| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp_1ceph-featureset024-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ (3 more messages) | 00:48 |
|---|---|---|
| *** rlandy has joined #oooq | 01:05 | |
| *** rlandy has quit IRC | 01:19 | |
| *** apetrich has quit IRC | 02:42 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp_1ceph-featureset024-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ (3 more messages) | 02:48 |
| *** skramaja has joined #oooq | 03:25 | |
| *** udesale has joined #oooq | 03:53 | |
| *** ykarel has joined #oooq | 03:59 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates, tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp_1ceph-featureset024-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ (3 more messages) | 04:48 |
| *** quique|rover|off is now known as quiquell|rover | 05:40 | |
| *** apetrich has joined #oooq | 05:46 | |
| *** jfrancoa has joined #oooq | 05:57 | |
| *** sanjayu_ has joined #oooq | 06:08 | |
| *** ratailor has joined #oooq | 06:12 | |
| *** jtomasek has joined #oooq | 06:13 | |
| *** holser_ has joined #oooq | 06:15 | |
| *** dtrainor has quit IRC | 06:33 | |
| *** quiquell|rover is now known as quique|rover|brb | 06:43 | |
| *** chkumar|off is now known as chkumar|ruck | 06:44 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates, tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp_1ceph-featureset024-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ (3 more messages) | 06:48 |
| *** quique|rover|brb is now known as quiquell|rover | 07:00 | |
| *** sanjayu_ has quit IRC | 07:08 | |
| *** bogdando has joined #oooq | 07:43 | |
| *** chem has joined #oooq | 07:51 | |
| chkumar|ruck | weshay: what was the task about mirrors? | 07:57 |
| *** ykarel is now known as ykarel|lunch | 07:59 | |
| bogdando | PTAL https://review.openstack.org/#/c/595265/ https://review.openstack.org/#/c/592996/ and https://review.openstack.org/#/c/576746/ | 08:04 |
| *** gkadam has joined #oooq | 08:16 | |
| *** ykarel|lunch is now known as ykarel | 08:21 | |
| *** sanjayu_ has joined #oooq | 08:29 | |
| *** fhubik has joined #oooq | 08:30 | |
| ssbarnea | chkumar|ruck: i am not sure, I seen that web assigned it to me but is not clear to me what needs to be done about it. | 08:38 |
| ssbarnea | btw, can someone give me a hint on why virt-resize failed here https://seashells.io/v/uwKySCy3 ? | 08:39 |
| ssbarnea | i have the impression that this is another case of image ownership issue. i checked and the images are owned by root user. | 08:47 |
| *** tosky has joined #oooq | 08:48 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates, tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp_1ceph-featureset024-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ (4 more messages) | 08:48 |
| *** apetrich has quit IRC | 08:49 | |
| ssbarnea | chkumar|ruck: any idea? | 08:49 |
| chkumar|ruck | ssbarnea: can yourun the particular command manually and check what happened | 08:51 |
| chkumar|ruck | during virt-resize i hitted the same problem | 08:51 |
| ssbarnea | just did it an got a much better error due to LIBGUESTFS_DEBUG=1 | 08:51 |
| ssbarnea | LIBGUESTFS_TRACE=1 vars. | 08:51 |
| ssbarnea | this reminds me about that CR that didn't get merged. | 08:52 |
| chkumar|ruck | try to sudo and then re-run the playbook | 08:52 |
| chkumar|ruck | it vanishes | 08:52 |
| chkumar|ruck | I have seen that issue like 2 montsh back | 08:52 |
| ssbarnea | ahh, damn. it was using command instead of shell, and environment is ignored. | 08:53 |
| chkumar|ruck | ssbarnea: shared a doc when I was tryoing to experiment with shift on stack | 08:54 |
| chkumar|ruck | it has notes | 08:54 |
| *** chkumar|ruck has quit IRC | 08:58 | |
| sshnaidm | bogdando, if you want non-destructive teardown - why not to use specific set of tags/vars? for example "full teardown" and "lite teardown". Then let's just organize current set of tasks and mark them with these "lite" or "full" | 08:59 |
| *** chandankumar has joined #oooq | 08:59 | |
| *** chandankumar is now known as chkumar|ruck | 09:00 | |
| sshnaidm | bogdando, users removal will be "full", directories etc - "lite" | 09:00 |
| bogdando | I suppose the changing teardown-all into those two, would bring too much of the changes to spread around | 09:00 |
| bogdando | I'd like to keep tags unchanged | 09:00 |
| bogdando | if possible | 09:01 |
| sshnaidm | bogdando, we can just add teardown-lite, when teardown-all will remain its default | 09:01 |
| bogdando | there is too much magic rely on tags | 09:01 |
| bogdando | literally everythere | 09:01 |
| sshnaidm | bogdando, yeah, maybe better to use var | 09:01 |
| *** ykarel is now known as ykarel|away | 09:01 | |
| bogdando | the idea is to keep the default UX non destructive, no special flags | 09:01 |
| bogdando | if users knows about a flag, highly likely he knows already how to not shoot into his leg | 09:02 |
| bogdando | a user* | 09:02 |
| bogdando | that's not the situation the patch is intended to fix | 09:02 |
| bogdando | so the default command becomes non destructive if a few conditions met | 09:03 |
| bogdando | makes sense? | 09:03 |
| bogdando | but ofc I could just go with a flag | 09:03 |
| sshnaidm | bogdando, it changes the default - to be destructive. When you introduce the liter teardown - you give users additional option not to destroy everything, while other that don't need this will not have to change their workflow. | 09:03 |
| sshnaidm | bogdando, this I call good user experience | 09:03 |
| bogdando | I think they won't change it as the conditions won't trigger changes | 09:04 |
| bogdando | won't have to change anything* | 09:04 |
| sshnaidm | bogdando, not really according to your patches | 09:04 |
| bogdando | let me double check... | 09:04 |
| sshnaidm | bogdando, these conditions are actually magic | 09:04 |
| *** ykarel|away has quit IRC | 09:05 | |
| bogdando | https://review.openstack.org/#/c/592996/ , right? | 09:05 |
| sshnaidm | bogdando, it's better to use well defined option than start to think now when and how use quickstart | 09:05 |
| sshnaidm | bogdando, https://review.openstack.org/#/c/593567/2 | 09:06 |
| sshnaidm | bogdando, these conditions are just guessing, why to make AI there instead of cli option? | 09:06 |
| bogdando | sshnaidm: virthost != 'localhost' blocks anyone using quickstart the old way from that change | 09:07 |
| *** dtrainor has joined #oooq | 09:07 | |
| sshnaidm | bogdando, what do you mean? | 09:07 |
| bogdando | if a user specifies that option, and the non root user matches $USER env var, he enters nuke-strikes-safe zone | 09:07 |
| bogdando | automagically | 09:07 |
| bogdando | that's how I see the case | 09:08 |
| bogdando | cuz those two conditions basically mean I deploy locally, on my devbox | 09:08 |
| bogdando | using the logged user | 09:08 |
| *** kopecmartin|off is now known as kopecmartin | 09:09 | |
| ssbarnea | based on execution logs I believe that environment variables are not setup, even if documentation stated that it should also work for command https://docs.ansible.com/ansible/2.6/user_guide/playbooks_environment.html | 09:09 |
| bogdando | and only if we want a CI job deploying locally with zuul user, then I propose adding a flag to overwrite that safe guard | 09:09 |
