sshnaidm|afk is now known as sshnaidm | 14:51 | |
| *** ykarel is now known as ykarel|away | 15:08 | |
| marios | panda: o/ are you available for a quick call at some point please should be brief. i am adding a ppc_manifests switch into the manifest push so we can disable ppc but keep the rest with a switch. having some issues decoding that product https://github.com/rdo-infra/ci-config/blob/a0e0049280218c37b9f9a3ad5ee2d8f884437e6b/ci-scripts/infra-setup/roles/promoter/molecule/container-push/playbook.yml#L126 in | 15:16 |
| marios | particular where containers is set. so it fails now when ppc is disabled cos it tries to check the ppc ones too. | 15:16 |
| weshay|ruck | rfolco, marios ok.. have a build running w/o patches | 15:18 |
| weshay|ruck | Error: No matching repo to modify: influxdb. | 15:18 |
| weshay|ruck | on the base container | 15:19 |
| weshay|ruck | I think rfolco mentioned that | 15:19 |
| weshay|ruck | rfolco, we should probably update the tripleo-ci-testing pin on centos-8 | 15:20 |
| weshay|ruck | marios, rfolco k.. we have wf on the tripleo-repos patch | 15:24 |
| marios | weshay|ruck: ack cool will be one less thing | 15:24 |
| panda | marios: today ? | 15:24 |
| marios | panda: sure if you can | 15:25 |
| chkumar|rover | weshay|ruck, I am logging off now, will follow the pad tomorrow | 15:25 |
| chkumar|rover | see ya | 15:25 |
| chkumar|rover | weshay|ruck, please take care of the cix | 15:25 |
| weshay|ruck | chkumar|rover, ya.. I got you :) | 15:25 |
| marios | panda: if not then i'll have to reserve a slot for tomorrow morning np | 15:25 |
| weshay|ruck | chkumar|rover, thanks.. get some rest | 15:25 |
| chkumar|rover | weshay|ruck, on tempest side cix, all patches are about to merge, final is still pending | 15:25 |
| chkumar|rover | weshay|ruck, this two jobs needs attension http://zuul.opendev.org/t/openstack/builds?job_name=tripleo-ci-centos-7-scenario002-standalone&job_name=tripleo-ci-centos-7-scenario004-standalone&branch=stable%2Fstein&pipeline=check | 15:26 |
| panda | marios: in 30 minutes ok or too late for you ? | 15:26 |
| chkumar|rover | red from long time | 15:26 |
| chkumar|rover | or file a bug tomorrow | 15:26 |
| chkumar|rover | weshay|ruck, on scenar02 adlee is looking | 15:26 |
| weshay|ruck | chkumar|rover, nice | 15:26 |
| chkumar|rover | that's it | 15:26 |
| marios | panda: yes fine thanks | 15:26 |
| *** chkumar|rover is now known as raukadah | 15:27 | |
| raukadah | see ya | 15:27 |
| rlandy | raukadah: weshay|ruck: I'm merging the review to add the fs021 job definition | 15:28 |
| *** fmount has quit IRC | 15:52 | |
| *** dtantsur|brb is now known as dtantsur | 15:53 | |
| *** fmount has joined #oooq | 15:53 | |
| panda | marios: where ? | 16:00 |
| marios | panda: o/ so perhaps we can skip the call... worked out the thing i was stuck on. if you like though you can have a look at that please https://review.rdoproject.org/r/#/c/24807/ | 16:00 |
| marios | panda: if you would like to we can still talk but I got past the issue i had | 16:00 |
| marios | panda: i added an extra test in the playbook for the container push scenario so we can check if its ok | 16:01 |
| panda | marios: looking | 16:01 |
| marios | panda: ack no rush can be later but thanks and sorry to mess you around (and thanks for being available to talk anyway) | 16:02 |
| marios | panda:++ | 16:02 |
