| *** tosky has quit IRC | 00:04 | |
| *** derekh_afk has quit IRC | 00:20 | |
| *** saneax has quit IRC | 00:40 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone @ https://review.openstack.org/604298, stable/pike: tripleo-ci-centos-7-scenario002-multinode-oooq-container @ https://review.openstack.org/602248, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone, tripleo-ci-centos-7-scenario007-multinode-oooq- (1 more message) | 00:49 |
|---|---|---|
| rlandy | weshay: you still around? | 01:03 |
| rlandy | weshay: if so, at what point do we expect the zuul change to be applied if the change is mode to tripleo-quickstart/tripleo-ci? | 01:04 |
| rlandy | not dlrn | 01:04 |
| *** rlandy is now known as rlandy|biab | 01:09 | |
| weshay | rlandy|biab, I think the ci repos need to updated w/ the zuul change right after invoking the script | 01:44 |
| *** weshay is now known as weshay_pto | 01:44 | |
| *** chem` has joined #oooq | 02:33 | |
| *** chem has quit IRC | 02:37 | |
| *** rlandy|biab is now known as rlandy | 02:45 | |
| *** rlandy has quit IRC | 02:45 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone @ https://review.openstack.org/604298, stable/pike: tripleo-ci-centos-7-scenario002-multinode-oooq-container @ https://review.openstack.org/602248, master: tripleo-quickstart-extras-gate-master-delorean-full-featureset052, tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci- (1 more message) | 02:49 |
| *** ykarel|away has joined #oooq | 03:04 | |
| *** udesale has joined #oooq | 03:51 | |
| *** chem` has quit IRC | 04:17 | |
| *** ykarel|away is now known as ykarel | 04:32 | |
| *** chandankumar has joined #oooq | 04:33 | |
| *** chandankumar is now known as chkumar|ruck | 04:36 | |
| *** ykarel has quit IRC | 04:37 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone, tripleo-ci-centos-7-scenario007-multinode-oooq-container @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-quickstart-extras-gate-master-delorean-full-featureset052, (1 more message) | 04:49 |
| *** ykarel has joined #oooq | 04:53 | |
| *** ratailor has joined #oooq | 05:22 | |
| *** saneax has joined #oooq | 05:41 | |
| *** jfrancoa has joined #oooq | 06:20 | |
| *** jaganathan has quit IRC | 06:30 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone, tripleo-ci-centos-7-scenario007-multinode-oooq-container @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-quickstart-extras-gate-master-delorean-full-featureset052, (1 more message) | 06:49 |
| Tengu | panda|rover|off: it was https://review.openstack.org/#/c/618092/ | 06:51 |
| Tengu | already merged. | 06:51 |
| *** apetrich has joined #oooq | 07:03 | |
| *** quiquell|off is now known as quiquell | 07:09 | |
| *** saneax has quit IRC | 07:21 | |
| *** jtomasek has joined #oooq | 07:27 | |
| *** ykarel is now known as ykarel|lunch | 07:58 | |
| quiquell | marios: How can I help with the 001 thing to standalone ? | 08:05 |
| quiquell | marios: good morning btw | 08:05 |
| *** jbadiapa has quit IRC | 08:07 | |
| chkumar|ruck | quiquell, marios \o/ | 08:09 |
| quiquell | chkumar|ruck: o/ | 08:12 |
| *** bogdando has joined #oooq | 08:18 | |
| *** amoralej|off is now known as amoralej | 08:20 | |
| *** holser_ has joined #oooq | 08:26 | |
| marios | quiquell: o/ morning | 08:35 |
| marios | quiquell: not really much at the moment for that - alex/wes were discussing last night about the override for featureset_overide here https://review.openstack.org/616872 | 08:35 |
| marios | quiquell: i will update that there is a nit as folco pointed out but after this just a small matter of making the job green :) | 08:36 |
| marios | quiquell: so last run was like http://logs.openstack.org/72/616872/26/check/tripleo-ci-centos-7-scenario001-standalone/e2a851b/logs/undercloud/home/zuul/standalone_deploy.log.txt.gz#_2018-11-15_22_46_48 | 08:36 |
| quiquell | marios: ack, | 08:36 |
| marios | quiquell: if it settles today wrt the approach (looks like we are mostly agreeing now, this featureset override was the last detail i think for now) then we can just split the others | 08:37 |
| marios | quiquell: so you can take scen2 for example | 08:37 |
| marios | quiquell: so maybe that is something to do too i mean start looking at the other scenarios we have 4 for this sprint but lets see how many we do :) | 08:38 |
| quiquell | marios: Will take scenario002 and will try to replicate to 0001 stuff | 08:39 |
| quiquell | marios: It's that ok ? | 08:39 |
| marios | quiquell: right sounds good man | 08:39 |
| quiquell | marios: So I have to replicate first the 002 environment yaml I suppose first | 08:39 |
| quiquell | marios: then the featureset | 08:39 |
| quiquell | marios: then the job | 08:39 |
| marios | quiquell: well no featureset | 08:39 |
| marios | quiquell: that was the point of featureset_override | 08:39 |
| marios | quiquell: so we don't need it sec | 08:39 |
| quiquell | marios: Ahh cool cool | 08:39 |
| quiquell | marios: So we use directly the 052 featureset with featureset_override in the job | 08:40 |