| ssbarnea | maybe it was fixed at some point? i remember reading somewhere that it was not working for commands but was working for shell. | 09:09 |
| bogdando | ssbarnea: USER is always set up | 09:09 |
| bogdando | and lookup plugins picks it up | 09:09 |
| bogdando | that's not the environment module, but a lookup | 09:10 |
| *** dtantsur|afk is now known as dtantsur | 09:10 | |
| ssbarnea | bogdando: I think we are talking about different things. See https://github.com/openstack/tripleo-quickstart/blob/master/roles/libvirt/setup/undercloud/tasks/main.yml#L204 -- i know for sure that command run without these variables being set. | 09:11 |
| bogdando | ssbarnea: I'm not sure I follow which patch do we discuss | 09:15 |
| sshnaidm | bogdando, I do NOT want to guess and then to be caught by some different situation which we didn't count in this set of conditions, and I do not think it's good to guess here instead of giving user to decide | 09:15 |
| bogdando | sshnaidm: ack, will update the patch then | 09:15 |
| sshnaidm | thanks | 09:16 |
| bogdando | ssbarnea: to fix https://github.com/openstack/tripleo-quickstart/blob/master/roles/libvirt/setup/undercloud/tasks/main.yml#L204 there is other patch I pushed | 09:18 |
| bogdando | https://review.openstack.org/#/q/topic:command_vs_shell+(status:open+OR+status:merged) | 09:18 |
| bogdando | sshnaidm: ^^ btw PTAL :) | 09:18 |
| bogdando | and one more thing as we have a review session here https://review.openstack.org/#/c/605021/ :) | 09:19 |
| ssbarnea | sshnaidm: chkumar|ruck : I can confirm that https://review.openstack.org/#/c/605021/ is fixing my error, lets review it and merge it. | 09:24 |
| *** ssbarnea|bkp has quit IRC | 09:28 | |
| sshnaidm | bogdando, "which does not happen with 'command'." - why? | 09:28 |
| bogdando | sshnaidm: just tested locally | 09:30 |
| bogdando | that's how it works | 09:30 |
| sshnaidm | bogdando, me too.. works great with environment.. hmm | 09:30 |
| bogdando | not the command module | 09:30 |
| bogdando | only shell does | 09:30 |
| sshnaidm | bogdando, does it work? | 09:31 |
| sshnaidm | - name: dd | 09:31 |
| sshnaidm | command: echo $VVV | 09:31 |
| sshnaidm | environment: | 09:31 |
| sshnaidm | VVV: zero | 09:31 |
| bogdando | sshnaidm: https://pastebin.com/WKFJ9FFr | 09:33 |
| sshnaidm | bogdando, does my task work for you? | 09:35 |
| bogdando | nevermind, I think it works | 09:36 |
| bogdando | https://pastebin.com/jYkK6Mqv worked well, sshnaidm, ssbarnea | 09:36 |
| *** dsneddon has quit IRC | 09:37 | |
| ssbarnea | bogdando: yep, I did something almost identical and tested, and environment variables are set for both command and shell, so it must be something else. The reality is still there: the output of the command when run inside ansible was very different than the one run outside ansible. the ansible one behaved like debug was not there, with a non informative enough error. | 09:39 |
| chkumar|ruck | ssbarnea: yes, | 09:40 |
| sshnaidm | ssbarnea, make sure you have all dependencies installed, I remember some problem with virt-resize whn not having all libguestfs tools installed | 09:41 |
| chkumar|ruck | ssbarnea: panda|off is working on a job to test the libvirt reproducer in ci https://review.openstack.org/#/c/604232/ | 09:41 |
| ssbarnea | sshnaidm: is not the first time I use it on this machine, sudo patch worked. and I will buy panda a beer the first time we meet, we really need th repro testing in CI. | 09:43 |
| ssbarnea | it could have saved so many hours | 09:43 |
| bogdando | sshnaidm: thanks for reviews! I hope we can let https://review.openstack.org/#/c/592996/ and https://review.openstack.org/#/c/576746/ in, at least | 09:43 |
| sshnaidm | bogdando, so you don't add undercloud here? https://review.openstack.org/#/c/576746/10/roles/tripleo-inventory/tasks/inventory.yml | 09:47 |
| bogdando | sshnaidm: https://review.openstack.org/#/c/605379/ if you don't mind to not change https://review.openstack.org/#/c/592996/4 which passed CI, has +2/+1 et al... | 09:48 |
| bogdando | sshnaidm: not virthost_with_private_key is a subcase of virthost == 'localhost' | 09:50 |
| bogdando | we do not use a key for the latter as well | 09:50 |
| bogdando | but we need to add the full case | 09:50 |
| bogdando | to make things working | 09:51 |
| bogdando | basically, that *adds* undercloud for the broader case | 09:51 |
| bogdando | which is not having a key *or* running with localhost virthost | 09:51 |
| ssbarnea | bogdando: sorry for adding a -1 on https://review.openstack.org/#/c/592996/4/playbooks/cleanup-destructive.yml@24 -- but i think we need that option. | 09:54 |
| bogdando | no worries, just wanted to not burn CI again and ask for restored +2 | 09:54 |
| sshnaidm | bogdando, maybe I miss something, but I don't see another "add host" from undercloud there, how is that added then? | 09:56 |
| sshnaidm | bogdando, nevermind, found | 09:59 |
| *** jfrancoa has quit IRC | 10:04 | |
| *** jfrancoa has joined #oooq | 10:05 | |
| ssbarnea | bogdando: please have a look at https://review.openstack.org/#/c/596428 which was my attempt to fix the same bug with resize, 4 weeks ago! For me that fixed sorted the issue and seems less risky than running with sudo, not clear to me why Sagi downvoted it. If you could have a look it would be great. | 10:09 |
| sshnaidm | ssbarnea, you just need to read comments and you will understand why I downvoted it | 10:10 |
| ssbarnea | sshnaidm: as i was hit by the same bug again, here is the log https://seashells.io/v/uwKySCy3 | 10:12 |
| sshnaidm | ssbarnea, ok, please try to find a root cause | 10:13 |
| ssbarnea | i already found the root cause: the qcow images are owned by root, so that is why resize fails. running with become works, or changing ownership also works. either approach is ok. | 10:14 |
| ssbarnea | bogdan was hit by the same issue, and he picked another approach. i don't came much which one will go in as long we have a fix. | 10:14 |
| bogdando | ssbarnea: it's too fragile to change an owner of a file | 10:15 |
| bogdando | sadly, I'm pretty sure that breaks for over places in quickstart | 10:15 |
| bogdando | and no one knows there | 10:16 |
| bogdando | where* | 10:16 |
| sshnaidm | ssbarnea, seems like it's red herring, not the root cause, more deep investigation is required | 10:16 |
| bogdando | sshnaidm: the sudo fix helps at least me and Tengu to deploy with libvirt reproducer script | 10:18 |
| bogdando | so the case is very clear | 10:18 |
| bogdando | just fetch and run the repro script with -l | 10:18 |
| *** fhubik is now known as fhubik|brb | 10:18 | |
| * Tengu can confirm | 10:18 | |
| bogdando | not being root | 10:18 |
| Tengu | although the probable root cause is "undercloud.qcow2 and undercloud_resized.qcow2 belongs to root for some reason" | 10:19 |
| bogdando | there is so much places in quickstart things belong to root, I mean really... | 10:19 |
| Tengu | probably one of the reasons why I don't use it ^^' | 10:19 |
| bogdando | that's fine, maybe, by design, just want to make repro working :) | 10:19 |
| Tengu | that said: https://review.openstack.org/#/c/596428/ seems to address the same issue in a more.... "user" way. | 10:20 |
