| *** weshay|ruck is now known as weshay_PTO | 00:28 | |
| *** fultonj has joined #oooq | 02:04 | |
| *** fultonj has quit IRC | 02:06 | |
| *** ykarel|away has joined #oooq | 03:10 | |
| *** udesale has joined #oooq | 03:44 | |
| *** skramaja has joined #oooq | 03:47 | |
| *** ykarel|away is now known as ykarel | 04:00 | |
| *** agopi has quit IRC | 04:15 | |
| *** ykarel has quit IRC | 04:58 | |
| *** ykarel has joined #oooq | 05:19 | |
| *** jbadiapa has joined #oooq | 06:02 | |
| *** chandankumar is now known as chkumar|ruck | 06:17 | |
| chkumar|ruck | quiquell|off: Hello | 06:26 |
|---|---|---|
| chkumar|ruck | ykarel: Hello | 06:27 |
| ykarel | chkumar|ruck, hi | 06:27 |
| chkumar|ruck | ykarel: please have a look at this bug https://bugs.launchpad.net/tripleo/+bug/1816026 | 06:27 |
| openstack | Launchpad bug 1816026 in tripleo "multinode promotion jobs timing out at overcloud deploy" [Critical,In progress] - Assigned to chandan kumar (chkumar246) | 06:27 |
| chkumar|ruck | ykarel: it is kind of weired, when I tried to run it in test project, it failed with tempest docker error and overcloud deploy passed | 06:28 |
| ykarel | looking | 06:28 |
| chkumar|ruck | ykarel: I am not sure what is leading to this error https://logs.rdoproject.org/openstack-periodic/git.openstack.org/openstack-infra/tripleo-ci/master/periodic-tripleo-ci-centos-7-multinode-1ctlr-featureset019-master/cf698b1/logs/undercloud/var/log/containers/mistral/engine.log.txt.gz?level=ERROR | 06:28 |
| ykarel | chkumar|ruck, so tempest is failing because it's not using podman | 06:38 |
| ykarel | it' using docker which is removed from undercloud :- https://review.openstack.org/#/c/636442/ | 06:39 |
| ykarel | and as overcloud is passing with the mistral revert, that mistral patch is somewhat related | 06:40 |
| ykarel | good to ask someone from mistral as issue seems around https://github.com/openstack/tripleo-common/blob/master/workbooks/deployment.yaml#L404 | 06:42 |
| ykarel | ^^ is returning True now somehow | 06:43 |
| ykarel | chkumar|ruck, the error you pointed is obvious when it tries to access a blank list | 06:44 |
| ykarel | running_config_download_workflows = [] | 06:44 |
| *** jtomasek has joined #oooq | 06:49 | |
| chkumar|ruck | ykarel: yes that is weired | 07:04 |
| chkumar|ruck | ykarel: regading tempest tempest can run on podman | 07:04 |
| chkumar|ruck | ykarel: support was already added for the same | 07:04 |
| ykarel | where? | 07:04 |
| chkumar|ruck | ykarel: https://review.openstack.org/#/c/605356/ | 07:05 |
| ykarel | chkumar|ruck, and where you enabled it | 07:06 |
| ykarel | the patch is just adding support | 07:06 |
| ykarel | https://review.openstack.org/#/c/605356/55/roles/validate-tempest/templates/run-tempest.sh.j2@68 | 07:07 |
| chkumar|ruck | ykarel: it depends upon undercloud_container_cli | 07:08 |
| ykarel | yup | 07:08 |
| ykarel | where you set it to podman for master | 07:08 |
| chkumar|ruck | ykarel: if it is defined in fs everything is taken care of | 07:09 |
| ykarel | chkumar|ruck, but most of the fs don't have it | 07:09 |
| ykarel | so docker is used | 07:09 |
| ykarel | and it's failing | 07:09 |
| chkumar|ruck | ykarel: then it is abug in fs side | 07:09 |
| ykarel | :) | 07:10 |
| chkumar|ruck | ykarel: http://codesearch.openstack.org/?q=undercloud_container_cli&i=nope&files=.yml&repos=tripleo-quickstart | 07:10 |
| chkumar|ruck | ykarel: let me check fs16 | 07:10 |
| chkumar|ruck | ykarel: let me fix that part | 07:14 |
| chkumar|ruck | if there is a containerized uc there we need to set undercloud_container_cli | 07:15 |
| *** jfrancoa has joined #oooq | 07:17 | |
| *** sanjayu__ has joined #oooq | 07:17 | |
| *** apetrich has joined #oooq | 07:19 | |
| *** quiquell|off is now known as quiquell|rover | 07:23 | |
| quiquell|rover | chkumar|ruck: o/ | 07:23 |
| quiquell|rover | chkumar|ruck: https://github.com/gnocchixyz/gnocchi/pull/1017 merged | 07:25 |
| quiquell|rover | but we need a promotion | 07:25 |
| chkumar|ruck | quiquell|rover: we have to wait till they create a new tag | 07:25 |
| chkumar|ruck | quiquell|rover: gnocchi is pinned in rdoinfo | 07:25 |
| chkumar|ruck | quiquell|rover: we have a serious bug in our fs | 07:26 |
| quiquell|rover | chkumar|ruck: Ahh yep, rdoinfo maybe we can unpin it ? | 07:26 |
| quiquell|rover | chkumar|ruck: what happend ? | 07:26 |
| quiquell|rover | the timeout ? | 07:26 |
| chkumar|ruck | quiquell|rover: where containerized_undercloud is defined, there undercloud_container_cli is not defined | 07:26 |
| chkumar|ruck | in most of the fs | 07:26 |
| quiquell|rover | chkumar|ruck: maybe we can unpin it from rdoinfo so we just depend on promotions ? | 07:26 |
| chkumar|ruck | quiquell|rover: let me try that for rdoinfo | 07:27 |
| *** jtomasek has quit IRC | 07:27 | |
| quiquell|rover | chkumar|ruck: the cli is docker by default I think | 07:27 |
| quiquell|rover | chkumar|ruck: humm no this is standalone | 07:27 |
| chkumar|ruck | quiquell|rover: I am taking from fs16 | 07:27 |
| quiquell|rover | chkumar|ruck: so we remove the pin or we update the pin ? | 07:27 |
| apetrich | morning chkumar|ruck | 07:27 |
| *** jtomasek has joined #oooq | 07:27 | |
| apetrich | what can I do to help you with https://bugs.launchpad.net/tripleo/+bug/1816026 ? | 07:28 |
| openstack | Launchpad bug 1816026 in tripleo "multinode promotion jobs timing out at overcloud deploy" [Critical,In progress] - Assigned to chandan kumar (chkumar246) | 07:28 |
| quiquell|rover | chkumar|ruck:f fs016 is not mistral ? | 07:28 |
| chkumar|ruck | quiquell|rover: check my https://bugs.launchpad.net/tripleo/+bug/1816026 comment 6 | 07:28 |
| chkumar|ruck | quiquell|rover: tempest error you will get it | 07:28 |
| quiquell|rover | chkumar|ruck: ack so is not mistral | 07:29 |
| chkumar|ruck | quiquell|rover: if I run the job seperatly it passes the overcloud | 07:29 |
| chkumar|ruck | but in promotion it is returning timedout | 07:29 |
| quiquell|rover | chkumar|ruck: if you run it at check it uses diferent release file | 07:31 |
| quiquell|rover | chkumar|ruck: And it means different mistral version | 07:31 |
| quiquell|rover | chkumar|ruck: and different vars | 07:32 |
| chkumar|ruck | quiquell|rover: https://github.com/gnocchixyz/gnocchi/pull/1018 already backported to stable/4.3 | 07:32 |
| chkumar|ruck | quiquell|rover: let me just re-run the job | 07:32 |
| quiquell|rover | chkumar|ruck: you have to hack this https://github.com/openstack-infra/tripleo-ci/blob/master/roles/common/vars/main.yaml#L14 | 07:34 |
| quiquell|rover | chkumar|ruck: to use same release file at periodic | 07:34 |