| marios | https://tree.taiga.io/project/tripleo-ci-board/task/363?kanban-status=1447278 | 08:40 |
| marios | quiquell: read here i explained a bit in the description | 08:40 |
| quiquell | marios: ack | 08:40 |
| marios | quiquell: right chkumar|ruck++ idea | 08:41 |
| hubbot1 | marios: chkumar|ruck's karma is now 3 | 08:41 |
| marios | very good one | 08:41 |
| quiquell | marios: You have remember me that I have to talk with upgrade guys about reusing featurests | 08:41 |
| marios | though it caused quite a bit of discussion in the end we convinced weshay_pto :) | 08:41 |
| quiquell | marios: chkumar|ruck++ already tested at puppet projects | 08:41 |
| hubbot1 | quiquell: chkumar|ruck's karma is now 4 | 08:41 |
| quiquell | marios: That's difficult :-) | 08:41 |
| chkumar|ruck | quiquell, marios thanks :-) | 08:42 |
| chkumar|ruck | quiquell, marios https://docs.openstack.org/tripleo-docs/latest/contributor/check_gates.html#featureset-override | 08:42 |
| chkumar|ruck | advertized here | 08:42 |
| chkumar|ruck | we can expand the docs here for scenario job and list all the jobs at one place | 08:43 |
| marios | chkumar|ruck: yeah but trouble is weshay_pto doesn't like it as a general purpose thing (well in this case he didn't like the environment file override because others will use it and then we have problems if env files are maintained) | 08:43 |
| quiquell | marios: Humm, also you need the job yaml to run this, with quickstart I think you just pass featuresets and release file | 08:44 |
| quiquell | marios: I mean it means dependency to reproducer to run this | 08:44 |
| marios | chkumar|ruck: in the end the compromise was we don't pass the feature directly through the override but instead expose standalone_environment_files and then translate it i.e. like here https://review.openstack.org/#/c/616872/27/zuul.d/standalone-jobs.yaml where you add it to your job then we munge it here https://review.openstack.org/#/c/616872/27/playbooks/tripleo-ci/run-v3.yaml | 08:44 |
| marios | standalone_environment_files is a new ci specific thing which you can override and it gets passed to the deployment as standalone_custom_env_files | 08:45 |
| marios | quiquell: ^ | 08:45 |
| chkumar|ruck | marios, but there is a problem in the long run | 08:45 |
| chkumar|ruck | marios, we end up adding lots of vars there | 08:45 |
| marios | chkumar|ruck: right and its one of those "you had to be there" to know wtf it does but ... | 08:45 |
| * marios shrug | 08:46 | |
| marios | i don't know what the right way is. alex point was exactly that i.e. why do we keep adding vars and overriding things but weshay_pto point is also valid | 08:46 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone, tripleo-ci-centos-7-scenario007-multinode-oooq-container @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-quickstart-extras-gate-master-delorean-full-featureset052, (1 more message) | 08:49 |
| chkumar|ruck | marios, in devstack jobs, https://github.com/openstack-dev/devstack/blob/master/.zuul.yaml#L107 | 08:51 |
| chkumar|ruck | they include the roles and i think they just set it there | 08:52 |
| chkumar|ruck | as a vars | 08:52 |
| *** ykarel|lunch is now known as ykarel | 08:54 | |
| marios | chkumar|ruck: cool | 08:54 |
| chkumar|ruck | marios, I am not sure it will work in our case | 08:55 |
| chkumar|ruck | marios, as our roles are in TQE and TQ, may we once merge these two repo in TOCI, that will open few more posibilties | 08:55 |
| quiquell | marios: So we merge the two environment files with featurest_override option ? | 08:56 |
| quiquell | marios: Ahh no | 08:57 |
| marios | quiquell: well not merge. the idea is to pass the standalone_custom_env_files to the deployment. standalone_environment_files is just something we made up | 08:57 |
| *** tosky has joined #oooq | 08:57 | |
| *** saneax has joined #oooq | 08:57 | |
| marios | quiquell: this ismy v23 fro yesterday little different but maybe clearer as to the point here https://review.openstack.org/#/c/616872/21/playbooks/tripleo-ci/run-v3.yaml | 08:58 |
| marios | quiquell: i was just using regex/replace s/standalone_environment/standalone_custom_env_files | 08:58 |
| *** holser_ has quit IRC | 08:59 | |
| quiquell | marios: Too much noise to just have an absolute path | 08:59 |
| marios | quiquell: right i was adding the path too like https://review.openstack.org/#/c/616872/21/zuul.d/standalone-jobs.yaml | 09:01 |
| marios | quiquell: so in the job you just give the name | 09:01 |
| quiquell | marios: Prefer that, or even implement it as part of Standalone, like adding it | 09:02 |
| quiquell | marios: I think it's done in other featuresets that use thth | 09:02 |
| quiquell | tht | 09:02 |
| marios | quiquell: yeah actually that isn't about usability primarily but rather to resttrict it. so whichever env they want to specify HAS to live in usr/shar/tht/ci/environments | 09:02 |
| quiquell | marios: Also why we don't use standalone_custom_env_files directly as name in featureset_override ? | 09:03 |
| marios | quiquell: to restrict its use | 09:03 |
| quiquell | marios: Is uglier but explicit | 09:03 |