| bogdando | I think Sagi is right and that breaks the module | 10:20 |
| Tengu | probably missing the remote_src: true flag. | 10:21 |
| quiquell|rover | bogdando, Tengu, sshnaidm: to fix timeouts https://review.openstack.org/#/c/605377/ | 10:22 |
| quiquell|rover | marios: ^ | 10:22 |
| ssbarnea | AFAIK, at least in theory, reproducer should need root access to the machine on which it is running on, as it either does its job with openstack cloud or libvirt which is recommended not to be run as root anyway. i looked inside the temp directory (workspace) and found that the only two files not owned by my user were the qcow2 images,... I don't think that these files should be owned by root. | 10:22 |
| Tengu | anyway. one or the other should be accepted/merged in order to make the reproducer work as expected when using libvirt. | 10:22 |
| Tengu | quiquell|rover: for real? :) | 10:22 |
| Tengu | ssbarnea: +1 | 10:23 |
| quiquell|rover | Tengu: yep... response from mistral before websocket :-/ tripleo was too fast (that's new) | 10:23 |
| Tengu | o___O | 10:23 |
| * Tengu is sitting, that's a good news, else he would have fall on the floor | 10:23 | |
| quiquell|rover | Tengu: or websocket is way way way slow :-) | 10:24 |
| Tengu | :) | 10:24 |
| quiquell|rover | Tengu: anywais let's merge this, therve did great job debugging it with me | 10:24 |
| Tengu | indeed | 10:24 |
| Tengu | therve++ on that | 10:25 |
| quiquell|rover | Tengu: who else have super powers at tripleoclient ? | 10:25 |
| Tengu | quiquell|rover: not me. but bogdando does, guess he's waiting for CI result. | 10:26 |
| Tengu | quiquell|rover: for now I don't have any core role on any repository. | 10:26 |
| quiquell|rover | bogdando: damn you are always a fast +2 :-) | 10:26 |
| Tengu | XD | 10:26 |
| quiquell|rover | bogdando: getting old ? | 10:27 |
| bogdando | +1 ;) | 10:27 |
| sshnaidm | bogdando, what is full quickstart command that you run on localhost? is it from root user or non-root? | 10:27 |
| bogdando | sshnaidm: which patch is it? | 10:27 |
| Tengu | sshnaidm: non-root (stack user with full sudo), command: bash reproducer-quickstart.sh -w $(pwd)/work -v true -a -l | 10:28 |
| sshnaidm | bogdando, testing https://review.openstack.org/#/c/576746/ | 10:28 |
| bogdando | sshnaidm: same as Tengu says | 10:28 |
| Tengu | sshnaidm: and trying to run that one to be precise: http://logs.openstack.org/17/600517/29/check/tripleo-ci-centos-7-undercloud-containers/492e203/logs/reproducer-quickstart.sh | 10:28 |
| Tengu | if it's still available - but there are many other changes you can try I guess :) | 10:28 |
| bogdando | sshnaidm: thanks! | 10:29 |
| sshnaidm | bogdando, Tengu are we talking about reproducer or quickstart.sh? | 10:29 |
| Tengu | sshnaidm: reproducer that runs quickstart | 10:29 |
| sshnaidm | bogdando, I meant in patch https://review.openstack.org/#/c/576746/ - how do you run quickstart.sh there? | 10:38 |
| Tengu | oh. :) | 10:39 |
| sshnaidm | bogdando, because it fails for me with: ./quickstart.sh -R master-tripleo-ci --bootstrap --no-clone 127.0.0.1 | 10:40 |
| chkumar|ruck | ssbarnea: do we have a working standalone fs28 job in ci? I need to play with it | 10:41 |
| chkumar|ruck | *f28 | 10:41 |
| sshnaidm | bogdando, also with ./quickstart.sh -R master-tripleo-ci --bootstrap --no-clone localhost : http://paste.openstack.org/show/730933/ | 10:42 |
| ssbarnea | chkumar|ruck: nople, we havve 6 wip patches for f28 and I think this will be a card for the next sprint. | 10:42 |
| ssbarnea | this is how i found that issue with reproducer, while trying to work on f28 patches. | 10:43 |
| sshnaidm | Tengu, can you please paste your _quickstart.log when you run reproducer and it fails for you? | 10:44 |
| sshnaidm | Tengu, without any other fixes/patches | 10:44 |
| sshnaidm | or bogdando ^^ | 10:44 |
| sshnaidm | ssbarnea, why do you ignore my comment to patch 596428? | 10:45 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates, tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (2 more messages) | 10:48 |
| ssbarnea | sshnaidm: not ignoring, just addresing each comment in order, check last version and you will see the issue addressed via remote_src: true | 10:54 |
| Tengu | ssbarnea: err ok. wait. | 11:03 |
| Tengu | sshnaidm: -^ (sorry) | 11:04 |
| Tengu | ssbarnea: ah, so I was right earlier with the "remote_src" :). good | 11:05 |
| Tengu | sshnaidm: hm, was something merged in the meanwhile? because apparently right now it seems to work as expected. | 11:06 |
| sshnaidm | Tengu, that's I call magic | 11:06 |
| Tengu | unless my proxy is messing with me, but it should not for SSL thingy. | 11:06 |
| sshnaidm | ssbarnea, try please to use the latest code | 11:07 |
| Tengu | sshnaidm: did you patch anything? care to share? | 11:07 |
| sshnaidm | Tengu, nope, me did nothing.. just blocked other patches | 11:08 |
| Tengu | o_O | 11:08 |
| Tengu | well, something must have changed. yesterday we just couldn't do anything. | 11:09 |
| sshnaidm | Tengu, I think there was some weshay patch to install lacking dependency for virt-resize to work, not sure if it's merged, lemme check.. | 11:09 |
| sshnaidm | Tengu, well, it wasn't.. | 11:11 |
| *** udesale has quit IRC | 11:12 | |
| sshnaidm | Tengu, so the problem is with environment, seems like if it worked once it'll continue to work.. need to check it on clean environment from scratch | 11:12 |
| Tengu | sshnaidm: hmm I dropped all the "work" directory. | 11:13 |
| sshnaidm | Tengu, I mean virtbox itself | 11:13 |
| Tengu | ah. well. nope, won't reinstall it right now ;) | 11:13 |
| chkumar|ruck | quiquell|rover: we need to cover program call today | 11:38 |
| quiquell|rover | chkumar|ruck: program call is about rocky ? | 11:40 |
| chkumar|ruck | quiquell|rover: yes | 11:41 |
| chkumar|ruck | quiquell|rover: we are green there currrently periodic job is also running | 11:41 |
| quiquell|rover | we have even phase2 rocky promotions, so all green | 11:41 |
| quiquell|rover | chkumar|ruck: doc is updated, looks like weshay will cover it | 11:42 |
| quiquell|rover | Let's just connect | 11:42 |
| chkumar|ruck | quiquell|rover: yup | 11:42 |
| quiquell|rover | chkumar|ruck: let's have a quiet rest of week please :-) | 11:43 |
| * chkumar|ruck is going to take a deep sleep tomorrow | 11:43 | |
| chkumar|ruck | quiquell|rover: last night I had this https://www.zomato.com/photos/pv-res-50708-u_NzA4OTE1NDMwMz and I was not able to walk, to digest it I walked around 5 Km | 11:45 |
| bogdando | sshnaidm: 'sudo: a password is required' shows the environment preparations issues | 11:46 |
| bogdando | points to some steps missing, as we normally do for quickstart. | 11:46 |
| bogdando | may be it's allowing sudo w/o a password? | 11:46 |
| sshnaidm | bogdando, all users have passwordless sudo | 11:46 |
| sshnaidm | sshnaidm, see the link to ansible docs I posted | 11:46 |
| sshnaidm | bogdando, ^^ | 11:47 |
| quiquell|rover | chkumar|ruck: that's massive !!! you are still alive ? | 11:48 |
| bogdando | sshnaidm: perhaps it needs https://review.openstack.org/#/c/593103/ then | 11:48 |
| quiquell|rover | chkumar|ruck: damn now I am hungry | 11:48 |
| bogdando | I'll test this on a fresh vm, sshnaidm | 11:49 |