| *** Goneri has joined #oooq | 16:02 | |
| *** Trevor_V has joined #oooq | 16:03 | |
| *** openstackstatus has joined #oooq | 16:04 | |
| *** ChanServ sets mode: +v openstackstatus | 16:04 | |
| weshay|ruck | rfolco, do you know anything about the influxdb error? | 16:05 |
| weshay|ruck | Error: No matching repo to modify: influxdb. | 16:05 |
| *** TrevorV has quit IRC | 16:07 | |
| rfolco | weshay|ruck, yes, I am mocking it like I did with the others | 16:08 |
| weshay|ruck | rfolco, ah geez.. | 16:08 |
| rfolco | weshay|ruck, we need a patch somewhere | 16:08 |
| weshay|ruck | k.. /me looks at your hackmd | 16:08 |
| marios | weshay|ruck: rfolco: panda: created a new task for that ppc thing * https://tree.taiga.io/project/tripleo-ci-board/task/1511 (under consolidate promoter tests * https://tree.taiga.io/project/tripleo-ci-board/us/1493 ) | 16:09 |
| weshay|ruck | marios, thanks | 16:09 |
| rfolco | weshay|ruck, for now I am just creating a fake repo and doing this in dockerfile | 16:10 |
| rfolco | COPY influxdb.repo /etc/yum.repos.d/. | 16:10 |
| rfolco | I can't find why cinder is being skipped | 16:10 |
| panda | marios: who sets the ppc_manifest var ? | 16:13 |
| marios | panda: well for now we can default it in the vars/main. afterwards we want to add some conditional set_fact in the mainfest_push.yml... weshay|ruck was thinking along the lines of 'if the first 5 ppc containers fail, disable it' | 16:14 |
| rfolco | weshay|ruck, we still have bizarre conditions in kolla code like this | 16:14 |
| rfolco | {% if distro_package_manager == 'dnf' or base_distro not in ['centos'] %} | 16:14 |
| rfolco | {% set cinder_base_packages = cinder_base_packages + [ | 16:14 |
| rfolco | 'qemu-img' | 16:14 |
| rfolco | ] %} | 16:14 |
| marios | panda: commented there https://review.rdoproject.org/r/#/c/24807/1/ci-scripts/container-push/roles/containers-promote/tasks/manifest-push.yml | 16:15 |
| sshnaidm | zbr, wrt: https://review.opendev.org/#/c/705623/ I think it's worth to leave redirection to files in commands themselves, the construction "cmd" > "file" makes it overcomplicated, error prone, unpredictable results. | 16:18 |
| sshnaidm | zbr, and you don't need this to run it in parallel | 16:18 |
| weshay|ruck | rfolco, marios related patch https://review.opendev.org/#/c/704970/3/kolla/template/repos.yaml | 16:19 |
| sshnaidm | zbr, you don't have locks for same file, so why not to include them in the "cmd" body | 16:19 |
| *** rascasoft_ has joined #oooq | 16:19 | |
| owalsh | weshay|ruck: is it ok to recheck https://review.opendev.org/704880? | 16:19 |
| *** rascasoft has quit IRC | 16:19 | |
| weshay|ruck | owalsh, no need.. it should kick after https://review.opendev.org/#/c/704919/ merges | 16:20 |
| weshay|ruck | owalsh, I'll watch it.. I'm only 90% sure of that fact :) | 16:21 |
| owalsh | weshay|ruck: ack, didn't realise zuul was that smart :-) | 16:21 |
| zbr | i disagree here, doing the redirection ourselves does simplify the command definitions and assure consistency of how they are handled. | 16:21 |
| weshay|ruck | rfolco, marios so influx works on rhel8 | 16:22 |
| zbr | for the moment parallel execution is not part of the change, it is considered but not part. | 16:22 |
| zbr | also this approach is much safer than current one where is very easy for a user to miss adding a redirection and spam ansible with too much stdout output. | 16:23 |