| *** panda|rover|off is now known as panda | 09:03 | |
| quiquell | marios: It's part of the disccussion I haven't read I suppose :-/ | 09:03 |
| quiquell | marios: btw found little nit that can break other jobs | 09:04 |
| *** chkumar|ruck has quit IRC | 09:04 | |
| quiquell | marios: Why do we need the restriction ? | 09:05 |
| quiquell | marios: Going to read a little before talk :-) | 09:06 |
| marios | quiquell: because this is not meant to be a general purpose 'override environment files in your jobs' thing but specifically a 'override the standalone env files, only for these scenario jobs' | 09:07 |
| marios | we don't want people using it | 09:07 |
| marios | quiquell: now it made more sense when we even assumed tht/ci/environment in the path and you only specify scenario1.yml BUT now | 09:08 |
| marios | quiquell: latest version is like this https://review.openstack.org/#/c/616872/27/zuul.d/standalone-jobs.yaml so we are basically allowing any env under tht, making it more general again | 09:08 |
| marios | quiquell: thus we may as well us the original name i think. i'll add that comment. but honestly i just want to make our minds up and go with it.. :/ | 09:08 |
| quiquell | marios: we review those, is not like the are going to break them | 09:09 |
| marios | too much discussion... | 09:09 |
| quiquell | marios: Yep sorry :-/ it's only that add more ad-hoc stuff to the CI workflow | 09:09 |
| quiquell | :-/ | 09:09 |
| marios | quiquell: yeah indeed. | 09:09 |
| marios | quiquell: as i said it made more sense before but if we are going to allow it unrestrricted may as well use tha original name gonna comment | 09:10 |
| quiquell | marios: Humm I understand it now, is not only about this job, but jobs people do, like it starts to scalate and peple add env file and weird paths and so on | 09:11 |
| quiquell | marios: And then CI is impossible to understand or follow | 09:11 |
| quiquell | marios: something like this ? | 09:11 |
| marios | quiquell: right | 09:11 |
| quiquell | marios: like you can have an empty featurest and put all the featureset thing in the env file and wrong stuff like that | 09:12 |
| marios | quiquell: so first of all restrict to be something under tht root at least and not random location/env file . but then the problem is the env file in tht might change/folkd don't maintain it then we have problem | 09:12 |
| quiquell | marios: we are the owner not only of featuresets but also featureset attribtues, that's right ? | 09:13 |
| marios | quiquell: commented here https://review.openstack.org/#/c/616872/27/zuul.d/standalone-jobs.yaml | 09:14 |
| *** chandankumar has joined #oooq | 09:15 | |
| quiquell | marios: understood | 09:16 |
| quiquell | marios: It woul dbe nice to have a featurest_override role, with all this logic | 09:16 |
| quiquell | marios: we are already restricing the attribute we override so from the beginning is not just a mapping, so feels like it's going to have more logic there | 09:17 |
| marios | quiquell: right we are changing it by restricting thus it becomes a new thing, it isn't the normal standalone_custom_env_files that we have in the standalone deployment. we are using a new variable to create that standalone_custom_env_files and pass it on | 09:18 |
| *** udesale has quit IRC | 09:18 | |
| quiquell | marios: thanks for the explanation, will add two task to 002, job and env file. | 09:20 |
| quiquell | marios: also the we will have to override tempest, isn't it ? | 09:20 |
| *** holser_ has joined #oooq | 09:20 | |
| chandankumar | quiquell, since for tempest, in future we will move to os_tempest so overrides value change or we can featureset_override role to do the translation | 09:21 |
| chandankumar | by mapping each tempest var to specific os_tempest var | 09:21 |
| chandankumar | this will also help not to change anything | 09:21 |
| quiquell | chandankumar: so tempest disappear from featureset and we just configure os_tempest ? | 09:22 |
| chandankumar | quiquell, nope | 09:22 |
| chandankumar | quiquell, temepst will be there just validate-tempest role will be removed and we can use os_tempest role https://github.com/openstack/openstack-ansible-os_tempest/ | 09:23 |
| chandankumar | os_tempest vars is different from validate_tempest var | 09:23 |
| chandankumar | since you are going to featureset_override role there I am thinking to map validate_tempest vars to os_tempest var | 09:23 |
| quiquell | chandankumar: Humm, the less mapping we have the better, that's good idea temporally but in feature the vars have to be os_tempest vars | 09:24 |
| chandankumar | yes | 09:24 |
| quiquell | chandankumar: But I suppose it's going to take time to have everything ported to os_tempest | 09:25 |
| chandankumar | quiquell, we can switch to os_tempest vars without breaking | 09:25 |
| chandankumar | quiquell, yes in future | 09:25 |
| chandankumar | future ~= 3-4 months | 09:25 |
| quiquell | chandankumar: ack, the featurest_override will be good place for those temporal mappings | 09:25 |