| sshnaidm | bogdando, that's why I asked - how do you run it? Does it fail for you too? | 11:49 |
| bogdando | yeah, will check that and fix dependencies | 11:49 |
| sshnaidm | bogdando, thanks | 11:49 |
| weshay | sshnaidm, Tengu that is what I've seen, once it works once you no longer hit issues. I figured we kill that this sprint | 11:53 |
| weshay | sshnaidm, Tengu still bogdan has a change to move everything to shell and my change for deps | 11:53 |
| weshay | both should get reviewed | 11:53 |
| chkumar|ruck | quiquell|rover: yes I am | 11:53 |
| bogdando | weshay: oops, I abandoned that move to shell, the issue is wrong, and environment works with the command module :) | 11:54 |
| sshnaidm | weshay, it's not the reason | 11:54 |
| weshay | ah k | 11:54 |
| chkumar|ruck | quiquell|rover: they were surving as much as I can eat | 11:54 |
| chkumar|ruck | quiquell|rover: I call it Andhara food | 11:55 |
| weshay | sshnaidm, Tengu https://review.openstack.org/#/c/600836/ | 11:55 |
| Tengu | weshay: hmm I think I have that libguestfs thingy on my node by default - I already change image content with my tripleo-lab. | 11:55 |
| ssbarnea | weshay: i see you moved the meeting to avoid the clash, good. | 11:57 |
| sshnaidm | weshay, Tengu I think it's different issue.. | 11:57 |
| *** trown|outtypewww is now known as trown | 11:59 | |
| ssbarnea | i remembered that I wanted to ask you something: wouldn't be better if the temp folder used by reproducer would be "reproducable" instead of random? reproducable meaning that the same script should have a the same temp file, so re-running it would not generate a new file. | 12:00 |
| ssbarnea | there are multiple ways to obtain this: pregenerating a random name when we generate the file with ansilbe, or at runtime to use a sha1 of its own sourcecode for example. | 12:02 |
| weshay | ssbarnea, you do not need to join this meeting | 12:06 |
| weshay | low value | 12:06 |
| weshay | ssbarnea, usually the ruck will join | 12:07 |
| sshnaidm | bogdando, btw, did you try one of custom teardowns? https://github.com/openstack/tripleo-quickstart/blob/master/quickstart.sh#L245 is it a problem to use one of them? | 12:07 |
| sshnaidm | bogdando, for example "nodes" | 12:07 |
| ssbarnea | weshay: yep, lcuky me, you do nothave the same option. | 12:07 |
| bogdando | sshnaidm: good idea | 12:08 |
| bogdando | just not sure it does what I want :) | 12:08 |
| sshnaidm | bogdando, well, it shouldn't remove user, etc | 12:08 |
| ssbarnea | I updated bogdando fix at https://review.openstack.org/#/c/605021/ to include link to bug, we will need to revote. | 12:09 |
| bogdando | right, but I'd like to recreate all the libvirt things not only nodes | 12:09 |
| bogdando | and inventory and ssh config | 12:09 |
| bogdando | I think only nodes teardown won't do that, sshnaidm | 12:09 |
| sshnaidm | bogdando, ok, but it's not a problem to add there additional option with your set | 12:10 |
| sshnaidm | bogdando, that will include your set of removals | 12:10 |
| bogdando | sure, but adding more options sometimes is not good | 12:10 |
| sshnaidm | bogdando, I mean we have this already in design, just need to extend it | 12:10 |
| bogdando | that would also work for me | 12:11 |
| bogdando | whatever not forcing me to rebase my fork to keep deploying things | 12:11 |
| *** apetrich has joined #oooq | 12:11 | |
| weshay | sshnaidm, I need rocky ovb jobs in sova please | 12:15 |
| weshay | sshnaidm, I sent a pull | 12:15 |
| sshnaidm | weshay, yeah, will merge | 12:16 |
| *** dtantsur is now known as dtantsur|brb | 12:18 | |
| *** rlandy has joined #oooq | 12:18 | |
| bogdando | sshnaidm: that's weird, but I can see the default stack user created but not granted sudo | 12:20 |
| rlandy | weshay: hello - is panda|off out today? | 12:20 |
| *** quiquell|rover is now known as quique|rover|lch | 12:20 | |
| bogdando | seems like that a side issue for quickstart | 12:20 |
| bogdando | please check if sudo works really for the stack user at your env | 12:20 |
| *** ratailor has quit IRC | 12:20 | |
| weshay | rlandy, welcome back.. he should be in but he was a little sick | 12:20 |
| weshay | haven't seen anything | 12:20 |
| weshay | from him | 12:20 |
| *** panda|off is now known as panda | 12:21 | |
| panda | rlandy: I'm not | 12:21 |
| bogdando | sshnaidm: basically, the task 'Ensure XDG_RUNTIME_DIR is set' is executed before 'Grant sudo privileges to non-root user'. And I should also fix that! | 12:21 |
| rlandy | panda: hello - just wanted to catch up on the zuulv3 stuff | 12:22 |
| sshnaidm | bogdando, hmm.. seems like it wasn't supposed to run as this user.. | 12:22 |
| rlandy | what can I do to help there? | 12:22 |
| panda | rlandy: have you read the latest updates ? | 12:23 |
| rlandy | I read them last night | 12:23 |
| rlandy | you moved the parents to config | 12:23 |
| bogdando | sshnaidm: why? | 12:23 |
| bogdando | you didn't put a custom non_root_user for the command | 12:24 |
| *** agopi has quit IRC | 12:24 | |
| rlandy | panda: wasn;t sure if that would be for all jobs or only periodic | 12:24 |
| bogdando | so we expect it takes defaults | 12:24 |
| panda | rlandy: ok, wee need to push the change to move partes to config, but we're able to test without it for now, I creted test patches in testproject to test pairs of same job | 12:24 |
| panda | rlandy: all jobs, because periodic depend on the base rdo jobs | 12:24 |
| panda | rlandy: and rdo-jobs jobs are not included in rdo-infra/config repo | 12:25 |
| rlandy | panda: even check jobs that run on rdocloud - ovb? | 12:25 |
| sshnaidm | bogdando, ok, let's try to make sudo before | 12:25 |
| sshnaidm | bogdando, wait.. but I already had passwordless sudo in this point | 12:25 |
| sshnaidm | bogdando, I think the problem is in "become" working with "local connection" | 12:26 |
| panda | rlandy: I still have incostistend ovb results, but yeah, check jobs have tripleo-ci-base-rdo parent, and also periodic should have tripleo-ci-base-rdo parents | 12:26 |
| rlandy | panda: why move one nodeset? | 12:27 |
| sshnaidm | weshay, https://review.rdoproject.org/r/#/c/16485/ | 12:28 |
| panda | rlandy: isn't it the only one we are using ? | 12:28 |
| rlandy | panda: for ovb | 12:28 |
| rlandy | for multinode and singlenode we use another set | 12:28 |
| panda | rlandy: ok then, we need to move all that the parents are using, there's no way to test that patch until it's merged, so we need extra checks and eyes on that. | 12:28 |
| rlandy | panda: let's bj for five minutes are the meeting | 12:29 |
| bogdando | sshnaidm: "become" working with "local connection" problem? omfg | 12:29 |
| * bogdando away in rtfms... | 12:29 | |
| rlandy | after | 12:29 |
| panda | rlandy: sure | 12:29 |
| weshay | quique|rover|lch, 1-1 | 12:29 |
| rlandy | panda: nvm - it's wednesday - no meeting | 12:29 |
| rlandy | I keep thinking it's monday | 12:29 |
| panda | rlandy: retrospective | 12:30 |
| *** udesale has joined #oooq | 12:30 | |
| rlandy | I am so mixed up :( | 12:30 |
| marios | panda: in half hour tho? | 12:30 |
| weshay | rlandy, that's because you've been sleeping in a hut | 12:30 |
| panda | marios: yes | 12:30 |
| panda | a pizza hut ? | 12:30 |
| rlandy | weshay: lol - these broken weeks are doing me in | 12:30 |