| zbr | by keeping the redirection in one place, we assure it is kept that way. | 16:23 |
| *** udesale_ has quit IRC | 16:24 | |
| weshay|ruck | rfolco, when you have a sec can you please update https://hackmd.io/dSagCbocQ4KSVEZR1uf8Tw | 16:24 |
| rlandy | weshay|ruck: raukadah: https://code.engineering.redhat.com/gerrit/#/c/191279/ - let's see how far that gets | 16:28 |
| rlandy | I have some suspicion that we will have to do more steps to get the right release file on the virt undercloud. but we'll see | 16:28 |
| rfolco | weshay|ruck, will do asap... I just wanted to get some progress with master, but some are skipped somehow... I can't find where | 16:29 |
| *** jmasud has joined #oooq | 16:29 | |
| weshay|ruck | rlandy++ | 16:29 |
| weshay|ruck | re: influx CentOS one is not usable due to wrong arch name used (arm64 instead of | 16:30 |
| weshay|ruck | aarch64). | 16:30 |
| weshay|ruck | rfolco, added influx to the skip list got passed it | 16:35 |
| weshay|ruck | also noticed.. kolla creates the influx.repo for the containers | 16:35 |
| weshay|ruck | name = InfluxDB Repository - RHEL $releasever | 16:35 |
| weshay|ruck | baseurl = https://repos.influxdata.com/rhel/$releasever/$basearch/stable | 16:35 |
| weshay|ruck | enabled = 1 | 16:35 |
| weshay|ruck | gpgcheck = 1 | 16:35 |
| weshay|ruck | gpgkey = https://repos.influxdata.com/influxdb.key | 16:35 |
| weshay|ruck | [zuul@wes base]$ | 16:35 |
| rfolco | weshay|ruck, influx is just noise. | 16:35 |
| weshay|ruck | rfolco, marios job is getting to the build https://review.rdoproject.org/zuul/stream/cdb92414fad74d388fae1a1177d3f945?logfile=console.log | 16:36 |
| weshay|ruck | heh.. then failing | 16:36 |
| weshay|ruck | https://logserver.rdoproject.org/95/24795/3/check/tripleo-build-containers-centos-8-master/cdb9241/logs/build.log | 16:38 |
| weshay|ruck | 2020-02-05 16:36:06 | sudo: buildah: command not found | 16:39 |
| *** panda is now known as panda|bbl | 16:40 | |
| weshay|ruck | rfolco, sorry pita here.. | 16:42 |
| weshay|ruck | rfolco, not seeing where we ensure buildah is installed | 16:42 |
| weshay|ruck | thinking to add to the bindep | 16:42 |
| weshay|ruck | maybe it's preinstalled in rhel? | 16:42 |
| rfolco | weshay|ruck, https://github.com/openstack/tripleo-ci/blob/master/roles/build-containers/tasks/pre.yaml#L46 | 16:43 |
| rfolco | should have installed | 16:43 |
| weshay|ruck | thanks | 16:44 |
| rfolco | https://review.rdoproject.org/r/#/c/24775/11/zuul.d/build-containers.yaml@57 | 16:44 |
| rfolco | weshay|ruck, I just don't know why it didn't install it | 16:44 |
| weshay|ruck | 020-02-05 16:34:36.803151 | TASK [build-containers : Install Buildah] | 16:44 |
| weshay|ruck | 2020-02-05 16:34:36.840759 | primary | skipping: Conditional result was False | 16:44 |
| weshay|ruck | rfolco, I'll get it | 16:44 |
| weshay|ruck | rlandy, rhel bm needs this patch https://review.opendev.org/#/c/702100/ | 16:51 |
| rlandy | ok - adding | 16:52 |
| *** bogdando has quit IRC | 16:54 | |
| *** marios is now known as marios|out | 16:56 | |
| weshay|ruck | rfolco, does the containers build need to run as root or sudo? | 17:08 |
| rfolco | weshay|ruck, manually I run as normal user | 17:09 |
| rfolco | openstack overcloud container image build --kolla-config-file /home/centos/build/kolla-build.conf --exclude opendaylight --use-buildah > build.log 2>&1 | 17:09 |