| quiquell | chandankumar: That's enough to maintain a mapping, and remove it later on | 09:25 |
| chandankumar | yes | 09:26 |
| *** chandankumar is now known as chkumar|ruck | 09:27 | |
| quiquell | ssbarnea|bkp2: You there ? | 09:28 |
| *** derekh has joined #oooq | 09:38 | |
| *** chem has joined #oooq | 09:39 | |
| *** panda is now known as panda|rover | 09:39 | |
| *** dtantsur|afk is now known as dtantsur | 09:41 | |
| *** agopi_ has joined #oooq | 09:51 | |
| ssbarnea|bkp2 | quiquell: hi! now I am. | 09:51 |
| *** jbadiapa has joined #oooq | 09:52 | |
| quiquell | ssbarnea|bkp2: Are we going to activate tempest for f28 job ? | 09:53 |
| quiquell | panda|rover: ^? | 09:53 |
| *** rascasoft has quit IRC | 09:53 | |
| quiquell | ssbarnea|bkp2: Ahh I see we are ok | 09:53 |
| *** agopi|brb has quit IRC | 09:53 | |
| ssbarnea|bkp2 | quiquell: yes. when exactly, I do not know exactly with which settings but I would assume the final goal is to match the centos one. | 09:54 |
| quiquell | ssbarnea|bkp2: ack, about the paunch thing, looks like it's passing without it, it's using python2 ? | 09:54 |
| quiquell | ssbarnea|bkp2: Do we have a review here ? https://tree.taiga.io/project/tripleo-ci-board/task/377?kanban-status=1447276 | 09:59 |
| ssbarnea|bkp2 | let me get back to it, i have review | 09:59 |
| chkumar|ruck | ssbarnea|bkp2, let me and arxcruz|summit and mkopec, know if you need any help on f28 tempest | 09:59 |
| ssbarnea|bkp2 | https://review.openstack.org/#/c/617338/ -- lets assure this merges asap. | 10:00 |
| quiquell | ssbarnea|bkp2: Alex say something about paunch not being used as module | 10:01 |
| quiquell | ssbarnea|bkp2: Also the error is not because of this | 10:02 |
| chkumar|ruck | ssbarnea|bkp2, please have a look at https://review.openstack.org/#/c/618163/ | 10:02 |
| ssbarnea|bkp2 | obviously that it was not used as module because it was not possible before that patch. | 10:02 |
| quiquell | ssbarnea|bkp2: I mean | 10:02 |
| quiquell | ssbarnea|bkp2: Ahh wait I remember we have a review that calls paunch-3 | 10:03 |
| quiquell | ssbarnea|bkp2: That's why is working but you don't like suffixes | 10:03 |
| ssbarnea|bkp2 | i don't like that approach but they are not conflicting. | 10:04 |
| quiquell | ssbarnea|bkp2: So this is more of good to have | 10:04 |
| quiquell | ssbarnea|bkp2: Using "python3" is using suffix too | 10:04 |
| ssbarnea|bkp2 | well, all good tools have it. even ansible was close to having it but i got stuck due to the multiple CLI commands debate. | 10:05 |
| *** holser_ has quit IRC | 10:05 | |
| quiquell | puff what a pain... | 10:05 |
| ssbarnea|bkp2 | yes, but is one simple pattern reusable everywhere, without extra conditions. | 10:05 |
| quiquell | At pure python3 stuff we will use directly the commands ? | 10:05 |
| quiquell | And expect they don't break stuff with python4 :-) | 10:06 |
| *** udesale has joined #oooq | 10:10 | |
| *** dtantsur is now known as dtantsur|afk | 10:10 | |
| ssbarnea|bkp2 | panda|rover: i think you can help with https://review.openstack.org/#/c/617338/ | 10:13 |
| ssbarnea|bkp2 | quiquell: you may not have read the PEP related to command names, the plan is to go back to "python" when py2 will be out. | 10:13 |
| quiquell | ssbarnea|bkp2: Yep that what I read, so we will have to replace stuff around again | 10:13 |
| quiquell | ssbarnea|bkp2: python3 command will still be present ? | 10:13 |
| ssbarnea|bkp2 | i have some reasons why I push for the python_cmd approach.... is more future proof. the current "-3" suffix is more of future-poof. | 10:14 |
| ssbarnea|bkp2 | yep | 10:14 |
| quiquell | ssbarnea|bkp2: ack, thanks man, let's try to merge that | 10:15 |
| ssbarnea|bkp2 | thus is why I am trying to use module calling, when possible. | 10:15 |
| quiquell | ssbarnea|bkp2: But we have to ensure that commind calling is working too | 10:16 |
| marios | panda|rover: o/ | 10:16 |
| quiquell | ssbarnea|bkp2: I think that's Alex point | 10:16 |
| marios | wana talk? | 10:16 |
| marios | panda|rover: or would you prefer to do it at 1900 when i'm leaving ;() | 10:17 |
| marios | panda|rover: :) | 10:17 |
| *** udesale has quit IRC | 10:21 | |
| *** holser_ has joined #oooq | 10:28 | |
| chkumar|ruck | quiquell, panda|rover How can I pull the list of failed nodes and stacks which are not able to delete by script? | 10:37 |
| quiquell | chkumar|ruck: | 10:37 |
| quiquell | chkumar|ruck: You need the nodepool access and you can do a openstack stack list | 10:38 |
| quiquell | chkumar|ruck: There you will see DELETE_FAILED or similar | 10:38 |
| chkumar|ruck | quiquell, which I donot have | 10:38 |
| chkumar|ruck | access to nodepool | 10:39 |
| panda|rover | chkumar|ruck: you don't have nodepool tenant password ? | 10:39 |
| *** saneax has quit IRC | 10:39 | |
| chkumar|ruck | panda|rover, nope | 10:39 |
| quiquell | chkumar|ruck: priv | 10:39 |
| chkumar|ruck | panda|rover, last time it happened, quiquell done it | 10:39 |
| panda|rover | quiquell: priv on internal IRC please ... | 10:39 |
| panda|rover | quiquell: or use dcc chat | 10:40 |
| quiquell | panda|rover: ack | 10:40 |