| panda | I'd be mixed up too | 12:31 |
| weshay | a pizza hut would be awesome | 12:31 |
| rlandy | panda: ok - retro is only in half an hour - can we meet now for 5 minutes? | 12:31 |
| panda | rlandy: ok, 5 minutes, then lunch for me, then retro | 12:32 |
| panda | rlandy: bj/gcerami | 12:33 |
| weshay | quique|rover|lch, chkumar|ruck any info on queens fs001 jobs timing out? | 12:38 |
| weshay | in the deploy step? | 12:38 |
| chkumar|ruck | weshay: I am seening something similar in fs002 and fs020 also http://logs.rdoproject.org/openstack-periodic-24hr/git.openstack.org/openstack-infra/tripleo-ci/master/legacy-periodic-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset020-queens/aebdd5c/logs/undercloud/home/zuul/overcloud_prep_images.log.txt.gz and | 12:41 |
| chkumar|ruck | http://logs.rdoproject.org/openstack-periodic-24hr/git.openstack.org/openstack-infra/tripleo-ci/master/legacy-periodic-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset002-queens-upload/fd3bef2/logs/undercloud/home/zuul/overcloud_prep_images.log.txt.gz | 12:41 |
| chkumar|ruck | it is coming in overcloud prep image step | 12:42 |
| weshay | chkumar|ruck, prep-images is not the same | 12:42 |
| weshay | as the deploy time out | 12:42 |
| * weshay looks | 12:42 | |
| weshay | saw this error last night w/ rdo folks | 12:42 |
| weshay | chkumar|ruck, ya.. prep-images on queens 002 and 20 | 12:43 |
| *** quique|rover|lch is now known as quiquell|rover | 12:44 | |
| weshay | and 001 | 12:44 |
| weshay | and 035 | 12:44 |
| weshay | chkumar|ruck, is there a bug on that for queens? | 12:44 |
| weshay | 4 jobs is a good enough pattern | 12:44 |
| quiquell|rover | weshay: About timeouts in the overcloud deploy we have this https://review.openstack.org/#/c/605377/ | 12:44 |
| quiquell|rover | weshay: It's master, but suppose it's needed for master | 12:45 |
| quiquell|rover | s/master/queens/ | 12:45 |
| weshay | quiquell|rover, 1-1? | 12:45 |
| chkumar|ruck | weshay: https://bugs.launchpad.net/tripleo/+bug/1792872 I am not sure this one is enough | 12:46 |
| openstack | Launchpad bug 1792872 in tripleo "[queens] overcloud prepare image failed by giving IronicAction.node.set_provision_state failed: 'NoneType' object has no attribute '__getitem_" [Critical,Incomplete] - Assigned to Quique Llorente (quiquell) | 12:46 |
| chkumar|ruck | weshay: if needed, we can reupdate this | 12:46 |
| weshay | chkumar|ruck, k.. thank you | 12:46 |
| weshay | chkumar|ruck, quiquell|rover I would guess we need something backported from rocky --> queens and pike | 12:47 |
| weshay | chkumar|ruck, see the same issue in pike | 12:47 |
| weshay | master and rocky are clean | 12:47 |
| quiquell|rover | weshay: this timeout is different | 12:48 |
| quiquell|rover | :-/ | 12:48 |
| chkumar|ruck | weshay: need to find out | 12:48 |
| weshay | quiquell|rover, chkumar|ruck so just make sure you guys can hand off to the next pair | 12:48 |
| chkumar|ruck | https://groups.google.com/forum/m/#!topic/ansible-project/ehrb6AEptzA -> molecule and ansible lint coming to ansible | 12:48 |
| hubbot1 | FAILING CHECK JOBS on stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-3nodes-multinode, legacy-tripleo-ci-centos-7-containers-multinode-upgrades-pike-branch, legacy-tripleo-ci-centos-7 (3 more messages) | 12:48 |
| weshay | kopecmartin, ^ | 12:48 |
| chkumar|ruck | time to introduce to tripleo also | 12:48 |
| quiquell|rover | chkumar|ruck: Let's create a sprint20 etherpad with the stuff we have so next ones can continue | 12:49 |
| quiquell|rover | weshay: ^ | 12:49 |
| weshay | +1 | 12:49 |
| chkumar|ruck | quiquell|rover: on it | 12:49 |
| weshay | quiquell|rover, please jump in my bluejeans | 12:49 |
| weshay | it's ur 1-1 time | 12:50 |
| quiquell|rover | Ok | 12:50 |
| chkumar|ruck | quiquell|rover: weshay https://review.rdoproject.org/etherpad/p/ruckrover-sprint20 | 12:53 |
| rfolco | all: https://trello.com/b/0VFswmht/rdo-infra-retrospective?menu=filter&filter=label:Sprint%2019 | 12:54 |
| rfolco | pls write down your retro thoughts / cards | 12:55 |
| *** jfrancoa has quit IRC | 12:57 | |
| *** jfrancoa has joined #oooq | 12:58 | |
| *** skramaja has quit IRC | 12:59 | |
| arxcruz | rfolco: what's the bj ? | 13:00 |
| rfolco | same... matt's | 13:01 |
| marios | weshay: connection issues reconnecting | 13:01 |
| weshay | quiquell|rover, chkumar|ruck panda mtg | 13:02 |
| weshay | ssbarnea, ^ | 13:02 |
| weshay | arxcruz, ^ | 13:02 |
| panda | rfolco: we'll never abandon your patch, it contains the translations of all legacy jobs. | 13:03 |
| weshay | chkumar|ruck, quiquell|rover brains fried? | 13:04 |
| weshay | https://bluejeans.com/7050859455/ | 13:04 |
| bogdando | sshnaidm: so, please double check with sudo su - stack if it really has sudo | 13:06 |
| bogdando | I see no that error anymore, but I have to add stack to wheel group | 13:07 |
| bogdando | maybe it's my env specific | 13:07 |
| sshnaidm | bogdando, ok, will run it again | 13:07 |
| *** sshnaidm is now known as sshnaidm|mtg | 13:07 | |
| sshnaidm|mtg | bogdando, you need to ensure though that libvirt jobs don't fail, and they do | 13:12 |
| bogdando | got logs? | 13:12 |
| sshnaidm|mtg | bogdando, in your patch | 13:13 |
| sshnaidm|mtg | bogdando, I mean libvirt jobs that ran for your patch | 13:13 |
| bogdando | oh... ok, looking | 13:13 |
| bogdando | yeah, got it | 13:13 |
| marios | rfolco: https://www.youtube.com/watch?v=X21mJh6j9i4 never say no to panda | 13:16 |
| bogdando | sshnaidm|mtg: do you mean tripleo-quickstart-gate-master-delorean-quick-basic, tripleo-quickstart-gate-ocata-delorean-quick-basic libvirt jobs? | 13:26 |
| *** chkumar|ruck is now known as chandankumar | 13:26 | |
| bogdando | those are passing | 13:26 |
| *** vinaykns has joined #oooq | 13:27 | |
| *** agopi has joined #oooq | 13:27 | |
| *** panda has quit IRC | 13:32 | |
| *** dtantsur|brb is now known as dtantsur | 13:33 | |
| sshnaidm|mtg | bogdando, they failed in https://review.openstack.org/#/c/576746/ | 13:56 |
| bogdando | sshnaidm|mtg: looks green atm | 13:57 |
| sshnaidm|mtg | bogdando, ok, cool | 13:57 |
| *** udesale has quit IRC | 14:15 | |
| *** panda has joined #oooq | 14:21 | |
| sshnaidm|mtg | bogdando, tried and still fails, I can give you access to this environment to see if you want. Just send me your key | 14:42 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (3 more messages) | 14:48 |
| quiquell|rover | jaosorior: +w timeout fix https://review.openstack.org/#/c/605377/ | 15:00 |
| ssbarnea | jaosorior: quiquell|rover : this should be good to go in https://review.openstack.org/#/c/602015/ | 15:04 |
| jaosorior | quiquell|rover: can you ask the infra folks to priotize https://review.openstack.org/#/c/605377/ in the zuul queue? | 15:05 |
| jaosorior | the magic incantation is to askf or an "infra-root" | 15:05 |
| marios | panda: o/ add me to taiga please thanks | 15:06 |
| rlandy | panda: https://review.rdoproject.org/r/#/c/16412/ says nodeset is already defined - removing that | 15:10 |
| rlandy | already defined in rdo-jobs | 15:10 |
| rlandy | not sure if it's picking it up there | 15:11 |
| rlandy | in such a case, we could leave the nodesets there | 15:11 |