| *** quiquell|rover is now known as quiquell|rover|b | 07:36 | |
| *** dtantsur|afk is now known as dtantsur | 07:42 | |
| *** sshnaidm|off is now known as sshnaidm | 07:56 | |
| chkumar|ruck | apetrich: Hello | 07:57 |
| chkumar|ruck | apetrich: regarding mistral issue https://bugs.launchpad.net/tripleo/+bug/1816026/comments/8 | 07:57 |
| openstack | Launchpad bug 1816026 in tripleo "multinode promotion jobs timing out at overcloud deploy" [Critical,In progress] - Assigned to chandan kumar (chkumar246) | 07:57 |
| *** ykarel is now known as ykarel|lunch | 07:58 | |
| chkumar|ruck | apetrich: in the check job old commit is used https://github.com/openstack/mistral/commit/58d6634702014eb3c2bf70dbe487fe5983cef8e9 but when this https://github.com/openstack/mistral/commit/a39db2d3dc21afbe8b9e1d6f8dd870da272b9671 is used, it is leading to timedout | 07:58 |
| chkumar|ruck | in periodic pipeline | 07:58 |
| chkumar|ruck | apetrich: i think revert fixes the issue or we need other fix | 07:59 |
| chkumar|ruck | apetrich: please have a look | 07:59 |
| *** holser_ has joined #oooq | 07:59 | |
| apetrich | chkumar|ruck, I am. I'm trying to talk to other mistral folk but I think it is better to revert first | 08:01 |
| quiquell|rover|b | I think issue is just python cannot discover type of empty list | 08:04 |
| quiquell|rover|b | Does not have to be difficult to put a patch for it with a unit test | 08:04 |
| quiquell|rover|b | So we don't have to revert | 08:04 |
| *** kopecmartin|off is now known as kopecmartin | 08:04 | |
| *** quiquell|rover|b is now known as quiquell|rovee | 08:05 | |
| quiquell|rovee | Damn | 08:05 |
| *** quiquell|rovee is now known as quiquell|rover | 08:05 | |
| *** gkadam has joined #oooq | 08:06 | |
| *** gkadam has quit IRC | 08:09 | |
| *** panda|off is now known as panda | 08:09 | |
| *** rascasoft has joined #oooq | 08:09 | |
| panda | zbr: marios rfolco my user story is Done, so I'm available to help for the promotion stuff | 08:19 |
| marios | panda: ack thanks. my job merged but i didn't see it yet (can't quickly find in https://review.rdoproject.org/zuul/builds yet but didn't dig too much | 08:21 |
| marios | panda: so i'll be doing that today and then debug | 08:21 |
| marios | panda: folcos change need reviewing we said we'd merge those | 08:22 |
| marios | and zbr was gonna incorporate his findings into folco change | 08:22 |
| marios | s | 08:22 |
| panda | marios: your job is periodic, will run 4 times a day. When did it merge ? | 08:22 |
| marios | panda: yah but friday | 08:23 |
| *** amoralej|off is now known as amoralej | 08:23 | |
| marios | oh | 08:23 |
| marios | panda: i think friday lemme check again | 08:23 |
| marios | panda: yeah https://review.rdoproject.org/r/#/c/18729/ | 08:23 |
| panda | marios: weird, I don't see it too. https://logs.rdoproject.org/openstack-periodic/git.openstack.org/openstack-infra/tripleo-ci/master/ | 08:26 |
| panda | I'm afraid it fails in a way that it cant even output logs | 08:28 |
| quiquell|rover | marios: I think is here the place to look https://softwarefactory-project.io/zuul/t/rdoproject.org/builds | 08:29 |
| panda | there's a pipeline build right now in https://review.rdoproject.org/zuul/status | 08:30 |
| panda | but it's not there, so it's not running | 08:30 |
| quiquell|rover | the job is there https://softwarefactory-project.io/zuul/t/rdoproject.org/job/periodic-tripleo-centos-7-master-containers-build-push | 08:30 |
| marios | quiquell|rover: panda thanks just finishing some reviews and switching to in a bit | 08:31 |
| marios | quiquell|rover: thanks for pointer there | 08:31 |
| quiquell|rover | panda, marios: Have to be something with the config | 08:32 |
| panda | quiquell|rover:yep, I think it has not been picked up properly ... | 08:33 |
| quiquell|rover | Going to ask around | 08:33 |
| quiquell|rover | maybe they can check scheduler logs | 08:33 |
| *** tosky has joined #oooq | 08:33 | |
| quiquell|rover | panda, marios: here we have "Â3" attempts | 08:35 |
| quiquell|rover | 3 attempts | 08:35 |
| quiquell|rover | periodic-tripleo-centos-7-master-containers-build-push | 08:35 |
| quiquell|rover | ahh this is config | 08:37 |
| *** chem has joined #oooq | 08:37 | |
| marios | quiquell|rover: panda check https://review.openstack.org/637490 if you have 2 mins please (e.g. we should see scen1/4 running in https://review.openstack.org/#/c/637196/ | 08:39 |
| arxcruz | panda: sshnaidm can one of you guys +w https://review.openstack.org/#/c/636579/ ? | 08:40 |
| quiquell|rover | marios: commented | 08:42 |
| panda | arxcruz: done | 08:42 |
| *** bogdando has joined #oooq | 08:42 | |
| arxcruz | panda: ty | 08:42 |
| marios | quiquell|rover: thanks done | 08:47 |
| zbr | rfolco: panda did you had time to look at my change at https://review.openstack.org/#/c/636144/ ? i have no problems incorporating my changes to folco one but I want to get you to check it first. | 08:49 |
| panda | zbr: checking if I checked | 08:50 |
| zbr | i still have one weird issue where apparently openstack overcloud container image build returns 0 on case of total failure. see http://logs.openstack.org/44/636144/22/check/tripleo-build-containers-fedora-28/5f850ce/job-output.txt.gz#_2019-02-17_18_35_35_491820 | 08:51 |
| *** jpena|off is now known as jpena | 08:52 | |
| *** ykarel|lunch is now known as ykarel | 08:59 | |
| zbr | rfolco: are you also encountering http://logs.openstack.org/44/636144/19/check/tripleo-build-containers-fedora-28/fe7b613/job-output.txt.gz#_2019-02-17_13_09_43_706478 ? | 09:03 |
| marios | [docs - standalone jobs] o/ folks check it if you have 10 mins isn't too long :) comments please and lets merge it if you like it thanks https://review.openstack.org/#/c/635881/ http://logs.openstack.org/81/635881/2/check/openstack-tox-docs/597378b/html/ci/standalone_scenario_jobs.html | 09:04 |
| marios | quiquell|rover: panda rfolco weshay_PTO sshnaidm zbr ** ^ | 09:04 |
| zbr | it is a bit weird because I am cherry picking the change that adds supports for "base=fedora" and installing kolla with it. still at runtime, I get this. is like another copy of kolla would have being used. | 09:05 |
| quiquell|rover | marios: talking with amoralej and #zuul looks like base job have "files" section so it does not do periodic, but this can be a zuul bug | 09:08 |
| quiquell|rover | marios: let me reproduce something simmilar | 09:08 |
| quiquell|rover | marios: can you test files: [] in the push job ? | 09:11 |
| *** ccamacho has joined #oooq | 09:13 | |
