| *** tpsilva has quit IRC | 00:00 | |
| *** jcoufal has quit IRC | 00:03 | |
| *** sthussey has quit IRC | 00:06 | |
| *** jlibosva has joined #openstack-infra | 00:09 | |
| *** agopi__ is now known as agopi | 00:12 | |
| *** jlibosva has quit IRC | 00:14 | |
| *** jcoufal has joined #openstack-infra | 00:15 | |
| *** spartakos has quit IRC | 00:19 | |
| *** jcoufal has quit IRC | 00:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack-infra/project-config master: Remove sphinx publishing for openstack-ansible-os_monasca-agent https://review.openstack.org/607740 | 00:25 |
|---|---|---|
| *** diablo_rojo has quit IRC | 00:26 | |
| *** jlibosva has joined #openstack-infra | 00:30 | |
| *** jamesmcarthur has joined #openstack-infra | 00:31 | |
| *** tristanC has quit IRC | 00:33 | |
| *** tristanC has joined #openstack-infra | 00:33 | |
| *** jlibosva has quit IRC | 00:35 | |
| *** rlandy|afk has quit IRC | 00:35 | |
| *** anteaya has quit IRC | 00:38 | |
| *** codrcodz has joined #openstack-infra | 00:39 | |
| *** codrcodz has left #openstack-infra | 00:41 | |
| *** longkb has joined #openstack-infra | 00:50 | |
| *** longkb has quit IRC | 00:51 | |
| *** jlibosva has joined #openstack-infra | 00:51 | |
| *** longkb has joined #openstack-infra | 00:52 | |
| *** dave-mccowan has quit IRC | 00:52 | |
| *** adriant has quit IRC | 00:54 | |
| *** gyee has quit IRC | 00:54 | |
| *** studarus has joined #openstack-infra | 00:54 | |
| *** felipemonteiro has quit IRC | 00:55 | |
| *** jlibosva has quit IRC | 00:57 | |
| *** adriant has joined #openstack-infra | 01:00 | |
| *** jtomasek has joined #openstack-infra | 01:03 | |
| *** jtomasek has quit IRC | 01:07 | |
| *** jlibosva has joined #openstack-infra | 01:13 | |
| *** felipemonteiro has joined #openstack-infra | 01:13 | |
| *** jlibosva has quit IRC | 01:17 | |
| *** felipemonteiro has quit IRC | 01:18 | |
| *** mrsoul has quit IRC | 01:20 | |
| *** jlibosva has joined #openstack-infra | 01:34 | |
| *** longkb has quit IRC | 01:37 | |
| *** longkb has joined #openstack-infra | 01:37 | |
| *** ijw has quit IRC | 01:37 | |
| *** jlibosva has quit IRC | 01:38 | |
| *** annp has joined #openstack-infra | 01:43 | |
| *** felipemonteiro has joined #openstack-infra | 01:49 | |
| *** jlibosva has joined #openstack-infra | 01:55 | |
| *** hongbin has joined #openstack-infra | 01:55 | |
| *** studarus has quit IRC | 02:00 | |
| *** jlibosva has quit IRC | 02:00 | |
| *** jtomasek has joined #openstack-infra | 02:02 | |
| *** jtomasek has quit IRC | 02:07 | |
| *** fuentess has quit IRC | 02:16 | |
| *** jlibosva has joined #openstack-infra | 02:16 | |
| *** jlibosva has quit IRC | 02:21 | |
| *** Bhujay has joined #openstack-infra | 02:24 | |
| *** jlibosva has joined #openstack-infra | 02:38 | |
| *** jlibosva has quit IRC | 02:42 | |
| *** ramishra has joined #openstack-infra | 02:45 | |
| *** roman_g_ has quit IRC | 02:47 | |
| *** yamamoto has quit IRC | 02:47 | |
| *** yamamoto has joined #openstack-infra | 02:47 | |
| *** jlibosva has joined #openstack-infra | 02:59 | |
| *** bhavikdbavishi has joined #openstack-infra | 03:00 | |
| *** bhavikdbavishi1 has joined #openstack-infra | 03:00 | |
| *** Bhujay has quit IRC | 03:03 | |
| *** jlibosva has quit IRC | 03:04 | |
| *** bhavikdbavishi has quit IRC | 03:04 | |
| *** bhavikdbavishi1 is now known as bhavikdbavishi | 03:04 | |
| *** udesale has joined #openstack-infra | 03:14 | |
| *** jlibosva has joined #openstack-infra | 03:20 | |
| *** jlibosva has quit IRC | 03:25 | |
| *** hongbin has quit IRC | 03:27 | |
| *** jamesmcarthur has quit IRC | 03:39 | |
| *** jlibosva has joined #openstack-infra | 03:41 | |
| *** felipemonteiro has quit IRC | 03:45 | |
| *** jamesmcarthur has joined #openstack-infra | 03:46 | |