| quiquell|rover | jaosorior: I have a tattoo with that after this ruck/rovering :-) | 15:11 |
| rlandy | panda: so either we leave the nodesets all in rdo-jobs or move them all | 15:14 |
| ssbarnea | if i remember well the only trick to priorietize was to abandon changes from the queue and restore them after the event. | 15:15 |
| *** chandankumar is now known as chkumar|off | 15:26 | |
| *** quiquell|rover is now known as quique|rover|off | 15:27 | |
| *** quique|rover|off is now known as quique|off | 15:28 | |
| *** kopecmartin is now known as kopecmartin|ruck | 15:28 | |
| panda | rfolco: mercy! next time a break | 15:28 |
| sshnaidm|mtg | weshay, panda so we choose work more close with rdo-infra, but more far from rest of tripleo? | 15:29 |
| *** sshnaidm|mtg is now known as sshnaidm | 15:29 | |
| *** marios|rover has joined #oooq | 15:29 | |
| rfolco | panda, :) | 15:29 |
| rlandy | sshnaidm: that was exactly my concern | 15:30 |
| rlandy | exchanging one problem for another | 15:30 |
| sshnaidm | rlandy, and not sure it worth it | 15:30 |
| *** dsneddon has joined #oooq | 15:31 | |
| bogdando | ok can we at least let https://review.openstack.org/#/c/595566/ in | 15:32 |
| *** jtomasek has quit IRC | 15:34 | |
| *** sanjayu_ has quit IRC | 15:35 | |
| weshay | sshnaidm, I made the call here if you want to chat about it | 15:36 |
| sshnaidm | weshay, where? | 15:37 |
| weshay | sshnaidm, https://bluejeans.com/u/whayutin/ | 15:37 |
| *** amoralej has joined #oooq | 15:38 | |
| panda | rlandy: mmhh, so we are including the ndesets from rdo-jobs, but not the base jobs, mmhhh. In any case, since the periodic base will be there anyway, it makes sens t ohave all the bases in one single point | 15:39 |
| rlandy | panda: ok - then we will need a depends-on job that removes those nodesets from rdo-projects | 15:43 |
| rlandy | can commit that | 15:43 |
| rlandy | panda: secondly - to clean up the card, | 15:43 |
| rlandy | I am going to strike out the reviews I have other than the WIP one. | 15:44 |
| rlandy | and add your reviews to the list | 15:44 |
| rlandy | as well as the test reviews - marked as test reviews | 15:44 |
| *** jtomasek has joined #oooq | 15:44 | |
| panda | rlandy: do you want to try and create the card in taiga ? I can help in half hour | 15:51 |
| rlandy | panda: good idea. I'll try that once I get this next review in | 15:52 |
| panda | ok | 15:52 |
| rlandy | panda: in the main project - not the test one? | 15:52 |
| *** panda is now known as panda|bbl | 15:52 | |
| panda|bbl | rlandy: you can try in the test one if you want a place where you can screw up things | 15:52 |
| panda|bbl | I'll be here again in half hour and we can put it in the real project | 15:53 |
| * rlandy likes to live in places "where you can screw up things" | 15:53 | |
| rlandy | cool | 15:53 |
| *** holser_ has quit IRC | 15:54 | |
| *** jfrancoa has quit IRC | 15:54 | |
| *** jfrancoa has joined #oooq | 15:56 | |
| weshay | rlandy, panda|bbl https://review.rdoproject.org/r/#/c/16306 | 16:05 |
| rascasoft | hey rlandy, kopecmartin|ruck I see this conditional for containers in quickstart: when: release in ['ocata', 'pike', 'queens', 'rocky'] | 16:05 |
| rascasoft | question is why we're not considering master? | 16:06 |
| rlandy | weshay: need me to workflow that or do you have permissions there? | 16:08 |
| weshay | I don't have perm | 16:08 |
| rlandy | rascasoft: can you link the exact line - master might have a different workflow | 16:08 |
| *** dtantsur is now known as dtantsur|afk | 16:08 | |
| rlandy | done | 16:09 |
| rascasoft | rlandy, overcloud-prep-containers/tasks/main.yml | 16:09 |
| rascasoft | rlandy, https://github.com/openstack/tripleo-quickstart-extras/blob/master/roles/overcloud-prep-containers/tasks/main.yml#L6 | 16:09 |
| rascasoft | (and also line 11 of course) | 16:09 |
| *** fhubik|brb has quit IRC | 16:15 | |
| sshnaidm | bogdando, and what do you override it with? https://review.openstack.org/#/c/595566/5 | 16:17 |
| bogdando | sshnaidm: that's my secret :) | 16:18 |
| bogdando | jokes off, see the comments | 16:18 |
| bogdando | there is even benchmark results | 16:18 |
| sshnaidm | bogdando, which comment? | 16:19 |
| bogdando | sshnaidm: Patch Set 2, 23. Aug 15:01 | 16:20 |
| *** kopecmartin|ruck is now known as kopecmartin|off | 16:20 | |
| rlandy | rascasoft: hmm ... looking at the full review https://github.com/openstack/tripleo-quickstart-extras/commit/c4ec5a3b9fbd833a3384e062022d95c4e2364d28 | 16:20 |
| rlandy | containers-prepare-parameter.yaml for master | 16:21 |
| bogdando | there is no good value to override it, AFAIK, so couldn't think of anythin better than that, sshnaidm | 16:21 |
| bogdando | but that's still opted out for anyone how to use that option... | 16:21 |
| rlandy | panda|bbl: no good moving the nodesets | 16:21 |
| bogdando | defaults just works, x4 times slower tho | 16:21 |
| rlandy | I think we should just leave them in rdo-jobs | 16:21 |
| sshnaidm | bogdando, with LIBGUESTFS_BACKEND_SETTINGS=network_bridge=virbr0 ? | 16:22 |
| bogdando | yupp | 16:22 |
| bogdando | that's my override | 16:22 |
| sshnaidm | bogdando, maybe it's worth to mention this in comment in code, so that people will enjoy it too :) | 16:23 |
| bogdando | okaay | 16:23 |
| bogdando | in a follow up ok? let's save some CI resources and wes's time :D | 16:23 |
| *** gkadam has quit IRC | 16:25 | |
| sshnaidm | bogdando, who cares about wes' time.. | 16:27 |
| sshnaidm | :D | 16:27 |
| sshnaidm | bogdando, also you need to override LIBGUESTFS_BACKEND_SETTINGS by something meaningful, if you just set LIBGUESTFS_BACKEND_SETTINGS= it won't work, still will use force_tcg | 16:30 |
| sshnaidm | bogdando, maybe we need an option not to set it at all | 16:31 |
| sshnaidm | bogdando, for following up patch | 16:32 |
| *** panda|bbl is now known as panda | 16:35 | |
| panda | rlandy: what's the error you see ? | 16:35 |
| rlandy | panda: see the errors in the various patchsets | 16:35 |
| rlandy | the last patch set puts back your change | 16:36 |
| rlandy | but I think it's better to keep the nodesets in one place | 16:36 |
| rlandy | so remove tripleo-rdo-nodesets.yaml | 16:36 |
| rlandy | and add this dep https://review.rdoproject.org/r/#/c/16492/1/zuul.d/nodesets.yaml | 16:36 |
| rascasoft | rlandy, precious as usual, thanks I identified my problem: I was overriding container_args | 16:37 |
| rlandy | if zuul does not object | 16:37 |
| *** trown is now known as trown|lunch | 16:40 | |
| rlandy | panda: agree? | 16:43 |
| *** agopi has left #oooq | 16:44 | |
| *** agopi has joined #oooq | 16:44 | |
| bogdando | sshnaidm: that way with virbr0 it also works | 16:45 |
| bogdando | https://review.openstack.org/#/c/605483/ | 16:45 |
| bogdando | weshay: hi. why I cannot add you to review of a patch? Are you hiding from us? :D | 16:46 |
| weshay | bogdando, heh.. I decided to stop killing my personal email | 16:46 |
| bogdando | :D | 16:46 |
| weshay | whayutin@redhat.com | 16:46 |
| bogdando | can I know the new target for my killer bombs? | 16:47 |
| *** sanjayu_ has joined #oooq | 16:48 | |
| bogdando | the address you posted, filters out a 2 values with none of them working :[ | 16:48 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (3 more messages) | 16:48 |