| chkumar|ruck | quiquell|rover: apetrich any other places https://logs.rdoproject.org/openstack-periodic/git.openstack.org/openstack-infra/tripleo-ci/master/periodic-tripleo-ci-centos-7-multinode-1ctlr-featureset016-master/7e19d72/ I can look for more info like all the info related to overcloud deploy apart from mistral | 09:19 |
| quiquell|rover | humm | 09:26 |
| quiquell|rover | we didnt' have nothing in the gates this weekend | 09:26 |
| quiquell|rover | weird | 09:26 |
| chkumar|ruck | quiquell|rover: another weird http://zuul.openstack.org/status | 09:26 |
| chkumar|ruck | quiquell|rover: check tripleo queue | 09:26 |
| chkumar|ruck | quiquell|rover: 81 hr 46 min | 09:26 |
| chkumar|ruck | https://review.openstack.org/#/c/636934/ on this patch | 09:27 |
| quiquell|rover | chkumar|ruck: where do you see 81 hr ? | 09:27 |
| quiquell|rover | ahh yep | 09:27 |
| quiquell|rover | ... | 09:27 |
| quiquell|rover | well is the new job | 09:28 |
| quiquell|rover | good catch though | 09:28 |
| chkumar|ruck | :-) | 09:28 |
| marios | quiquell|rover: thanks ! sure can add it | 09:29 |
| marios | chkumar|ruck: ack added for next reviews cos i really need to switch now ! | 09:29 |
| quiquell|rover | marios: For me this is a zuul bug, periodic have to run always no mather what do we have at files | 09:29 |
| quiquell|rover | chkumar|ruck: can you put that job to the wiki ? | 09:33 |
| zbr | rfolco: are you working on addresing comments on https://review.openstack.org/#/c/637212 -- if not I can help. | 09:36 |
| marios | quiquell|rover: but i don't see any files | 09:36 |
| marios | quiquell|rover: panda at https://github.com/rdo-infra/review.rdoproject.org-config/blob/master/zuul.d/tripleo-rdo-base.yaml https://github.com/rdo-infra/review.rdoproject.org-config/blob/master/zuul.d/jobs.yaml | 09:36 |
| marios | quiquell|rover: ah we need it empty | 09:37 |
| marios | i get it now | 09:37 |
| marios | :) k adding | 09:37 |
| zbr | panda: please give some advise regarding ^^ -- regarding if it would not be better to just have --mirror instead of variable params on each base os: like --centos-mirror and --fedora-mirrors .... | 09:37 |
| quiquell|rover | marios: ack, let me know and we open a zuul bug | 09:37 |
| marios | quiquell|rover: well we won't know until we merge it | 09:37 |
| *** derekh has joined #oooq | 09:37 | |
| marios | quiquell|rover: luckilly its override https://zuul-ci.org/docs/zuul/user/config.html#attr-job.files not merging so should work if that is indeed the problem | 09:38 |
| quiquell|rover | marios: we can reproduce at reproducer | 09:40 |
| quiquell|rover | marios: want to try ? | 09:40 |
| quiquell|rover | marios: we can put in place a fake periodic pipeline | 09:40 |
| quiquell|rover | lot of fun :-) | 09:40 |
| marios | quiquell|rover: yah i need to re-setup (migrated laptop) so i should do that | 09:41 |
| marios | quiquell|rover: but i am kinda liking not having selinux alerts :D ;) | 09:41 |
| quiquell|rover | marios: you can run it at quickvm https://operations.cee.redhat.com/#/quickvm | 09:42 |
| quiquell|rover | marios: super fast to startup a f28 image | 09:42 |
| marios | quiquell|rover: nice i didn't know that (yah was thinking of grabbing a beaker box might do that) | 09:42 |
| quiquell|rover | marios: no need for beaker, well those image at quickvm are not very powerful | 09:43 |
| quiquell|rover | marios: but zuul is attacking your RDO tenant so no issue | 09:43 |
| marios | quiquell|rover: yah i mean some of the beaker boxes are pretty small like even 4gb ram and they come on a timer (like 5 days is max /99 hours but you can extend once for another 99) | 09:44 |
| panda | zbr: yes, distro is automatically detected, but how would you specify different mirrors for fedora or centos ? | 10:00 |
| zbr | panda: you can never specify multiple base OS mirrors, you can specify only one, for current OS. so it makes no sense to add one extra argument for each OS. | 10:01 |
| zbr | you just specify --distro --mirror | 10:01 |
| zbr | in fact distro is optional, and the safe mirror can also cover two OS-es at the same time. which means that tripleo-repos could be make to work on multiple platforms without having to call it with different parameters. | 10:03 |
| zbr | panda: makes sense what I say? i can make a poc change to demostrate it in few minutes. | 10:05 |
| rfolco | zbr, can you help with https://review.openstack.org/#/c/637212 please | 10:06 |
| zbr | rfolco: sure! | 10:07 |
| rfolco | zbr, please keep in mind --centos-mirror may need to exist for compatibility | 10:09 |
| zbr | rfolco: i know, not dropping any old behavior. | 10:10 |
| rfolco | zbr, thx for your help :) | 10:11 |
| quiquell|rover | rfolco: the f28 containers jobs has 82 hr at zuul queue | 10:14 |
| quiquell|rover | rfolco: https://review.openstack.org/636934 | 10:15 |
| rfolco | quiquell|rover, will abandon that change, I have no idea why | 10:15 |
| quiquell|rover | rfolco: ack | 10:16 |
| rfolco | quiquell|rover, done thanks | 10:16 |
| rfolco | zbr, as for https://review.openstack.org/#/c/636160, the ci job is defined in a separate change... I will merge into a single change | 10:20 |
| rfolco | marios, panda ^ | 10:20 |
| zbr | rfolco: yep, i do not see a reason to keep them separated, import it form my job, it avoids repetitio with yaml anchors. | 10:21 |
| marios | rfolco: ack i commented and revoted | 10:25 |
| rfolco | thx hope this gets in good shape today for merging | 10:25 |
| panda | zbr: rfolco wanna meet ? | 10:41 |
| rfolco | panda, up to you, I am making changes to the patch based on marios/zbr comments. | 10:42 |
| rfolco | panda, can we meet 1 hour before scrum ? | 10:42 |
| panda | rfolco: ok, the fact is taht I'm not sure about the interaction between all your patches now, so I wanted to understand what we need to review | 10:45 |
| rfolco | panda, ok, give me more few minutes, I can point you to the right sequence | 10:46 |
| *** holser_ is now known as holser|lunch | 10:48 | |
| panda | marios: sorry I'm lost with your topic, do you have reviews for me ? | 10:50 |
| marios | panda: which topic. https://review.rdoproject.org/r/18889 or https://review.openstack.org/#/c/635881 maybe | 10:51 |
| panda | marios: the first one, ok I see. That's weird. Anyway now that I had more time to spend on it, I have to ask something that may be missing. But if it's missing we'll need to address it probably next sprint | 10:53 |
| *** chem has quit IRC | 10:53 | |
| panda | marios: from which hash is that job building containers ? | 10:54 |
| *** chem has joined #oooq | 10:55 | |
| panda | rfolco: it can be in IRC if it's better for you | 10:56 |
| marios | panda: latest i think. or at least "whatever the upstream job does" | 10:57 |