| *** jlibosva has quit IRC | 03:46 | |
| *** ykarel|away has joined #openstack-infra | 04:02 | |
| *** jlibosva has joined #openstack-infra | 04:03 | |
| *** adam_g has joined #openstack-infra | 04:06 | |
| *** jlibosva has quit IRC | 04:08 | |
| *** jamesmcarthur has quit IRC | 04:08 | |
| *** armax_ has joined #openstack-infra | 04:14 | |
| *** kukacz has quit IRC | 04:14 | |
| *** ianychoi_ has joined #openstack-infra | 04:15 | |
| *** electrofelix has quit IRC | 04:15 | |
| *** kukacz has joined #openstack-infra | 04:16 | |
| *** armax has quit IRC | 04:17 | |
| *** armax_ is now known as armax | 04:17 | |
| *** ianychoi has quit IRC | 04:17 | |
| openstackgerrit | Merged openstack-infra/project-config master: Remove docs from os_monasca-agent https://review.openstack.org/607706 | 04:18 |
| *** jamesmcarthur has joined #openstack-infra | 04:20 | |
| *** jlibosva has joined #openstack-infra | 04:24 | |
| *** jamesmcarthur has quit IRC | 04:25 | |
| *** jamesmcarthur has joined #openstack-infra | 04:26 | |
| *** jlibosva has quit IRC | 04:28 | |
| *** jamesmcarthur has quit IRC | 04:31 | |
| *** jamesmcarthur has joined #openstack-infra | 04:31 | |
| *** jlibosva has joined #openstack-infra | 04:35 | |
| *** jlibosva has quit IRC | 04:39 | |
| *** tdasilva has quit IRC | 04:39 | |
| *** armax has quit IRC | 04:41 | |
| *** jamesmcarthur has quit IRC | 04:49 | |
| *** felipemonteiro has joined #openstack-infra | 04:50 | |
| *** jamesmcarthur has joined #openstack-infra | 04:52 | |
| *** jlibosva has joined #openstack-infra | 04:56 | |
| *** jlibosva has quit IRC | 05:01 | |
| *** jamesmcarthur has quit IRC | 05:02 | |
| *** udesale has quit IRC | 05:09 | |
| *** jamesmcarthur has joined #openstack-infra | 05:10 | |
| *** slaweq has joined #openstack-infra | 05:11 | |
| *** udesale has joined #openstack-infra | 05:15 | |
| *** slaweq has quit IRC | 05:16 | |
| *** jamesmcarthur has quit IRC | 05:16 | |
| *** jlibosva has joined #openstack-infra | 05:17 | |
| *** jamesmcarthur has joined #openstack-infra | 05:19 | |
| *** jlibosva has quit IRC | 05:22 | |
| *** ykarel|away is now known as ykarel | 05:23 | |
| *** jamesmcarthur has quit IRC | 05:24 | |
| *** ijw has joined #openstack-infra | 05:25 | |
| *** ijw is now known as ijw_ | 05:25 | |
| *** ijw_ is now known as ijw | 05:26 | |
| *** jlibosva has joined #openstack-infra | 05:39 | |
| *** jlibosva has quit IRC | 05:43 | |
| *** quiquell|off is now known as quiquell | 05:49 | |
| *** jamesmcarthur has joined #openstack-infra | 05:51 | |
| *** udesale has quit IRC | 05:53 | |
| *** jamesmcarthur has quit IRC | 05:55 | |
| *** udesale has joined #openstack-infra | 05:58 | |
| *** jlibosva has joined #openstack-infra | 06:00 | |
| *** jtomasek has joined #openstack-infra | 06:05 | |
| *** jlibosva has quit IRC | 06:07 | |
| *** jamesmcarthur has joined #openstack-infra | 06:24 | |
| *** jlibosva has joined #openstack-infra | 06:24 | |
| *** florianf|afk is now known as florianf | 06:26 | |
| *** jlibosva has quit IRC | 06:28 | |
| *** aojea has joined #openstack-infra | 06:31 | |
| *** rcernin has quit IRC | 06:38 | |
| *** rcernin has joined #openstack-infra | 06:38 | |
| *** pcaruana has joined #openstack-infra | 06:40 | |
| *** jlibosva has joined #openstack-infra | 06:45 | |
| *** dpawlik has joined #openstack-infra | 06:46 | |
| *** rockyg has joined #openstack-infra | 06:48 | |
| *** jlibosva has quit IRC | 06:50 | |
| *** rockyg_ has joined #openstack-infra | 06:50 | |
| *** rockyg_ has quit IRC | 06:51 | |
| *** rockyg has quit IRC | 06:53 | |
| *** felipemonteiro has quit IRC | 06:53 | |
| *** eernst has joined #openstack-infra | 06:54 | |
| *** shachar has joined #openstack-infra | 06:54 | |
| *** shachar has quit IRC | 06:54 | |