| quiquell | chkumar|ruck: check email | 10:41 |
| *** saneax has joined #oooq | 10:41 | |
| chkumar|ruck | quiquell, not received yet | 10:42 |
| chkumar|ruck | panda|rover, pleas join #rhos-ops internally | 10:46 |
| chkumar|ruck | e*please | 10:46 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-fedora-28-standalone, tripleo-ci-centos-7-scenario007-multinode-oooq-container @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-fedora-28-standalone @ https://review.openstack.org/560445, (1 more message) | 10:49 |
| ssbarnea|bkp2 | tosky raised a good question regarding linting hardening: that it may make sense to attempt to make the linting fixing in separated change(s), to avoid polluting our changes. | 10:51 |
| ssbarnea|bkp2 | with the risk of alienating few, please check https://review.openstack.org/#/c/618306/ - panda|rover chkumar|ruck quiquell | 10:57 |
| panda|rover | ssbarnea|bkp2: can you make a bigger change ? | 11:02 |
| panda|rover | ssbarnea|bkp2: can you put some fixes for tripleo-heat-templates, puppet-tripleo, and linux kernel too please ? :) | 11:02 |
| ssbarnea|bkp2 | panda|rover: sure, just let me finish my cloning machine. | 11:03 |
| ssbarnea|bkp2 | the reality is that this once could be split in two, one part for flake8 and one for ansible-lint, but i think it does not work the extra load to do that. | 11:04 |
| quiquell | ssbarnea|bkp2 thinking agina about the paunch thing, maybe we can just change the RPM to make paunch command the python3 version, I know this is not correct, but the future is going to be like this | 11:14 |
| quiquell | ssbarnea|bkp2: I mean future is calling just 'command' | 11:15 |
| ssbarnea|bkp2 | i am not so sure able that, we will always have some kind of duality. when py2 will be dead, we will have py4, or some other py* that breaks things and we will have to run things in parallel. based on the extra roundtrips for making RPM, I am *sure* that we will face the need to install from git or wheel packages,... where we do *not* have the symlinking stuff. | 11:17 |
| chkumar|ruck | panda|rover, i have emailed the list of server/nodes to delete | 11:18 |
| ssbarnea|bkp2 | also module calling method is addressing *other* problems: ability to install the tool at user level and use it, without needing root. | 11:18 |
| quiquell | ssbarnea|bkp2: Didn't PEP say that future is using "command" and that we are not going to suffer the same again ? | 11:18 |
| quiquell | ssbarnea|bkp2: But this is more of how thinks are designed, looks like paunch is desgined to be a command (we can replace python with another language and it shoudl work | 11:19 |
| ssbarnea|bkp2 | in fact PSF has something baking for everyone is called a python launcher if i remember well, and guess what is another.... python_cmd tool. | 11:19 |
| panda|rover | chkumar|ruck: I blocked te-broker from destroying overcloud nodes, | 11:21 |
| panda|rover | chkumar|ruck: we are doing debugging with kieran | 11:21 |
| chkumar|ruck | panda|rover, ok | 11:21 |
| panda|rover | chkumar|ruck: some stacks will need to be manually destryed | 11:22 |
| quiquell | chkumar|ruck, panda|rover: Contact first with kforde | 11:22 |
| quiquell | He deal with that stuff | 11:22 |
| chkumar|ruck | panda|rover, I have dropped all info in email you are in cc | 11:23 |
| chkumar|ruck | as per kforde there is some Db needs adjusting | 11:23 |
| chkumar|ruck | oh not again vexhost deployment failed | 11:24 |
| panda|rover | quiquell: ?? | 11:24 |
| panda|rover | quiquell: what stuff ? | 11:24 |
| quiquell | panda|rover: deleting nodepool stacks that get stuck | 11:24 |
| chkumar|ruck | brb | 11:25 |
| panda|rover | quiquell: ok, don't touch the old stacks for now, I'can't select what te-broker is not going to destroy, so I modified destroy-env for now, until we're done | 11:25 |
| *** holser_ is now known as holser|lunch | 11:35 | |
| *** dtantsur|afk is now known as dtantsur | 11:37 | |
| *** quiquell is now known as quiquell|brb | 11:53 | |
| chkumar|ruck | panda|rover, can we sync? | 12:10 |
| panda|rover | chkumar|ruck: sure | 12:11 |
| panda|rover | chkumar|ruck: in your channel | 12:12 |
| chkumar|ruck | panda|rover, https://redhat.bluejeans.com/1571313919/6145/?src=meet_now | 12:12 |
| chkumar|ruck | panda|rover, https://review.openstack.org/#/c/618477/ | 12:13 |
| chkumar|ruck | panda|rover, https://logs.rdoproject.org/openstack-periodic/git.openstack.org/openstack-infra/tripleo-ci/master/periodic-tripleo-ci-centos-7-standalone-master/0323c3f/logs/undercloud/home/zuul/undercloud_install.log.txt.gz#_2018-11-16_12_03_35 | 12:22 |
| *** quiquell|brb is now known as quiquell | 12:25 | |
| ssbarnea|bkp2 | marios: quiquell|brb chkumar|ruck: can we please revote on https://review.openstack.org/#/c/618131/ -- it was ready to workflow but the linters change merged before it so I had to update it. | 12:25 |
| chkumar|ruck | panda|rover, https://review.rdoproject.org/zuul/builds?job_name=periodic-tripleo-ci-centos-7-standalone-master | 12:25 |
| quiquell | ssbarnea|bkp2: is failing like shit | 12:26 |