| weshay|ruck | http://pastebin.test.redhat.com/833386 | 17:10 |
| rfolco | weshay|ruck, try to run that command inside your base container, you'll see what is failing | 17:13 |
| rfolco | weshay|ruck, one of those repos probably doesn't exist | 17:13 |
| weshay|ruck | rfolco, just got perm errors | 17:14 |
| weshay|ruck | but will double check | 17:14 |
| weshay|ruck | thanks | 17:14 |
| weshay|ruck | rfolco, so do I need to copy the repos from /etc/yum.repos.d/ to the base container w/ these kolla names? | 17:17 |
| weshay|ruck | like .. PowerTools | 17:18 |
| *** marios|out has quit IRC | 17:18 | |
| rfolco | first time I built I used rhel's method, build a base one and kolla builds on top of it | 17:19 |
| rfolco | and I injected all repos there in the dockerfile | 17:19 |
| rfolco | then built the base one | 17:19 |
| rfolco | sudo buildah bud -t centos:8 . | 17:19 |
| rfolco | now I am trying like you | 17:19 |
| rfolco | without a base one | 17:20 |
| rfolco | so let me see what we need to do with repos | 17:20 |
| rfolco | weshay|ruck, yeah if you don't build a base one injecting repos, you'll miss opstools, then ha, influx, etc | 17:24 |
| weshay|ruck | rfolco, ah cool | 17:24 |
| *** soniya29 has joined #oooq | 17:24 | |
| rfolco | [root@b0ee3447d0f4 /]# dnf -y install dnf-plugins-core centos-release-opstools && dnf config-manager --enable PowerTools && dnf config-manager --disable influxdb && dnf -y distro-sync --security --sec-severity=Important --sec-severity=Critical && dnf clean all && rm -rf /var/cache/dnf | 17:24 |
| weshay|ruck | rfolco, so this file https://review.opendev.org/#/c/702993/17/roles/build-containers/templates/Dockerfile.centos8.j2 | 17:24 |
| rfolco | No match for argument: centos-release-opstools | 17:24 |
| *** aakarsh has joined #oooq | 17:25 | |
| sshnaidm | zbr, I think we have to do redirection in task, sorry, in your patch it gives unpredictable results, and this is really confusing for future users, if you want I can explain in mtg | 17:25 |
| rfolco | yes should be this one. As long as you build a base one called centos, it will be used by kolla | 17:26 |
| weshay|ruck | 2020-02-05 17:24:15.927847 | TASK [build-containers : Install Buildah] | 17:26 |
| weshay|ruck | 2020-02-05 17:24:27.472439 | primary | changed | 17:26 |
| weshay|ruck | fixed the install of buildah | 17:26 |
| sshnaidm | zbr, maybe if you run the whole cmd in subshell, then maybe.. not sure | 17:26 |
| weshay|ruck | rfolco, if you can pastebin what you think https://review.opendev.org/#/c/702993/17/roles/build-containers/templates/Dockerfile.centos8.j2 should look like | 17:26 |
| weshay|ruck | that would help us sync | 17:27 |
| zbr | sshnaidm: please commend on review with example. | 17:27 |
| sshnaidm | zbr, example of what? | 17:27 |
| zbr | of when it breaks | 17:28 |
| rfolco | weshay|ruck, line 127 is my dockerfile http://pastebin.test.redhat.com/833389 | 17:28 |
| weshay|ruck | oh I see | 17:28 |
| sshnaidm | zbr, can you please re-read my message again? | 17:29 |
| rfolco | weshay|ruck, re: buildah install, cool thanks | 17:29 |
| rfolco | weshay|ruck, don't ask me why, the base container I built w/ sudo buildah bud... then openstack build command without sudo | 17:30 |
| rfolco | I had to build* | 17:30 |
| weshay|ruck | rfolco, don't think we can use epel btw | 17:30 |