| panda | rlandy: agreed | 16:50 |
| rlandy | panda: ok - submitting that | 16:50 |
| rlandy | panda: also working on https://tree.taiga.io/project/gabrielecerami-testproject1/backlog - taiga card | 16:51 |
| rlandy | we can switch to the real project | 16:51 |
| rlandy | https://review.rdoproject.org/r/#/c/16412/ | 16:53 |
| rlandy | let's see what happens there | 16:53 |
| *** jfrancoa has quit IRC | 16:58 | |
| *** jfrancoa has joined #oooq | 16:58 | |
| *** jfrancoa has quit IRC | 17:02 | |
| rlandy | panda: https://tree.taiga.io/project/gabrielecerami-testproject1/us/13?no-milestone=1 | 17:12 |
| rlandy | pls review and move to main project of you agree | 17:12 |
| rlandy | weshay: ^^ fyi | 17:12 |
| * rlandy looks at OVB failure now | 17:12 | |
| *** bogdando has quit IRC | 17:25 | |
| *** trown|lunch is now known as trown | 17:44 | |
| rlandy | panda: what about the legacy-rdoinfo-tripleo jobs? | 17:53 |
| weshay | ssbarnea, do not use depends-on for patches on the same repo | 17:57 |
| panda | rlandy: not sure, marios translated them | 17:57 |
| weshay | sshnaidm, you need to rebase one on top of the other | 17:58 |
| rlandy | panda: ok - pls review the taiga card and I;ll look into adding the rdoinfo jobs | 17:58 |
| panda | rlandy: I'm reviewing | 17:58 |
| panda | rlandy: I've added fields and broken down the list in sub tasks. Thinking about what to do in this sprint | 17:59 |
| panda | rlandy: as a fist pass this sprint maybe it ok to just use taiga as trello replacement, without any of the additional features. | 18:00 |
| rook | rlandy: hey - setting up a existing openstack overcloud for vmbc did anyone ever get that done? | 18:00 |
| rlandy | panda: hmmm - I did n;t find a list feature to add reviews | 18:00 |
| rlandy | I think we should ask that at tomorrow's meeting | 18:01 |
| rlandy | rook: you mean to launch ovb on an existing overcloud? if so, no | 18:01 |
| rook | rlandy okay - i thought there was work to do this? | 18:02 |
| rlandy | rook: yeah - has not been a priority as of late though | 18:03 |
| rlandy | panda: where did the review list go? | 18:04 |
| rlandy | https://tree.taiga.io/project/gabrielecerami-testproject1/us/13?no-milestone=1 | 18:04 |
| rlandy | I see just test reviews | 18:04 |
| rlandy | https://tree.taiga.io/project/gabrielecerami-testproject1/task/14 | 18:05 |
| rlandy | oh I see what you did | 18:05 |
| panda | rlandy: in the tasks, I divided the card into sub task, some reviews are for the "change base" tasks | 18:05 |
| panda | rlandy: I'm just testing | 18:05 |
| rlandy | np | 18:05 |
| panda | I'm not sure this is the right approach, and not sure if it's the right approach for the first pass | 18:06 |
| rlandy | I;m looking into branch testing in testproject | 18:06 |
| panda | rlandy: last time I asked they said we could to by ourselve, but it's false I thin kwe need permission do add references before doing that | 18:06 |
| rlandy | panda: you mean there is no branch created yet? | 18:07 |
| rlandy | remotes/gerrit/liberty-patches | 18:07 |
| rlandy | remotes/gerrit/master | 18:07 |
| rlandy | remotes/gerrit/stable/liberty | 18:07 |
| rlandy | remotes/origin/HEAD -> origin/master | 18:07 |
| rlandy | remotes/origin/liberty-patches | 18:07 |
| rlandy | remotes/origin/master | 18:07 |
| rlandy | remotes/origin/stable/liberty | 18:07 |
| rlandy | ^^ shows this | 18:07 |
| panda | rlandy: yep, no branch | 18:08 |
| rlandy | k | 18:08 |
| panda | rlandy: https://review.rdoproject.org/r/16388 | 18:10 |
| panda | rlandy: that's what they meant | 18:11 |
| panda | rlandy: so the request is still there | 18:11 |
| panda | rlandy: nobody approved it yet | 18:11 |
| panda | oh wow | 18:11 |
| panda | it was created for gitnetics | 18:11 |
| panda | lol | 18:11 |
| rlandy | panda: doe it matter - will just ask #sf-dfg if we can merge that | 18:12 |
| panda | rlandy: merge that and maybe also add some more stables | 18:13 |
| panda | with horses too. | 18:13 |
| rlandy | and ponies | 18:16 |
| * rlandy wonders if that will work as a depends-on | 18:16 | |
| rlandy | or a rebase | 18:16 |
| rlandy | we only really need to test one branch | 18:17 |
| *** amoralej is now known as amoralej|off | 18:20 | |
| weshay | gitnetics | 18:21 |
| weshay | geez | 18:21 |
| weshay | blast from the past | 18:21 |
| weshay | rlandy, panda qq | 18:36 |
| weshay | rlandy, panda before we run https://github.com/openstack-infra/tripleo-ci/blob/master/toci_quickstart.sh#L99 | 18:37 |
| weshay | https://github.com/openstack-infra/tripleo-ci/blob/master/toci_quickstart.sh#L34 | 18:38 |
| weshay | don't we need to update the requirements file in tripleo-quickstart/requirements to point at /opt/stack/tripleo-quickstart-extras ? | 18:38 |
| rlandy | we do that somewhere else ... | 18:40 |
| weshay | ah I see that now | 18:41 |
| weshay | https://github.com/openstack-infra/tripleo-ci/blob/master/toci_quickstart.sh#L34 | 18:41 |
| weshay | ooops | 18:41 |
| weshay | git+file:///opt/stack/tripleo-quickstart-extras/#egg=tripleo-quickstart-extras | 18:41 |
| weshay | hrm | 18:42 |
| weshay | why is that not working | 18:42 |
| rlandy | are you testing through the reproducer? | 18:42 |
| weshay | yes | 18:42 |
| rlandy | so ... that's in ... | 18:43 |
| rlandy | https://github.com/openstack/tripleo-quickstart-extras/blob/master/roles/nodepool-setup/tasks/clone-ci-repos.yml#L32 | 18:44 |
| weshay | rlandy, ah cool | 18:45 |
| weshay | rlandy, my chanages to tqe are not getting dropped in | 18:45 |
| * weshay keeps looking | 18:45 | |
| rlandy | if changes are made to the setup, they need to be included before | 18:46 |
| weshay | rlandy, looking for changes to tqe/undercloud-setup | 18:46 |
| weshay | rlandy, the vxlan networking script is running for one node | 18:47 |
| weshay | standalone node | 18:47 |
| weshay | even though I have it turned off | 18:47 |
| weshay | werid | 18:47 |
| weshay | weir | 18:47 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (3 more messages) | 18:48 |
| rlandy | weshay: did you turn that off with --extra-vars? | 18:51 |
| rlandy | panda: well wrt stable branches patch ... tristanC is in Korea atm | 18:57 |
| weshay | rlandy, w/ a patch to the config | 19:02 |
| weshay | config/general_config/fs053 | 19:02 |
| weshay | config/general_config/fs052 | 19:02 |
| weshay | rather | 19:02 |
| rlandy | weshay: are we talking about your change that is not taking? | 19:11 |
| rlandy | which reproducer? libvirt? | 19:12 |
| rlandy | panda: taiga-pro, is it possible to move the card from testing to the real project? | 19:16 |
| *** toure is now known as toure|biab | 19:19 | |
| weshay | rlandy, it's something w/ how I have these reviews stacked | 19:31 |
| rlandy | weshay: are they all rebased? | 19:32 |
| weshay | rlandy, I tried... that.. do you start at the bottom or top? | 19:32 |
| weshay | bottom right | 19:32 |
| weshay | rlandy, http://paste.openstack.org/show/730959/ | 19:32 |
| * rlandy checks | 19:32 | |
| weshay | does not have the change present. .when I look w/ vi | 19:32 |
| * weshay tries rebase | 19:32 | |
| rlandy | if one is ahead of the other, the first will not take | 19:33 |
| weshay | rlandy, k.. trying agian | 19:34 |
| weshay | rlandy, thanks for the sanity check | 19:34 |