| *** slaweq has joined #openstack-infra | 06:57 | |
| *** jlibosva has joined #openstack-infra | 06:58 | |
| *** kjackal has quit IRC | 07:00 | |
| *** jlibosva has quit IRC | 07:03 | |
| *** eernst has quit IRC | 07:04 | |
| *** rcernin has quit IRC | 07:10 | |
| *** kjackal has joined #openstack-infra | 07:11 | |
| *** shachar has joined #openstack-infra | 07:15 | |
| *** shachar has quit IRC | 07:16 | |
| *** jlibosva has joined #openstack-infra | 07:18 | |
| *** jlibosva has quit IRC | 07:23 | |
| *** ykarel_ has joined #openstack-infra | 07:24 | |
| *** ykarel has quit IRC | 07:26 | |
| *** shardy has joined #openstack-infra | 07:26 | |
| *** ykarel__ has joined #openstack-infra | 07:28 | |
| *** ykarel_ has quit IRC | 07:30 | |
| *** udesale has quit IRC | 07:30 | |
| *** ykarel_ has joined #openstack-infra | 07:41 | |
| *** jklare has quit IRC | 07:43 | |
| *** ykarel__ has quit IRC | 07:44 | |
| *** tosky has joined #openstack-infra | 07:45 | |
| openstackgerrit | Merged openstack-infra/project-config master: Disable vexxhost mtl1 https://review.openstack.org/607737 | 07:45 |
| *** ykarel__ has joined #openstack-infra | 07:46 | |
| *** ykarel__ is now known as ykarel | 07:46 | |
| *** bhavikdbavishi has quit IRC | 07:46 | |
| *** ykarel has quit IRC | 07:47 | |
| *** ykarel_ has quit IRC | 07:48 | |
| ssbarnea|bkp2 | how do I add node (npm) tool to a job using openstack-tox-linters? | 07:52 |
| AJaeger | ssbarnea|bkp2: as rpm/deb package? | 07:52 |
| ssbarnea|bkp2 | AJaeger: yeah, in fact I do not care much how as long it becomes available when the job runs. | 07:52 |
| *** olivierb_ has joined #openstack-infra | 07:53 | |
| AJaeger | ssbarnea|bkp2: bindep.txt is the right file to add for those. | 07:53 |
| *** jlibosva has joined #openstack-infra | 07:53 | |
| ssbarnea|bkp2 | AJaeger: thanks! | 07:53 |
| *** udesale has joined #openstack-infra | 07:53 | |
| *** jpena|off is now known as jpena | 07:53 | |
| *** snapiri has quit IRC | 07:54 | |
| AJaeger | ssbarnea|bkp2: there's also https://zuul-ci.org/docs/zuul-jobs/jobs.html#job-nodejs-npm-run-lint | 07:54 |
| *** ykarel has joined #openstack-infra | 07:54 | |
| *** pcaruana has quit IRC | 07:55 | |
| *** amoralej|off is now known as amoralej | 07:55 | |
| ssbarnea|bkp2 | AJaeger: I wonder if it would not be a good idea to start asking suck questions on https://ask.openstack.org - to improve the KB. do you see it as a good idea or not really? Sadly our ask site is far less popular than stackoverflow. | 07:56 |
| *** pcaruana has joined #openstack-infra | 07:57 | |
| *** snapiri has joined #openstack-infra | 08:01 | |
| *** jklare has joined #openstack-infra | 08:09 | |
| *** ykarel_ has joined #openstack-infra | 08:12 | |
| AJaeger | ssbarnea|bkp2: that's worth a discussion for the broader team - yes, using that more would be great... | 08:13 |
| *** derekh has joined #openstack-infra | 08:13 | |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 08:14 |
| *** ykarel has quit IRC | 08:15 | |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 08:15 |
| *** ykarel__ has joined #openstack-infra | 08:20 | |
| *** aidin has joined #openstack-infra | 08:22 | |
| *** ykarel_ has quit IRC | 08:23 | |
| *** ykarel_ has joined #openstack-infra | 08:23 | |
| *** bhavikdbavishi has joined #openstack-infra | 08:25 | |
| *** gfidente has joined #openstack-infra | 08:25 | |
| *** ykarel__ has quit IRC | 08:26 | |
| ssbarnea|bkp2 | AJaeger: this would be a good subject for the weekly meeting, but I have no idea how to add something to the agenda, no etherpad link on https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Agenda_for_next_meeting | 08:26 |