| quiquell | ssbarnea|bkp2: http://logs.openstack.org/31/618131/4/check/tripleo-ci-centos-7-containers-multinode/433e323/job-output.txt.gz | 12:28 |
| quiquell | pardon my english | 12:29 |
| ssbarnea|bkp2 | quiquell: thanks! fixed. | 12:32 |
| quiquell | marios: Add you as the QE of scenario002 ? | 12:34 |
| marios | quiquell: sure | 12:36 |
| marios | ssbarnea|bkp2: so you decided to keep the 4 space indent? | 12:39 |
| * marios shakes fist | 12:39 | |
| ssbarnea|bkp2 | quiquell: speaking about shit... i just lost o huge stash and now I am digging inside git object cache to retrieve it... | 12:39 |
| ssbarnea|bkp2 | ... using git fsck for the first time but... I fount it. reminds me of msdos days of fixing FAT partition by hand. | 12:40 |
| quiquell | ssbarnea|bkp2: Hehe I think I have do both things, but totally forgot details :-) | 12:43 |
| ssbarnea|bkp2 | marios: not at all, i like 2 chars for yaml. only on ini files (tox.ini) I usually use 4. | 12:45 |
| ssbarnea|bkp2 | advise on yamllint: default settings are to check for spacing consistency inside the same file, meaning that it may decide to go to 2 or 4 based on what it finds first. thus we may see bit confusing errors but I hope everyone already knows we use "2". | 12:47 |
| marios | ssbarnea|bkp2: yes i am trolling you because of discussion here https://review.openstack.org/#/c/617118/1/roles/environment/tasks/main.yml | 12:47 |
| marios | ssbarnea|bkp2: and you 'fixed' from 4 to 2 here https://review.openstack.org/#/c/618131/3/roles/validate-tempest/tasks/tempest-rpm.yml but threw it all away in v 5 | 12:48 |
| marios | it could have been beautiful man | 12:48 |
| * marios back to regex | 12:48 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ (1 more message) | 12:49 |
| *** ratailor has quit IRC | 12:50 | |
| ssbarnea|bkp2 | i was curious about introduction of whitelist, i remember reading some comments about obliteration of white/black list in many codebases. i guess include/exclude are more neutral. | 12:50 |
| quiquell | marios: is Alex env file usable ? | 12:50 |
| marios | quiquell: how do you mean? it is being consumed by the job yes | 12:52 |
| marios | http://logs.openstack.org/72/616872/28/check/tripleo-ci-centos-7-scenario001-standalone/0134a94/logs/undercloud/home/zuul/standalone.sh.txt.gz | 12:52 |
| marios | quiquell: T^ like -e /usr/share/openstack-tripleo-heat-templates/ci/environments/scenario001-standalone.yaml \ | 12:52 |
| quiquell | marios: ack | 12:53 |
| chkumar|ruck | panda|rover, please have a look at this bug https://bugs.launchpad.net/tripleo/+bug/1803703 | 12:53 |
| openstack | Launchpad bug 1803703 in tripleo "[master] master standalone job is failing during undercloud install with Server Error for url: https://192.168.24.2:13696/v2.0/network" [Critical,Triaged] | 12:53 |
| panda|rover | chkumar|ruck: thanks | 12:54 |
| chkumar|ruck | see ya on monday, bye | 12:56 |
| *** chkumar|ruck has quit IRC | 12:56 | |
| weshay_pto | hey marios | 12:56 |
| marios | weshay_pto: o/ | 12:58 |
| weshay_pto | marios, want me to sum up what alex and I agreed on? I think he captured it well in the review | 12:58 |
| weshay_pto | He wants to be able to pass a list/array of files from tht to the standalone deployment command | 12:59 |
| weshay_pto | which I'm cool w/ as long as it's from tht | 12:59 |
| marios | weshay_pto: ack yeah all good seen it. commented here https://review.openstack.org/#/c/616872/27/zuul.d/standalone-jobs.yaml@94 | 12:59 |
| weshay_pto | if you want to chat I have a bit before I go | 12:59 |
| marios | weshay_pto: no its fine. in the end lets just do one thing. i think current form is fine. we DO have new param, because we are changing the original functionality of that standalone_custom_env_files and instead construct it from our param | 13:00 |
| marios | weshay_pto: we could also have argued to include ci/environments in the path but fine. ack on the list, i mean the original param is a list i just thought we'd need a single thing here for our override | 13:00 |
| weshay_pto | k | 13:01 |
| weshay_pto | {{ featureset_override | combine(custom_env, recursive=true) }} recursive is cool | 13:01 |
| weshay_pto | didn't know about that | 13:01 |
| marios | weshay_pto: quiquell has started poking at scen2 based on the current scen1 state, which i think is fairly well along now design wise esp. given this very latest discussion about the featureset | 13:02 |
| weshay_pto | sounds great | 13:02 |
| weshay_pto | there is a minor ceph issue on fs001 but you see that I'm sure | 13:02 |
| weshay_pto | panda|rover, need anything while i'm around? | 13:02 |
| weshay_pto | panda|rover, did you see inventory still failed on vexx | 13:03 |
| marios | weshay_pto: ack yeah had to read that task a couple times ;) | 13:03 |
| quiquell | weshay_pto: btw zuul-in-a-box deploying standalone and running tempest | 13:03 |
| marios | weshay_pto: but it seems to do the right thing :) passing it through http://logs.openstack.org/72/616872/28/check/tripleo-ci-centos-7-scenario001-standalone/0134a94/logs/undercloud/home/zuul/standalone.sh.txt.gz | 13:03 |