| rfolco | weshay|ruck, epel is fake | 17:30 |
| weshay|ruck | k | 17:31 |
| rfolco | those repos are fake except ceph_master | 17:31 |
| rfolco | its just to get something built | 17:31 |
| rfolco | weshay|ruck, also need to patch kolla... kolla/image/build.py, remove skips for centos8 like ceph, collectd | 17:32 |
| rfolco | weshay|ruck, on top of master ^ | 17:32 |
| weshay|ruck | which patch? | 17:32 |
| rfolco | I hacked it manually, if this works now, I'll send out the patch | 17:32 |
| *** rascasoft_ has quit IRC | 17:32 | |
| weshay|ruck | k | 17:33 |
| weshay|ruck | thanks | 17:33 |
| rfolco | sorry, didn't get a chance to structure the steps yet | 17:33 |
| rfolco | going on trial and error methodology | 17:33 |
| rfolco | weshay|ruck, for example, what we have for influx? you know? who knows ? | 17:34 |
| weshay|ruck | rfolco, /me experimenting w/ the skip list for now | 17:36 |
| weshay|ruck | we should get that w/ ceph | 17:36 |
| weshay|ruck | when the repos are avail | 17:36 |
| rfolco | weshay|ruck, ceph is in skip list, so barbican-api, cinder-* will be skipped | 17:37 |
| rlandy | weshay|ruck: 1-on-1? or do you want to skip it? | 17:37 |
| weshay|ruck | rlandy, I can chat | 17:37 |
| *** aakarsh has quit IRC | 17:38 | |
| *** dtantsur is now known as dtantsur|afk | 17:48 | |
| *** jaosorior has quit IRC | 17:49 | |
| *** jpena is now known as jpena|off | 17:53 | |
| weshay|ruck | http://git.app.eng.bos.redhat.com/git/tripleo-environments.git/tree/roles/prep-internal-host | 17:58 |
| weshay|ruck | rlandy, ^ | 17:59 |
| *** dsneddon has joined #oooq | 18:04 | |
| *** soniya29 has quit IRC | 18:07 | |
| *** tesseract has quit IRC | 18:11 | |
| *** chem has quit IRC | 18:16 | |
| *** amoralej is now known as amoralej|off | 18:16 | |
| *** dpawlik has quit IRC | 19:04 | |
| *** dpawlik has joined #oooq | 19:05 | |
| rfolco | weshay|ruck, RUN echo 'binary not yet available for centos' | 19:09 |
| rfolco | in nova-libvirt | 19:09 |
| *** holser has quit IRC | 19:15 | |
| rfolco | need to find out what changed in centos8 for libvirt | 19:20 |
| rfolco | https://github.com/openstack/kolla/blob/master/docker/nova/nova-libvirt/Dockerfile.j2#L46-L54 | 19:20 |
| rfolco | weshay|ruck, ^ | 19:20 |
| weshay|ruck | k.. sec in 1-1 | 19:27 |
| weshay|ruck | rfolco, 5min chat? | 19:34 |
| rfolco | yeah | 19:34 |
| rfolco | weshay|ruck, gmeet? | 19:35 |
| weshay|ruck | https://meet.google.com/twm-pdro-dbp?authuser=1 | 19:36 |
| weshay|ruck | rfolco, https://etherpad.openstack.org/p/kolla-centos8 | 19:42 |
| weshay|ruck | rlandy, sorry to bug you.. | 19:50 |
| weshay|ruck | what was the internal repo w/ Dockerfiles.. that's not in kolla? | 19:50 |
| * weshay|ruck looking at https://hackmd.io/6gE-LZYMSzmoP3IFt7puKg | 19:50 | |
| weshay|ruck | https://pkgs.devel.redhat.com/cgit/containers/ | 19:51 |
| weshay|ruck | maybe | 19:51 |
| weshay|ruck | 2020-02-05 19:25:18 | Stderr: 'error building at STEP "RUN dnf -y install ceph-common cyrus-sasl-scram libguestfs libvirt-client libvirt-daemon libvirt-daemon-config-nwfilter libvirt-daemon-driver-nwfilter openvswitch trousers qemu-kvm edk2-ovmf libseccomp openssl-perl openstack-nova-migration podman && dnf clean all && rm -rf /var/cache/dnf": error while running runtime: exit status 1\n' | 19:53 |
| rlandy | weshay|ruck: checking | 20:06 |