| marios | panda: but i'm guessing that's wrong :) | 10:58 |
| marios | panda: and yeah we can add override into the job definition but will not make it for wednesday | 10:58 |
| marios | panda: but this is precisely why i want to see the thing run. thats what we discussed merge and debug till wednesday but quickly losing time | 10:59 |
| quiquell|rover | panda, marios: I am preparing a test on periodic + files | 11:01 |
| panda | marios: absolutely, agreee. We need to start from "tripleo-ci-testing" to build, but we'll deal with that in the next sprint, now focus on moving this forward. Unless of course building from a different hash it needs a complete refactor. | 11:03 |
| panda | quiquell|rover: waht kidn of test ? | 11:03 |
| quiquell|rover | panda: locally fake periodic pipeline and test the files job attribute and inheritance | 11:04 |
| quiquell|rover | panda: to see if we are affected by it | 11:04 |
| quiquell|rover | panda: so we don't have to merge unneeded stuff | 11:05 |
| quiquell|rover | panda: at config | 11:05 |
| panda | quiquell|rover: you're trying to adapt the reproducer to run periodic jobs ? Seems difficult especially to get the exact environment we need in rdoproject. | 11:06 |
| marios | panda: well we're currently using a fake tag for push | 11:06 |
| marios | panda: like tripleo-ci-testing | 11:06 |
| marios | err | 11:06 |
| marios | ci-squad-testing | 11:06 |
| rfolco | panda, marios zbr: so this is the change: https://review.openstack.org/#/c/636160/ | 11:07 |
| rfolco | zbr is helping with https://review.openstack.org/#/c/637212, which not a blocker... tripleo-repos works with --centos-mirror <fedora_mirror> | 11:08 |
| panda | rfolco: ok so zbr will converge in that change 636160 and that will be the main focus | 11:09 |
| marios | ack rfolco will check afain | 11:09 |
| marios | again | 11:09 |
| panda | marios: mmhh, that is the tafg used bu container after they are built | 11:10 |
| panda | marios: maybe the right hash is pulled directly from tripleo-repos, but we needt o tell it to fetch that particular hash, so it may not even somethign on your patches that needs to be changed. | 11:11 |
| panda | marios: focus on your task an thfour reviews at me, I'll explore that problem, and share the results. | 11:11 |
| quiquell|rover | panda: no, just exercise periodic + files | 11:13 |
| quiquell|rover | panda: I don'w want to test our periodic jobs there | 11:14 |
| marios | panda: ack which 4 reviews? | 11:14 |
| marios | panda: am looking briefly here too but gonna update it for the merge conflict/don't want to get stuck here. wrt my task, ideally i want to test the files: [] locally, it seems to be a perfect use-case for the reproducer quiquell|rover - not sure how i'll run the periodic yet but i'll ask you quiquell|rover once i get there. panda do you think its not even possible? | 11:16 |
| panda | marios: (reproducer) I don't know, but since it's outside of MVP, and still not in RC1 even, I would not spend too much on it. | 11:17 |
| marios | panda: well right now what else am gonna do? otherwise we need to merge https://review.rdoproject.org/r/18889 and hope it runs :) | 11:18 |
| marios | panda: k might spend a bit more on the pacemaker stuff then | 11:18 |
| * rfolco takes car to repair service brb | 11:18 | |
| marios | rfolco_work_from_shop | 11:18 |
| panda | marios: yes ... | 11:20 |
| panda | ok, diving into https://review.openstack.org/636160 now | 11:22 |
| zbr | rfolco: panda marios: adding --mirror to tripleo-repos ready for review: https://review.openstack.org/#/c/637212/ | 11:24 |
| panda | zbr: wait | 11:25 |
| zbr | also I think we need to talk on BJ about the cherry picking aproach, ok at xx:45 (in ~20min from now) | 11:26 |
| panda | zbr: how is this going to be used ? | 11:26 |
| panda | zbr: ok we can talk about that then | 11:26 |
| zbr | panda: https://review.openstack.org/#/c/636144/ is using it, running now. but change also. needs to be backward compatible as we don't want to break that. | 11:27 |
| quiquell|rover | panda, marios: files: [] does not work also - ".*" either | 11:33 |
| panda | so files is not the problem | 11:33 |
| quiquell|rover | panda: it is | 11:33 |
| quiquell|rover | panda: does not trigger if base has files at periodic | 11:33 |
| *** udesale has quit IRC | 11:33 | |
| quiquell|rover | panda: if I remove files it works | 11:34 |
| *** udesale has joined #oooq | 11:34 | |
| marios | quiquell|rover: :/ but the override fails? | 11:34 |
| quiquell|rover | marios: I see the value | 11:34 |
| quiquell|rover | marios: Could be that files get aggregated not overriden | 11:34 |
| quiquell|rover | going to try again so I don't fuck up | 11:35 |
| marios | quiquell|rover: not according to docs | 11:35 |
| marios | https://zuul-ci.org/docs/zuul/user/config.html#attr-job.files | 11:35 |
| quiquell|rover | marios: do you have time fo rblue ? | 11:35 |
| panda | quiquell|rover: where do we have files: specified ? | 11:36 |
| quiquell|rover | I want another pair of eyes | 11:36 |
| marios | quiquell|rover: sure | 11:36 |
| quiquell|rover | panda: blue ? | 11:36 |
| quiquell|rover | so we test it ? | 11:36 |
| panda | quiquell|rover: ok | 11:36 |
| quiquell|rover | I can be doing something wrong | 11:36 |
| marios | quiquell|rover: where | 11:36 |
| quiquell|rover | marios, panda: I am at gcerami | 11:36 |
| panda | chez me | 11:37 |
| marios | panda: number pls | 11:37 |
| quiquell|rover | just gcerami | 11:37 |
| quiquell|rover | marios: https://bluejeans.com/u/gcerami/ | 11:38 |
| panda | marios: 3492508669 | 11:38 |
| marios | quiquell|rover: thanks panda yah i meant the nmeric id so i don't have to look in source :) | 11:38 |
| *** jpena is now known as jpena|brb | 11:42 | |
| chkumar|ruck | quiquell|rover: once the tripleo common works, then we can test telemetry stuff | 11:52 |
| rfolco | zbr, thanks for helping with --mirror patch | 11:52 |
| rfolco | zbr, marios panda I'm ready to talk about f28 patch. When is good time? | 11:56 |
| zbr | join meeting above ^ and wait for them to finish current subject. | 11:56 |
| marios | rfolco: we are talking righ tnow about the periodic build contgainers job and issue running it https://review.rdoproject.org/r/#/c/18889/ | 11:57 |
| rfolco | oh | 11:57 |
| marios | rfolco: at 13:38 < quiquell|rover> marios: https://bluejeans.com/u/gcerami/ | 11:57 |
| marios | 13:38 < panda> marios: 3492508669 | 11:57 |
| marios | rfolco join us for radio quiquell|rover | 11:58 |
| zbr | btw, i managed to fix my BJ and now I can click normal URLs and opening the client, not sure if others still have similar issue. | 11:58 |