| ssbarnea|bkp2 | back to npm subject, I got an interesting error after adding it to npm, something that makes be believe that the version installed is way too old: http://logs.openstack.org/04/607504/2/check/openstack-tox-linters/f0eeec6/job-output.txt.gz | 08:28 |
| ajo | hmm http://logs.openstack.org/61/607561/1/check/tox-py36/bc3166c/job-output.txt.gz#_2018-10-03_15_50_47_374384 | 08:29 |
| ajo | do I need to do anything special to enable a py36 job? | 08:29 |
| ajo | ubuntu-xenial | ERROR: py36: InterpreterNotFound: python3.6 | 08:29 |
| *** e0ne has joined #openstack-infra | 08:42 | |
| *** bhavikdbavishi has quit IRC | 08:44 | |
| *** panda|off is now known as panda | 08:44 | |
| *** ykarel__ has joined #openstack-infra | 08:45 | |
| *** udesale has quit IRC | 08:46 | |
| *** aidin has quit IRC | 08:46 | |
| *** ykarel_ has quit IRC | 08:48 | |
| *** ykarel_ has joined #openstack-infra | 08:49 | |
| *** udesale has joined #openstack-infra | 08:49 | |
| *** ykarel__ has quit IRC | 08:51 | |
| *** kashyap has joined #openstack-infra | 08:52 | |
| kashyap | Hey folks, is Storyboard still actively being developed? | 08:52 |
| kashyap | Is the plan still intact that all projects supposed to move to it sooner than later? | 08:53 |
| kashyap | I know some projects (Ironic?) have moved to it. | 08:53 |
| SotK | it is still actively developed yeah, and as far as I'm aware that is still the plan | 09:05 |
| cmurphy | ajo: you might want to switch to the openstack-tox-py36 job which uses the bionic nodeset, i don't know why the regular one tries to use xenial | 09:06 |
| *** ykarel_ has quit IRC | 09:07 | |
| *** ykarel_ has joined #openstack-infra | 09:07 | |
| kashyap | SotK: Do you know if the Storyboard folks intend the project to be used beyond OpenStack? E.g. QEMU upstream still uses LaunchPad, and if Storyboard suits their needs, they might consider it. | 09:08 |
| *** ykarel__ has joined #openstack-infra | 09:10 | |
| *** electrofelix has joined #openstack-infra | 09:12 | |
| *** ykarel_ has quit IRC | 09:13 | |
| *** ykarel has joined #openstack-infra | 09:18 | |
| *** ykarel__ has quit IRC | 09:19 | |
| frickler | cmurphy: ajo: tox-py36 is a generic zuul job not specifically tuned to our openstack infra. so it will need a special nodeset defined running bionic in order to work. using openstack-tox-py36 if possible seems indeed the best way forward | 09:20 |
| *** yamamoto has quit IRC | 09:22 | |
| *** yamamoto has joined #openstack-infra | 09:22 | |
| AJaeger | ssbarnea|bkp2: the link you give is a wiki, so just edit it... | 09:22 |
| AJaeger | frickler, if you have some review cycles, please look at : some smaller cleanups for review, please: https://review.openstack.org/#/c/607584/ https://review.openstack.org/#/c/607348/ https://review.openstack.org/#/c/607586/ | 09:23 |
| AJaeger | cmurphy, ajo, want to update the tox-py36 job? It lives in zuul-jobs repo... | 09:24 |
| cmurphy | AJaeger: from frickler's comment i'm not sure that would be appropriate, and neither it nor its ancestor jobs define any nodeset | 09:26 |
| AJaeger | cmurphy: you're right - it assumes a base job defined elsewhere - and that one uses xenial. So, let's leave it... | 09:27 |
| *** jesusaur has joined #openstack-infra | 09:30 | |
| *** ykarel has quit IRC | 09:32 | |
| *** ykarel has joined #openstack-infra | 09:32 | |
| ajo | cmurphy: oh, I can switch to openstack-tox-py36 if that works :-) | 09:33 |
| ajo | thanks frickler AJaeger cmurphy | 09:33 |
| AJaeger | ajo: best: use openstack-python36-jobs template | 09:34 |
| ajo | AJaeger: what does that template include? | 09:34 |
| ajo | I only have unit tests with py27,py35,py36 jobs in that repo | 09:34 |
| ajo | (so far) | 09:35 |
| ajo | btw, I also have tox-py27, tox-py35 | 09:35 |
| ajo | should I switch those too? | 09:35 |
| *** ykarel_ has joined #openstack-infra | 09:35 | |