| rlandy | weshay: k - if it doesn't work, I'll try it | 19:34 |
| weshay | rlandy, looks like that should work. | 19:46 |
| weshay | thanks rlandy | 19:46 |
| rlandy | weshay: cool | 19:46 |
| *** sanjayu_ has quit IRC | 19:46 | |
| weshay | rlandy, one more qq | 19:46 |
| weshay | oh man | 19:46 |
| weshay | http://git.openstack.org/cgit/openstack/tripleo-quickstart-extras/tree/roles/undercloud-setup/tasks/main.yml#n21 | 19:47 |
| rlandy | toci_vxlan_networking set that to false | 19:48 |
| weshay | ah that is getting me | 19:48 |
| weshay | http://git.openstack.org/cgit/openstack-infra/tripleo-ci/tree/toci-quickstart/config/testenv/multinode.yml | 19:48 |
| rlandy | both | 19:48 |
| weshay | rlandy, so I have that in the fs config | 19:48 |
| weshay | but maybe that's not good enough | 19:48 |
| weshay | because it's in the env file | 19:49 |
| rlandy | use --extra-vars will overwrite your config | 19:49 |
| rlandy | like we do in the v3 stuff | 19:49 |
| rlandy | in the reproducer | 19:49 |
| rlandy | you have extra-vars var | 19:50 |
| rlandy | add it to that | 19:50 |
| weshay | k | 19:50 |
| rlandy | set both to false | 19:50 |
| weshay | trying | 19:50 |
| * weshay wonders how undercloud only works | 19:50 | |
| rlandy | in ci? | 19:50 |
| rlandy | it runs off v3 | 19:50 |
| rlandy | and we set that to false in v3 | 19:51 |
| rlandy | weshay: how much damage do you think I could do if I merged https://review.rdoproject.org/r/#/c/16388/? | 19:51 |
| weshay | rlandy, no.. locally | 19:51 |
| weshay | rlandy, I'm not seeing how that could damage anytihng | 19:52 |
| weshay | other than people remembering things from 3 years ago | 19:52 |
| weshay | like gitnetics and foreman | 19:53 |
| rlandy | we need to test a stable branch and apparently tristanC is in Korea | 19:53 |
| weshay | rlandy, go 4 it | 19:54 |
| weshay | rlandy, where do those -e vars get passed to toci? | 19:55 |
| weshay | in the env? | 19:56 |
| * rlandy gets | 19:56 | |
| rlandy | "env_vars_to_src.sh" | 19:57 |
| rlandy | source ~/env_vars_to_src.sh; ANSIBLE_FORCE_COLOR=true /opt/stack/tripleo-ci/toci_gate_test.sh | 19:57 |
| rlandy | export EXTRA_VARS="\$EXTRA_VARS --extra-vars dlrn_hash_tag=48a0d56e4ba547ab10a35888138370fb1ec74a97_31a62456 " | 19:58 |
| rlandy | ^^ goes into env_vars_to_src.sh | 19:58 |
| rlandy | weshay; make sense? | 19:58 |
| weshay | aye | 19:58 |
| weshay | thanks | 19:58 |
| rlandy | this is the old style | 19:58 |
| weshay | rlandy, so I need to create an env file for standalone in toci | 19:58 |
| weshay | I guess | 19:59 |
| weshay | ya | 19:59 |
| rlandy | yeah - preferably | 19:59 |
| rlandy | although with the new way, we don't run that step anyways | 19:59 |
| weshay | rlandy, those don't get pushed over | 20:00 |
| weshay | export EXTRA_VARS="$EXTRA_VARS --extra-vars dlrn_hash_tag=0d5b1d18e88ad27e87368c8844c325cbfe603fdb_e25bec61 " | 20:00 |
| weshay | bash reproducer-quickstart.sh -v true -w /var/tmp/test/ -l -e image_fetch_dir=/var/tmp/test/ -e vxlan_networking=false -e toci_vxlan_networking=false | 20:00 |
| weshay | it's ok for now | 20:00 |
| weshay | need to get this moved to over to use jinja anyhoooooo | 20:01 |
| *** dsneddon has quit IRC | 20:02 | |
| *** dsneddon has joined #oooq | 20:02 | |
| panda | rlandy: no, it's not even possible to duplicate a card inside the same project | 20:04 |
| rlandy | I want to move it to the real project | 20:05 |
| rlandy | from the test project | 20:05 |
| rlandy | panda: btw ... merging https://review.rdoproject.org/r/#/c/16388/ - got tired of waiting to test branches | 20:06 |
| *** agopi has quit IRC | 20:06 | |
| weshay | \-/ | 20:09 |
| weshay | halloween yay! | 20:09 |
| weshay | headless? | 20:09 |
| *** toure|biab is now known as toure | 20:14 | |
| rlandy | gerrit/stable/queens | 20:17 |
| rlandy | ok | 20:17 |
| rook | rlandy weshay question regarding oooq, can we point to a downstream repo to run against? getting asked to do some osp9/10 work. | 20:30 |
| rook | basically have our perfci deploy osp9 run a set of tests, nuke, deploy osp10 rerun same set of tets | 20:31 |
| rook | tests* | 20:31 |
| weshay | rook, ya.. in theory that should work we have branched copies of tq and tqe | 20:31 |
| weshay | because the upstream loves to remove support for the older versions | 20:31 |
| rook | weshay yeah :/ the hope is agopi can kick this off -- but we were not sure if there was some extra vars he would need to pass | 20:32 |
| weshay | rook, note.. these are not tested much | 20:32 |
| rook | to tell oooq that this is a downstream | 20:32 |
| rook | :/ | 20:32 |
| rook | ok | 20:32 |
| weshay | https://code.engineering.redhat.com/gerrit/gitweb?p=tripleo-quickstart.git | 20:32 |
| weshay | https://code.engineering.redhat.com/gerrit/gitweb?p=tripleo-quickstart-extras.git | 20:32 |
| weshay | note the branches | 20:32 |
| rook | weshay: ack - thanks! | 20:34 |
| rook | i wonder how much pain this will be | 20:34 |
| weshay | rook, more people, more hardware, more time and this could be more sane | 20:34 |
| rook | weshay: i tell management that all the time | 20:34 |
| weshay | rook, who is asking you? | 20:34 |
| rook | and get laughed at | 20:35 |
| weshay | for it | 20:35 |
| rook | customer | 20:35 |
| weshay | rook, ya.. laugh right back at them | 20:35 |
| weshay | doh | 20:35 |
| rlandy | panda: did you notice this tox-doc failure before? https://review.rdoproject.org/r/#/c/16370/ | 20:41 |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (3 more messages) | 20:48 |
| *** agopi has joined #oooq | 20:56 | |
| *** trown is now known as trown|outtypewww | 20:59 | |
| agopi | weshay, looks like i don't ave access anymore on rhos-dev-jenkins to stop builds, should i open a ticket with ops team or is it possible for you me to give me back those privileges. | 21:01 |
| *** matbu has quit IRC | 21:10 | |
| *** matbu has joined #oooq | 21:11 | |
| panda | rlandy: yes, I didn't worry about it too much | 21:12 |
| *** panda is now known as panda|off | 21:13 | |
| agopi | weshay, I see stable/rhos-[7-14] except for stable/rhos-9 | 21:21 |
| agopi | https://code.engineering.redhat.com/gerrit/gitweb?p=tripleo-quickstart.git | 21:21 |
| agopi | ^ | 21:21 |
| agopi | nvm my bad | 21:21 |
| agopi | i see it | 21:21 |
| agopi | ugh | 21:21 |
| agopi | thanks weshay | 21:21 |
| agopi | weshay++ | 21:21 |
| hubbot1 | agopi: weshay's karma is now 10 | 21:21 |
| *** toure is now known as toure|gone | 21:48 | |
| rlandy | weshay: the rdoinfo jobs - where are they used? https://github.com/rdo-infra/rdo-jobs/blob/master/playbooks/legacy/rdoinfo-tripleo-master-testing-centos-7-multinode-1ctlr-featureset016/run.yaml | 22:05 |
| rlandy | rfolco: ^^ any idea? | 22:21 |
| rlandy | who is using rdoinfo | 22:22 |
| *** tosky has quit IRC | 22:37 | |
| hubbot1 | FAILING CHECK JOBS on master: tripleo-ci-centos-7-scenario000-multinode-oooq-container-upgrades, tripleo-ci-centos-7-scenario000-multinode-oooq-container-updates @ https://review.openstack.org/604298, stable/pike: legacy-tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001-pike, legacy-tripleo-ci-centos-7-ovb-1ctlr_1comp-featureset022-pike @ https://review.openstack.org/602248, master: tripleo-ci- (3 more messages) | 22:48 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!