| *** jtomasek has quit IRC | 20:06 | |
| rlandy | http://pkgs.devel.redhat.com/cgit/containers/openstack-nova-api | 20:06 |
| rlandy | weshay|ruck: they are all in their own repos | 20:06 |
| rlandy | base is in http://pkgs.devel.redhat.com/cgit/containers/openstack-base | 20:07 |
| rlandy | weshay|ruck: take 7 patches | 20:07 |
| *** jmasud has quit IRC | 20:11 | |
| weshay|ruck | rfolco, ok.. adapting this https://opendev.org/openstack/tripleo-ci/src/branch/master/roles/build-containers/tasks/main.yaml#L108-L144 | 20:12 |
| rfolco | weshay|ruck, yes. This is a way when you want a custom base | 20:12 |
| rfolco | weshay|ruck, or change that centos8.dockerfile + pre tasks | 20:13 |
| *** jmasud has joined #oooq | 20:14 | |
| *** holser has joined #oooq | 20:41 | |
| sshnaidm | rlandy, so do you want to change "release" to "baremetal_provision" in py script? | 20:43 |
| rlandy | sshnaidm: yes | 20:43 |
| rlandy | sshnaidm: I have another patch ready | 20:44 |
| sshnaidm | rlandy, cool | 20:44 |
| rlandy | but your patch is testing now | 20:44 |
| rlandy | sshnaidm: I'd like to see it complete testing before I switch | 20:44 |
| sshnaidm | rlandy, ack | 20:44 |
| weshay|ruck | rfolco, fyi.. ceph is not in the expected to build | 20:45 |
| weshay|ruck | list | 20:45 |
| rfolco | weshay|ruck, don't see any skips in the code or errors in the log | 20:51 |
| mjturek | rfolco: was able to upload a known working image today. Upon pulling it from rdoproject.org, I still get the EOF error | 20:58 |
| rfolco | mjturek, wtf | 20:59 |
| mjturek | rfolco: it's still that first chunk | 21:01 |
| *** jfrancoa has quit IRC | 21:01 | |
| rfolco | mjturek, what are you using to push/pull | 21:01 |
| rfolco | mjturek, nm, sorry I am mixing problems | 21:08 |
| mjturek | rfolco: docker for push and pull (just to be clear) | 21:14 |
| mjturek | rfolco: do you know what registry software they use? | 21:15 |
| baha | (Also, just to reiterate, we were able to push these images up to and down from dockerhub just fine) | 21:15 |
| rfolco | yeah must be on server side since I get the same EOF | 21:15 |
| rfolco | sir, unplug your modem and re-plug it back | 21:16 |
| rfolco | mjturek, I am almost giving you dummy suggestions like this ^ | 21:16 |
| mjturek | lololol | 21:17 |
| baha | are you sure your build server is connected to the internet | 21:18 |
| *** tosky has quit IRC | 21:18 | |
| baha | I mentioned this to mjturek, but I'm wondering if it's possible that the registry is tossing new uploads, since the base container isn't actually changing hash/version when we upload it. Would be something to check with the RDO folks | 21:23 |
| baha | If we had an upload that didn't work quite right and future uploads are being ignored, that could cause it | 21:24 |
| *** rfolco has quit IRC | 21:27 | |
| *** holser has quit IRC | 21:30 | |
| *** rfolco has joined #oooq | 21:31 | |
| *** rfolco has quit IRC | 21:35 | |
| mjturek | baha: unfortunately don't know enough about registries to know if that's plausible or not | 21:38 |
| baha | same | 21:38 |
| baha | I would hope checksumming would occur at some step in the chain to prevent that, but have no other immediate ideas | 21:39 |
| *** yolanda has joined #oooq | 22:35 | |
| *** Trevor_V has quit IRC | 22:40 | |
| *** rlandy is now known as rlandy|bbl | 23:25 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!