| weshay_pto | quiquell, oh fuck you.. that is amazing | 13:03 |
| weshay_pto | quiquell++ | 13:03 |
| hubbot1 | weshay_pto: quiquell's karma is now 13 | 13:03 |
| panda|rover | weshay_pto: yes, but we made progress at least, different error. We are working with kforde to debug the network failure in OVB, and standalone is failing in promotion, so to promote manually we would need to remove ovb and standalone from criteria | 13:04 |
| panda|rover | quiquell: zuul-in-a-box out of the box ? or with workarounds ? | 13:05 |
| weshay_pto | panda|rover, k.. and to handle the images | 13:06 |
| weshay_pto | panda|rover, by hand or by softlink | 13:06 |
| ssbarnea|bkp2 | quiquell: marios: it seems that godzilla fix passed linters so you are welcome to review it: https://review.openstack.org/#/c/618518/ | 13:06 |
| panda|rover | weshay_pto: so it's ok if we take an image that passes only fs002 and manually promote it ? even if the hash that built it is failing standalone ? | 13:09 |
| weshay_pto | panda|rover, I would not promote while the standalone is failing | 13:10 |
| weshay_pto | it's a gate job | 13:10 |
| weshay_pto | I saw chandan's bug on it | 13:10 |
| panda|rover | ssbarnea|bkp2: you forgot to include a patch in libc in your review | 13:10 |
| panda|rover | yep | 13:10 |
| weshay_pto | panda|rover, any upstream zuul job has to pass before we promote | 13:10 |
| ssbarnea|bkp2 | weshay_pto: now that I see you "out" of pto, i may try to remind about https://review.openstack.org/#/c/617338/ ? | 13:10 |
| weshay_pto | as we don't want to break the gate | 13:10 |
| weshay_pto | but ovb is optional at this point.. | 13:11 |
| weshay_pto | ssbarnea|bkp2, nag some folks in #tripleo man | 13:11 |
| weshay_pto | ssbarnea|bkp2, sbaker, mandre, dprince, alex etc | 13:11 |
| weshay_pto | ssbarnea|bkp2, k? | 13:12 |
| ssbarnea|bkp2 | weshay_pto: sure! i didn't know the magic names, usually I look at who +W last changes to the repo. | 13:12 |
| weshay_pto | ssbarnea|bkp2, aye np | 13:13 |
| weshay_pto | l8r I'll be back about 2pm MST.. take care | 13:16 |
| weshay_pto | have a good weekend | 13:16 |
| ssbarnea|bkp2 | marios: https://review.openstack.org/#/c/618477/ please. | 13:21 |
| *** trown|outtypewww is now known as trown | 13:31 | |
| *** quiquell is now known as quiquell|lunch | 13:32 | |
| *** quiquell|lunch is now known as quiquell | 13:35 | |
| *** holser|lunch is now known as holser_ | 13:36 | |
| *** dtrainor has joined #oooq | 13:41 | |
| *** rlandy has joined #oooq | 13:46 | |
| *** agopi_ is now known as agopi|brb | 13:46 | |
| *** dtantsur is now known as dtantsur|afk | 13:50 | |
| *** agopi|brb has quit IRC | 13:50 | |
| rlandy | marios: hello | 13:53 |
| panda|rover | ykarel: ping | 13:55 |
| ykarel | panda|rover, pong | 13:55 |
| marios | rlandy: o/ | 13:57 |
| panda|rover | ykarel: overcloud logs, I hoped it was quicker but We stil can't reach the overcloud nodes over ssh | 13:59 |
| panda|rover | ykarel: so this https://review.openstack.org/618205 works, but w'ell have to wait for the network problems to be resolved before we have them | 14:00 |
| *** amoralej is now known as amoralej|off | 14:03 | |
| panda|rover | amoralej|off: NOOOOOOOOOO | 14:03 |
| *** amoralej|off is now known as amoralej | 14:04 | |
| *** amoralej is now known as amoralej|lunch | 14:04 | |
| quiquell | panda|rover: Do we have to backport python3 changes to rocky ? | 14:04 |
| panda|rover | quiquell: no | 14:05 |
| panda|rover | quiquell: did you get any indication on this ? | 14:05 |
| panda|rover | rlandy: do you have time for me today ? Apparently we have packages in Ceotos7.6 repos and we need to start some tests | 14:08 |
| rlandy | panda|rover: in meeting - will ping you | 14:08 |
| quiquell | panda|rover: nop just pop up into my mind | 14:09 |
| panda|rover | rlandy: sure, I'm in meetint too, just malloc() your time. | 14:09 |
| *** quiquell is now known as quiquell|lunch | 14:09 | |
| *** agopi|brb has joined #oooq | 14:10 | |
| *** agopi|brb is now known as agopi | 14:10 | |
| ykarel | panda|rover, ack so u are running by turning of SSL, and now script is able to get overcloud_nodes list, but this time SSH error due to network issues | 14:13 |
| ykarel | ? | 14:13 |
| ykarel | panda|rover, do u know what is the network issue? | 14:13 |
| panda|rover | ykarel: I think I saw ssh problems even before that patch, but I thought it was related to this bug | 14:19 |
| panda|rover | ykarel: we are trying with kforde to hold a deploy and debug live | 14:20 |
| ykarel | panda|rover, ack, yes that would help a lot | 14:20 |
| *** ykarel is now known as ykarel|away | 14:26 | |
| *** derekh has quit IRC | 14:27 | |
| *** derekh has joined #oooq | 14:27 | |
| *** quiquell|lunch is now known as quiquell | 14:30 | |
| *** ykarel|away has quit IRC | 14:31 | |
| *** saneax has quit IRC | 14:32 | |
| *** vkapalav has joined #oooq | 14:35 | |
| quiquell | rlandy: the zuul-in-a-box thingy is deploy standalone and running tempest | 14:40 |
| quiquell | rlandy: executor get stuck though, I have to discover why | 14:40 |