| marios | rfolco: we cant hear you | 12:01 |
| marios | rfolco: can you hear us? | 12:01 |
| rfolco | yes I hear you | 12:02 |
| rfolco | rejoining | 12:02 |
| rfolco | hell | 12:03 |
| marios | rfolco: lemme know if you wana jump on a blue but sort it out firts :) gonna comment on your review now | 12:03 |
| marios | rfolco: brb 2 mins | 12:03 |
| zbr | rfolco: marios panda we still need to talk on BJ about f28 work. | 12:04 |
| rfolco | marios, zbr panda can you pls join https://bluejeans.com/u/rfolco I hope my mic works | 12:05 |
| panda | zbr: rfolco ok let's review it live | 12:05 |
| marios | rfolco: ack gimme number please? | 12:05 |
| rfolco | https://www.google.com/url?q=https%3A%2F%2Fbluejeans.com%2F5878458097&sa=D&ust=1550662302494000&usg=AFQjCNFOk55G73RcR1s_4lauOylkXNqymA | 12:06 |
| rfolco | oops | 12:06 |
| rfolco | 5878458097 | 12:06 |
| rfolco | marios, | 12:06 |
| marios | rfolco: joining thx | 12:06 |
| *** apetrich has quit IRC | 12:11 | |
| *** skramaja_ has joined #oooq | 12:15 | |
| *** skramaja has quit IRC | 12:15 | |
| zbr | https://review.openstack.org/#/c/636144/ | 12:18 |
| *** apetrich has joined #oooq | 12:25 | |
| quiquell|rover | sshnaidm: commented https://review.openstack.org/#/c/636588 | 12:28 |
| sshnaidm | quiquell|rover, I mean not buildable projects like tripleo-ci, ovb repo, etc | 12:29 |
| sshnaidm | quiquell|rover, not every project can be built, but it doesn't mean we need to fail if it's in zuul changes | 12:29 |
| quiquell|rover | sshnaidm: humm | 12:30 |
| *** apetrich has quit IRC | 12:30 | |
| quiquell|rover | sshnaidm: we inject them at other ways | 12:30 |
| quiquell|rover | sshnaidm: that's right | 12:30 |
| quiquell|rover | sshnaidm: maybe we need a list for those | 12:30 |
| sshnaidm | quiquell|rover, yeah, we have this list, but not sure it's full | 12:30 |
| quiquell|rover | sshnaidm: there is no transparent way to do it | 12:30 |
| quiquell|rover | sshnaidm: skip ? | 12:30 |
| chkumar|ruck | quiquell|rover: sshnaidm https://github.com/openstack/tripleo-quickstart-extras/blob/master/roles/build-test-packages/defaults/main.yml#L15 is the list na? | 12:31 |
| sshnaidm | quiquell|rover, https://github.com/openstack/tripleo-quickstart-extras/blob/master/roles/build-test-packages/defaults/main.yml#L15 | 12:31 |
| quiquell|rover | sshnaidm: but better to have false negative if the list is not full | 12:31 |
| quiquell|rover | sshnaidm: than false positive | 12:31 |
| quiquell|rover | sshnaidm: we can change the message and say that if project is being used by job it has to be added to the skip list | 12:32 |
| quiquell|rover | sshnaidm: what do you think ? | 12:32 |
| sshnaidm | quiquell|rover, it started with problems with github packages, right? | 12:32 |
| quiquell|rover | sshnaidm: yep | 12:32 |
| sshnaidm | quiquell|rover, and we can't build them.. | 12:32 |
| quiquell|rover | sshnaidm: like people adding Depends-On the CI is green | 12:32 |
| quiquell|rover | sshnaidm: but not even the review itself is used | 12:33 |
| quiquell|rover | sshnaidm: nope, at least for now more though have to be done about them | 12:33 |
| sshnaidm | quiquell|rover, I see.. ok, then you're right, better to fail if project is not in the list and not buildable | 12:33 |
| quiquell|rover | sshnaidm: Going to change the error message and reference the skiplist | 12:34 |
| sshnaidm | quiquell|rover, but it requires a testing with different dummy patches for repos | 12:34 |
| quiquell|rover | sshnaidm: totally will do | 12:34 |
| chkumar|ruck | quiquell|rover: what if they adds a depends on rdo spec file in the ci? | 12:35 |
| quiquell|rover | chkumar|ruck: we do test that | 12:35 |
| quiquell|rover | chkumar|ruck: This is new feature I thing | 12:35 |
| quiquell|rover | sshnaidm: ^ ? | 12:35 |
| quiquell|rover | chkumar|ruck: also mapping is going to fail | 12:36 |
| quiquell|rover | chkumar|ruck: I will put a test for that too | 12:36 |
| chkumar|ruck | quiquell|rover: cool! | 12:36 |
| quiquell|rover | To check that we fail at all possibilities | 12:36 |
| sshnaidm | yeah, I think it was something about it.. maybe jpena|brb or ykarel know better | 12:36 |
| marios | rfolco: here https://review.openstack.org/#/c/635222/ https://review.openstack.org/#/c/635222/22/playbooks/tripleo-buildcontainers/templates/kolla-build.conf.j2 | 12:38 |
| marios | zbr: that was the 5 minutes ringing | 12:39 |
| marios | ;D | 12:39 |
| chkumar|ruck | quiquell|rover: do we need to make some exceptions there also while failing github projects for example someday if we want to ceph-ansible depends on then? | 12:40 |
| quiquell|rover | chkumar|ruck: Now that I think of we have mapping with puppet-rabbitmq, this review is not enough | 12:41 |
| ykarel | sshnaidm, quiquell|rover what's up? | 12:42 |
| quiquell|rover | chkumar|ruck: for ceph-ansible if we want we fix | 12:42 |
| chkumar|ruck | quiquell|rover: ack! | 12:42 |
| sshnaidm | ykarel, is it possible to test rdo spec file in CI? | 12:42 |
| quiquell|rover | chkumar|ruck: Do you know anything about this queen tempest issue ? http://logs.rdoproject.org/openstack-periodic-24hr/git.openstack.org/openstack-infra/tripleo-ci/master/periodic-tripleo-ci-centos-7-multinode-1ctlr-featureset017-queens/b273262/logs/tempest.html | 12:43 |
| ykarel | sshnaidm, u mean via depends-on on rdo spec in an upstream project? | 12:43 |
| marios | panda: the base change is this https://review.openstack.org/#/c/636160/25/zuul.d/build-containers.yaml and https://review.openstack.org/#/c/636188/26/zuul.d/layout.yaml (though second review will merge into the first) | 12:43 |
| marios | panda: fwiw/if it helps | 12:44 |
| sshnaidm | ykarel, yes | 12:44 |
| marios | panda: i.e. new base & then this stuff that folco is showing will run under the f28 one only | 12:44 |
| ykarel | sshnaidm, i remmeber mwhahaha added some support, let me find it | 12:44 |
| chkumar|ruck | quiquell|rover: it is the same error which is coming for rocky and master for telemetry tempest plugin | 12:44 |
| sshnaidm | ykarel, yeah, also think there was something added.. | 12:44 |
| quiquell|rover | chkumar|ruck: ack so we ignore it ? | 12:44 |
| quiquell|rover | chkumar|ruck: or we state somewhere that it affect queens too ? | 12:45 |
| chkumar|ruck | quiquell|rover: better to state it affects queens too | 12:45 |
| chkumar|ruck | quiquell|rover: but from today it is start showing na? | 12:45 |
| quiquell|rover | chkumar|ruck: Yep today I think | 12:45 |
| ykarel | sshnaidm, https://review.openstack.org/#/c/591865/ | 12:45 |
| chkumar|ruck | quiquell|rover: need to check what got changed there | 12:45 |
| chkumar|ruck | quiquell|rover: may be we can open a seperate bug for this | 12:46 |
| sshnaidm | ykarel, cool | 12:46 |
| *** holser|lunch is now known as holser_ | 12:46 | |
| sshnaidm | chkumar|ruck, quiquell|rover we have this for testing rdo spec files: https://review.openstack.org/#/c/591865/3/roles/build-test-packages/tasks/dlrn-build.yml | 12:47 |
| quiquell|rover | sshnaidm: ack | 12:47 |
| chkumar|ruck | ack | 12:47 |
| *** apetrich has joined #oooq | 12:48 | |
| arxcruz | chkumar|ruck: please take a look at https://tree.taiga.io/project/tripleo-ci-board/task/664?kanban-status=1447274 and update if you have any other information when you have time please | 12:49 |
| quiquell|rover | chkumar|ruck: open the bug if you see it's needed | 12:49 |
| chkumar|ruck | quiquell|rover: sure | 12:49 |
| *** bogdando has quit IRC | 12:50 | |
| *** bogdando has joined #oooq | 12:50 | |
| *** rlandy has joined #oooq | 12:54 | |
| marios | zbr: sorry i don't mean to be rude i have to go look at this parenting before scrum | 13:05 |
| marios | zbr: talk again in less than an hour ;) | 13:05 |
| *** trown|outtypewww is now known as trown | 13:06 | |
| zbr | panda: let's wf one of the changes, so we would have only one to cherry pick: https://review.openstack.org/#/c/629679/20 | 13:06 |
| zbr | once of them is not problematic | 13:06 |
| rfolco | panda, marios I do not need to change layout... templates include - tripleo-build-containers-jobs | 13:09 |
| rfolco | zbr, ^ | 13:09 |
| marios | ack rfolco | 13:16 |
| *** amoralej is now known as amoralej|lunch | 13:17 | |
| rfolco | marios, point me to the no files patch pls | 13:18 |
| rlandy | marios: well, if you hack haproxy enough, ... tripleo-ci-centos-7-scenario009-standaloneSUCCESS in 1h | 13:20 |
| rlandy | panda: ^^ | 13:20 |
| panda | rlandy: wow | 13:20 |
| panda | rlandy: and this was approved by one of the shiftonstack folks ? | 13:21 |
| rlandy | panda: it gets to be a bigger wow - will go through it in the meeting | 13:21 |
| panda | rlandy: so the task was not parked | 13:21 |
| panda | rlandy: ok | 13:21 |
| rlandy | panda: thetask was parked but I didn't give up on it | 13:21 |
| rlandy | it wasn't my choice to park it | 13:21 |
| trown | sshnaidm: never seen anyone talk on #openshift-dev so I am responding here :P | 13:23 |
| trown | sshnaidm: openshift-ansible is the only way to install openshift on openstack atm, but after the new installer[1] has support for openstack, you are correct that openshift-ansible wont be supported anymore [1]https://github.com/openshift/installer | 13:24 |
| trown | sshnaidm: at least that is my understanding... | 13:25 |
| quiquell|rover | trown: they are going to use coreos operators ? | 13:25 |
| trown | quiquell|rover: ya, we have a operator for openstack: https://github.com/openshift/cluster-api-provider-openstack | 13:26 |
| quiquell|rover | make sense | 13:26 |
| trown | or really a plugin for the cluster-api operator... | 13:27 |
| *** jpena|brb is now known as jpena | 13:27 | |
| marios | rfolco: ack working on it | 13:28 |
| *** quiquell|rover is now known as quique|rover|eat | 13:28 | |
| marios | rlandy: thanks will check in a bit my scen3 was also green (no pcs) the others failed 1/4 on the redis vip. i have to revisit but somethign else came up first | 13:28 |
| rlandy | marios: np - I have some info that may be helpful - will talk about it in meeting | 13:31 |
| apetrich | quique|rover|eat, I updated your patch to fix some pep8 errors and added more comments. | 13:41 |
| *** skramaja_ has quit IRC | 13:44 | |
| *** skramaja has joined #oooq | 13:45 | |
| sshnaidm | trown, ack, thanks | 13:47 |
| *** agopi has joined #oooq | 13:59 | |
| *** skramaja has quit IRC | 14:04 | |
| marios | rfolco: 15:53 < openstackgerrit> Marios Andreou proposed openstack-infra/tripleo-ci master: Add new base for the containers-build and periodic without files: https://review.openstack.org/637543 | 14:06 |
| marios | new change neededi just spoke about is here ^ | 14:07 |
| marios | and another one will follow | 14:07 |
| *** quique|rover|eat is now known as quiquell|rover | 14:09 | |
| quiquell|rover | apetrich: thanks! | 14:09 |
| apetrich | no worries | 14:09 |
| zbr | panda: afaik, use of yaml constructs to avoid duplication in zuul is no-risk because that's if 100% implemented by yaml parser at load time, zuul has no idea about what you put inside, it just gets the expanded dictionary. | 14:13 |
| *** ykarel is now known as ykarel|away | 14:14 | |
| *** amoralej|lunch is now known as amoralej | 14:14 | |
| *** ykarel|away is now known as ykarel | 14:15 | |
| *** vinaykns has joined #oooq | 14:24 | |
| quiquell|rover | panda: Now that I am thinking about periodics | 14:26 |
| quiquell|rover | panda: why do we need hubbot rebasing if we have periodics ? | 14:27 |
| quiquell|rover | panda: we can implement noop with periodics too | 14:27 |
| chkumar|ruck | quiquell|rover: timed to head home | 14:28 |
| chkumar|ruck | see ya tomorrow | 14:28 |
| *** chkumar|ruck is now known as chandankumar | 14:28 | |
| quiquell|rover | chandankumar: bye, read you tomorrow | 14:29 |
| panda | quiquell|rover: test on the rebase are launched in ci.centos, and we don't have periodic there | 14:31 |
| quiquell|rover | panda: ack | 14:32 |
| panda | zbr: yes, but it's the first time we are using them, and all your patch is doing is replicating all the contents of the parent, and changing a couple of keys, which is exaclty what zuul does in the end. | 14:32 |
| zbr | panda: not the parent, the official job (centos-7). anyway, i am open for any approach as long we do not endup duplicating the list of file-patterns. | 14:34 |
| zbr | i do not love yaml expansion myself, i just hate duplication more ;) | 14:35 |
| panda | marios: another comment for your patch. We have all the base job in the same file | 14:36 |
| panda | marios: don't know if you're comfortable in doing so too | 14:36 |
| panda | zbr: that's more reasonable and I think we can accomplish that with a single anchor containing the list of files for both. | 14:38 |
| zbr | panda: i am adraid that not because yaml does anchors work with dicitonaries and not with lists. i think i tried this first. | 14:39 |
| zbr | i may be wrong | 14:39 |
| marios | panda: will have a look momentarily | 14:39 |
| quiquell|rover | sshnaidm: added test reviews as you commented https://review.openstack.org/636588 | 14:41 |
| *** ykarel is now known as ykarel|away | 14:42 | |
| *** ykarel|away has quit IRC | 14:52 | |
| *** dougbtv has joined #oooq | 14:53 | |