| quiquell | rlandy: do you want to sync ? | 14:40 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ (1 more message) | 14:49 |
| rlandy | quiquell: in meeting with marios | 14:50 |
| rlandy | panda|rover: fyi .. | 14:51 |
| rlandy | <kforde> just to say ovb is on cloud-d now ... I think your projects are there already | 14:51 |
| rlandy | <kforde> so you guys could start testing there | 14:51 |
| quiquell | rlandy: ack | 14:53 |
| *** amoralej|lunch is now known as amoralej | 15:00 | |
| *** ratailor has joined #oooq | 15:02 | |
| quiquell | marios: Can we split scenario001 job review into to https://review.openstack.org/#/c/616872 ? | 15:03 |
| quiquell | marios: one with the workflow change and the other with the job itself | 15:03 |
| quiquell | marios: so other scnearios depend only on the workflow change | 15:04 |
| *** ykarel|away has joined #oooq | 15:26 | |
| rlandy | panda|rover: ok - ready to meet when you are | 15:36 |
| *** ratailor has quit IRC | 15:39 | |
| marios | quiquell: ack sorry was on phone. sure we can do it, you mean just the run-v3 change? | 15:40 |
| quiquell | marios: Yep, so we can merge it before the others, if it's not breaking anything | 15:40 |
| marios | quiquell: yeah sec, doing | 15:41 |
| quiquell | marios: cool thanks | 15:41 |
| marios | quiquell: on its way | 15:48 |
| quiquell | ssbarnea|bkp2: the linting change has break upgrade job | 15:59 |
| quiquell | ssbarnea|bkp2: Nah false alarm | 16:02 |
| *** quiquell is now known as quiquell|off | 16:13 | |
| *** jfrancoa has quit IRC | 16:13 | |
| *** agopi is now known as agopi|food | 16:25 | |
| ssbarnea|bkp2 | marios: quiquell|off : we need to get https://review.openstack.org/#/c/618518/ in before it upsets too many people about having to fix linting in touched files. | 16:26 |
| *** d0ugal has quit IRC | 16:27 | |
| *** d0ugal has joined #oooq | 16:27 | |
| *** agopi|food has quit IRC | 16:33 | |
| *** agopi has joined #oooq | 16:33 | |
| *** agopi is now known as agopi|food | 16:34 | |
| *** bogdando has quit IRC | 16:42 | |
| *** holser_ has quit IRC | 16:42 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ (1 more message) | 16:50 |
| *** dtrainor has quit IRC | 16:51 | |
| *** agopi|food is now known as agopi | 17:06 | |
| rlandy | panda|rover: hello | 17:13 |
| rlandy | panda|rover: so you happen to know when the zuul applies the change if it is not in dlrn? | 17:14 |
| agopi | hey rlandy, I don't see any builds for openstack/browbeat here in https://review.rdoproject.org/zuul/builds?job_name=tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset053 | 17:16 |
| agopi | I don't see any changes to https://github.com/rdo-infra/review.rdoproject.org-config/blob/master/zuul.d/tripleo.yaml#L719 as well | 17:16 |
| agopi | is everything fine and is anything needed from my end? | 17:17 |
| *** ykarel|away has quit IRC | 17:28 | |
| *** trown is now known as trown|lunch | 17:44 | |
| *** quiquell|off has quit IRC | 17:51 | |
| *** derekh has quit IRC | 18:01 | |
| *** rfolco has quit IRC | 18:31 | |
| rlandy | agopi: not sure - been working exclusively on the reproducer - panda|rover?? if not - I'll look in a bit | 18:36 |
| agopi | oh okay cool rlandy. panda|rover, please lmk if you need anything from my side. | 18:37 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, master: tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ (1 more message) | 18:50 |
| *** amoralej is now known as amoralej|off | 18:55 | |
| *** trown|lunch is now known as trown | 19:11 | |
| *** apetrich has quit IRC | 19:28 | |
| *** saneax has joined #oooq | 19:37 | |
| *** apetrich has joined #oooq | 19:42 | |
| *** saneax has quit IRC | 20:09 | |
| *** saneax has joined #oooq | 20:21 | |
| *** saneax has quit IRC | 20:38 | |
| *** rlandy has quit IRC | 20:47 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario008-multinode-oooq-container, tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, stable/pike: tripleo-ci-centos-7-scenario004-multinode-oooq-container @ https://review.openstack.org/602248, master: tripleo-ci- (2 more messages) | 20:50 |
| panda|rover | FINALLY I could get rdoinfo job to launch quickstart ... was losing my sanity | 21:03 |
| *** panda|rover is now known as panda|rover|off | 21:03 | |
| *** agopi has quit IRC | 21:14 | |
| *** holser_ has joined #oooq | 21:18 | |
| *** holser__ has joined #oooq | 21:28 | |
| *** holser__ has quit IRC | 21:28 | |
| *** holser_ has quit IRC | 21:31 | |
| *** trown is now known as trown|outtypewww | 22:07 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario008-multinode-oooq-container, tripleo-ci-centos-7-scenario001-multinode-oooq-container, tripleo-ci-centos-7-scenario007-multinode-oooq-container, tripleo-ci-centos-7-standalone-upgrade @ https://review.openstack.org/604298, stable/pike: tripleo-ci-centos-7-scenario004-multinode-oooq-container @ https://review.openstack.org/602248, master: tripleo-ci- (2 more messages) | 22:50 |
| *** jtomasek has quit IRC | 23:15 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!