| arxcruz | panda: thanks for the review, I'll fix those :) | 14:57 |
| marios | panda: i don't see a new comment in https://review.openstack.org/637543 did you forget to post it | 15:04 |
| marios | panda: i am replying to your and zbr other comments now thanks and will update | 15:04 |
| marios | panda: please add your comment and i'll do in the update | 15:04 |
| panda | marios: added | 15:05 |
| marios | panda: k not sold completely ont he base thing | 15:05 |
| marios | panda: i mean is nice that all the container build stuff is in one file | 15:06 |
| marios | panda:and all the standalone stuff in another file? | 15:06 |
| marios | panda: instead of also base.yaml | 15:06 |
| panda | marios: it's also nice that we have all the bases in one file :) | 15:06 |
| marios | panda: i guess its a matter of opinion;) | 15:06 |
| panda | marios: trade offs, not a -1 from me | 15:06 |
| quiquell|rover | panda: Can I workflow this ? https://review.openstack.org/#/c/634725 | 15:06 |
| quiquell|rover | panda: I see the job passin gthere | 15:06 |
| marios | panda: ack lemme see what it lookslike and will update we can change if people feel strongly enough to -1 for a particular way | 15:07 |
| quiquell|rover | panda: is the scenario012 new job | 15:07 |
| panda | quiquell|rover: I wanted Derek to confirm it's legit success | 15:08 |
| panda | quiquell|rover: I'll +1 it, we arelady have quorum | 15:09 |
| panda | quiquell|rover: bot only after I see Derek commment positively | 15:09 |
| quiquell|rover | panda: ack | 15:09 |
| *** quiquell|rover is now known as quiquell|off | 15:11 | |
| jfrancoa | panda: marios: as the ruck and rover are not online, just a heads up https://bugs.launchpad.net/tripleo/+bug/1816453 . The standalone upgrade job is failing, I've opened a lp for it and I'll investigate where the problem is | 15:21 |
| openstack | Launchpad bug 1816453 in tripleo "Standalone upgrade job failing in task [Extra migration for nova tripleo/+bug/1656791]" [Medium,New] - Assigned to Jose Luis Franco (jfrancoa) | 15:21 |
| marios | jfrancoa: thanks | 15:22 |
| panda | jfrancoa: are you afraid of quiquell|off ? :) You waited for him to go offline. Behold! The Wrath of the rover! | 15:23 |
| jfrancoa | ahahha, it was just a coincidence, I saw it now...but even though yes, he's a bit scary...too much hair for me :-D | 15:27 |
| zbr | just got out of bj session with alex, he will try to merge the simple one and for the kolla one we can try current approach with one note: we may update it to run on rdo instead of upstream (if needed...) | 15:28 |
| marios | zbr: thanks for doing that call | 15:33 |
| marios | zbr: do you mean current approach just cherrypick/checkout as rfolco is doing now | 15:34 |
| marios | zbr: or do you mean fork as we discussed earlier | 15:34 |
| zbr | no fork | 15:34 |
| zbr | no fork, no branch | 15:34 |
| marios | fork me . k so we'll just ask for forgiveness later if there are objections i guess :D | 15:35 |
| marios | zbr: rfolco panda ^ | 15:35 |
| zbr | i already integrated most of my changes to folco, now waiting for test results. also rebased change. | 15:35 |
| zbr | i will ping you once I get the of from zuul | 15:36 |
| *** udesale has quit IRC | 15:41 | |
| sshnaidm | trown, and where are you talking about openshift? :) | 15:42 |
| trown | sshnaidm: for openshift on openstack vms... downstream #shiftstack channel | 15:43 |
| *** sanjayu__ has quit IRC | 15:44 | |
| rfolco | zbr, we may update to rdo or we should ? | 15:49 |
| *** ykarel|away has joined #oooq | 15:50 | |
| *** ykarel|away is now known as ykarel | 15:51 | |
| rfolco | zbr, will rebase https://review.openstack.org/#/c/636160/ on https://review.openstack.org/#/c/637543/, but will wait for zuul job to run first | 16:03 |
| rfolco | marios, panda fyi ^ | 16:03 |
| marios | rfolco: ack thanks | 16:07 |
| *** ykarel is now known as ykarel|away | 16:13 | |
| rlandy | rfolco; marios: of the cards in Ready to Review, afaict, only review.openstack.org/636160 is outstanding, correct? | 16:18 |
| rfolco | rlandy, this is the main change, yes. There a few ones around it. | 16:19 |
| marios | rlandy: yeah and rfolco will reparent that onto my new base https://review.openstack.org/637543 | 16:20 |
| rfolco | rlandy, I'm still waiting on zbr to tell me what are the kolla/tripleo-common patches that will be used | 16:20 |
| rlandy | rfolco: k - ping when ready for review | 16:20 |
| rfolco | rlandy, thanks | 16:21 |
| zbr | rfolco: rlandy : this needs to merge https://review.openstack.org/#/c/629679/ -- i removed depends-on from. | 16:27 |
| panda | arxcruz: comments on https://review.rdoproject.org/r/18795 | 16:47 |
| panda | any last review before I leave for the day ? | 16:47 |
| arxcruz | panda: no, thanks | 16:47 |
| arxcruz | panda: i'll reply tomorrow, i'm already late for my german class | 16:48 |
| rlandy | zbr: I don;t have +2 in tripleo-common | 16:49 |
| zbr | rlandy: once i get it verified from zuul i will ping those that do have. it should be very easy as it previously already had 2x+2 | 16:50 |
| *** panda is now known as panda|off | 16:58 | |
| *** chem has quit IRC | 17:14 | |
| *** jfrancoa has quit IRC | 17:16 | |
| *** bogdando has quit IRC | 17:22 | |
| zbr | marios: please check https://review.openstack.org/#/c/637543/3 now -- if I would have -1 on wofkflow i would block it. affected job is not even triggered by the change. | 17:30 |
| zbr | marios: anyway, I did incorporate the idea from that change into https://review.openstack.org/#/c/636160/ -- so we should be ok. | 17:45 |
| *** kopecmartin is now known as kopecmartin|off | 17:56 | |
| *** trown is now known as trown|lunch | 18:01 | |
| *** derekh has quit IRC | 18:02 | |
| chandankumar | sshnaidm: Hello | 18:06 |
| chandankumar | sshnaidm: is there a way to enable ara from tripleo-modify-image role? | 18:06 |
| chandankumar | *for | 18:06 |
| sshnaidm | chandankumar, iirc it's difficult because of parallel queries to sqlite from ara, this roles runs multiple times in parallel | 18:07 |
| *** amoralej is now known as amoralej|off | 18:07 | |
| sshnaidm | chandankumar, for enabling it there we need to use a different database | 18:08 |
| chandankumar | sshnaidm: what about trying these stuff https://github.com/openstack/ara-server ? | 18:08 |
| *** jpena is now known as jpena|off | 18:09 | |
| * chandankumar takes a look for different db | 18:09 | |
| sshnaidm | chandankumar, ara-server looks like nice solution too | 18:09 |
| chandankumar | sshnaidm: one server then having client to fetch the data for each node | 18:10 |
| chandankumar | sshnaidm: I will create an epic for that | 18:10 |
| sshnaidm | chandankumar, cool | 18:10 |
| *** ykarel|away has quit IRC | 18:16 | |
| *** ykarel|away has joined #oooq | 18:17 | |
| *** ykarel|away has quit IRC | 18:22 | |
| *** holser_ has quit IRC | 18:38 | |
| *** dtantsur is now known as dtantsur|afk | 18:47 | |
| *** dsneddon has joined #oooq | 18:59 | |
| *** trown|lunch is now known as trown | 19:06 | |
| rfolco | zbr, around? | 19:23 |
| zbr | rfolco: yep. still working on that change. | 19:23 |
| rlandy | rfolco: how did it go with the reproducer? | 19:23 |
| rfolco | so you saw the tripleo-repo error ? | 19:24 |
| zbr | rfolco: https://review.openstack.org/#/c/637608/ is supposed to fix it. | 19:24 |
| rfolco | rlandy, on hold for now... will resume asap | 19:24 |
| zbr | another example of what happens when we do not add unittesting around our modifications. | 19:25 |
| rfolco | zbr, before you submit can I please rebase on top of marios change ? | 19:26 |
| zbr | rfolco: no, because I already included that change. | 19:27 |
| zbr | rfolco: also no need to worry that it will merge too soon, already has two unmerged depends-on... ;) | 19:27 |
| rfolco | zbr, ok then, can we comment other jobs until we get a green run, then we just re-add overnight ? | 19:28 |
| rfolco | so we speed up test | 19:28 |
| zbr | rfolco: smart move! can you do it? | 19:28 |
| rlandy | panda|off: ping if you become panda on again | 19:29 |
| rfolco | zbr, sure, you want me to do with this patchset ? | 19:29 |
| rfolco | rlandy, to turn panda|off on, you have to do this panda|off panda|off panda|off | 19:30 |
| * rfolco hides | 19:30 | |
| * rlandy needs to switch time zones | 19:30 | |
| zbr | rfolco: lets disable other jobs only if this run fails, most jobs already started so we should not loose our position in queue. | 19:32 |
| rfolco | zbr, ok | 19:33 |
| zbr | while reading discussions about sprints and productivity I realised that we may have a much better way to improve productivity: start shipping beer to people doing reviews. what if people, at the end of week, would receive an amount of beer direct proportional with their reviewing contribution? | 19:37 |
| zbr | probably we need to put a cap on this as we may endup with one or two that will never be sober ;) | 19:37 |
| rfolco | zbr, if any other non-blocking error, please ignore and remove from the patch (like tripleo-repos) | 19:41 |
| rfolco | zbr, let's go straight to the point, we can fix tripleo-repos later | 19:42 |
| zbr | rfolco: no, you cannot. this bug is blocker. | 19:42 |
| rfolco | zbr, probably because you auto-detect, if you select -d fedora, it is not blocker | 19:43 |
| zbr | rfolco: look at last zuul execution on fedora job, this is how i found it. thus adding it with depends-on | 19:43 |
| zbr | rfolco: last version of the patch no longer has the --distro parma. but no need to worry, this patch will be merged quickly on tripleo-repos. is has very few tests on it. | 19:43 |
| rfolco | Error: 'don't worry' does not exist in your dict. | 19:44 |
| rfolco | zbr, ^ | 19:44 |
| rfolco | rlandy, reproducer stuck... openstacksdk is required for this module | 19:45 |
| rfolco | clean centos7 vm in rdocloud | 19:45 |
| zbr | i think we may need to add a trick to lower the use of fedora-28 nodes: make it a child of centos-7, so it will only run after the other one, if the other one passes. it will not make the final execution longer because there are plenty of other jobs that take longer than sum of these two. | 19:54 |
| rlandy | rfolco: what are you running on? | 19:57 |
| rlandy | trown: hi ... I'd like to test this out on the baremetal hardware we have ... https://review.openstack.org/#/c/632052/2/doc/source/install/advanced_deployment/deploy_openshift.rst | 20:04 |
| rlandy | trown: do you have a ready defined test I could use - what's the assumed starting point here? | 20:06 |
| rfolco | rlandy, in a clean centos vm, created cloud.yaml and then ran reproducer-zuul-based-quickstart.sh | 20:18 |
| rfolco | rlandy, (sorry for the delay, somebody knocked my door) | 20:19 |
| rlandy | ugh - weshay did this centos testing - we should have fixed that | 20:19 |
| * rlandy checks | 20:19 | |
| rlandy | https://github.com/rdo-infra/ansible-role-tripleo-ci-reproducer/blob/master/requirements.txt | 20:20 |
| rlandy | rfolco: can you tell me where it stopped | 20:20 |
| trown | rlandy: there is no automated tests for baremetal openshift ... the docs you pointed to should be correct though | 20:20 |
| trown | rlandy: starting point is an undercloud as far as I can tell | 20:20 |
| rfolco | rlandy, ASK [ansible-role-tripleo-ci-reproducer : Open needed ports at default security group] | 20:21 |
| rlandy | trown: k, so as long as I can get by undercloud install, we should be good to go | 20:21 |
| rfolco | rlandy, will start over | 20:21 |
| rlandy | rfolco:k - I also ran into that - but I had openstacksdk installed via rpm | 20:21 |
| rlandy | try install it via pip using --user | 20:22 |
| rfolco | ok, do you run any requirements.txt before starting ? r | 20:23 |
| rfolco | ok, do you run any requirements.txt before starting ? rlandy | 20:23 |
| rlandy | rfolco: no | 20:23 |
| rlandy | you should not need to run anything | 20:24 |
| rlandy | but to check try the install vai pip with --user | 20:24 |
| rfolco | rlandy, ok thanks | 20:24 |
| *** agopi has quit IRC | 20:37 | |
| panda|off | somebody rang ? | 20:47 |
| rlandy | panda|off: I did | 20:50 |
| rlandy | panda|off: but I can't really remember why :) | 20:51 |
| rlandy | I think to ask if you were planning to keep the pacemaker stuff in scenario012 | 20:51 |
| panda|off | rlandy: for now yes. Then it depends on derek, we'll probably discuss optimization before migrating test to standalone. | 20:53 |
| rlandy | panda|off: ack | 20:54 |
| panda|off | rfolco: last patchset in https://review.openstack.org/636160 is the good one ? | 21:05 |
| rfolco | panda|off, waiting for zuul to tell if it is good, previous one was bad, zbr fixed with a depends on | 21:06 |
| panda|off | :/ I wanna review stuff. Tomorrow then. | 21:15 |
| rlandy | trown: would I use a CNS or non-CNS deployment? | 21:29 |
| *** jtomasek has quit IRC | 21:40 | |
| trown | rlandy: probably non-CNS is less to go wrong | 21:54 |
| rlandy | k - I'm good with less going wrong | 21:55 |
| *** trown is now known as trown|outtypewww | 22:02 | |
| *** rascasoft has quit IRC | 23:06 | |
| *** rascasoft has joined #oooq | 23:44 | |
| *** agopi has joined #oooq | 23:47 | |
| *** rascasoft has quit IRC | 23:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!