| AJaeger | ajo: which repo? Have a change for me to review? | 09:37 |
| *** ykarel has quit IRC | 09:38 | |
| AJaeger | ajo: check https://docs.openstack.org/infra/openstack-zuul-jobs/project-templates.html#project_template-openstack-python-jobs for templates | 09:39 |
| *** ykarel_ is now known as ykarel | 09:43 | |
| *** Bhujay has joined #openstack-infra | 09:48 | |
| *** priteau has joined #openstack-infra | 09:48 | |
| SotK | kashyap: use outside OpenStack is very much welcomed :) | 09:49 |
| *** kjackal has quit IRC | 09:55 | |
| *** dtantsur|afk is now known as dtantsur | 09:55 | |
| *** longkb has quit IRC | 10:01 | |
| openstackgerrit | Tobias Henkel proposed openstack-infra/zuul master: Parallelize prime_installation_map https://review.openstack.org/607876 | 10:01 |
| *** udesale has quit IRC | 10:04 | |
| *** Bhujay has quit IRC | 10:06 | |
| *** kjackal has joined #openstack-infra | 10:08 | |
| *** udesale has joined #openstack-infra | 10:13 | |
| ajo | AJaeger: https://review.openstack.org/#/c/607561/1/.zuul.yaml | 10:15 |
| ajo | the change I was trying to do, it's rather simple project : | 10:16 |
| ajo | :) | 10:16 |
| AJaeger | ajo: ah, we have no templates for these jobs - only for the openstack-python ones. | 10:16 |
| ajo | AJaeger: it'd be adding a "template:" section before check and gate, right? | 10:17 |
| ajo | I suspect openstack-tox should also work | 10:17 |
| ajo | I'll try | 10:17 |
| AJaeger | ajo: sure, it will work just fine, the one change is that it allows to use constraints - but that's opt-in via your tox.ini file | 10:18 |
| ajo | :-) | 10:19 |
| ajo | AJaeger: publishing tarball is only for governance projects? | 10:19 |
| ajo | or also other projects can do it? | 10:19 |
| AJaeger | ajo: no, anybody can... | 10:21 |
| ajo | ohh | 10:21 |
| openstackgerrit | Stephen Finucane proposed openstack-dev/pbr master: Support wheel 0.32.0+ https://review.openstack.org/607894 | 10:21 |
| openstackgerrit | Stephen Finucane proposed openstack-dev/pbr master: tox: Suppress output https://review.openstack.org/607895 | 10:21 |
| ajo | AJaeger: I thought that was only for openstack governed projects :) | 10:22 |
| ajo | I'll try it | 10:22 |
| ajo | amoralej: ^ ;) | 10:22 |
| stephenfin | mtreinish: When you're about, can you take a look at https://review.openstack.org/607895 please? There's some stestr/subunit funkiness that I haven't been able to resolve | 10:22 |
| amoralej | ajo, cool | 10:23 |
| stephenfin | dhellmann, mordred: Need your input on https://review.openstack.org/607894 Wheel 0.32.0 introduces API breaks (Semantic versioning, anyone?) and I'm not sure if we need to provide wrappers for the older version as it only affects tests | 10:23 |
| stephenfin | AJaeger: That's the fix for pbr too, btw | 10:24 |
| dhellmann | stephenfin : ffs | 10:24 |
| * dhellmann goes to update the 100+ patches he just wrote to add a [wheel] section | 10:25 | |
| ajo | AJaeger: thanks, you're always very helpful, I submitted this one, let's see how it goes: https://review.openstack.org/#/c/607561/2/.zuul.yaml | 10:25 |
| *** roman_g has joined #openstack-infra | 10:25 | |
| stephenfin | dhellmann: To be clear, it's deprecated but not removed yet. Given that we don't version wheel, we might want to hold off on that | 10:25 |
| dhellmann | stephenfin : can we make pbr honor the old wheel section? | 10:26 |
| dhellmann | the *default* is to do versioned wheels, so I was updating setup.cfg everywhere to force universal | 10:26 |
| stephenfin | I haven't investigated that yet but...maybe? | 10:26 |
| AJaeger | stephenfin: thanks for looking into that! | 10:27 |
| dhellmann | wait, this just updates the tests? | 10:27 |
| AJaeger | ajo: that zuul.yaml looks fine... | 10:27 |
| stephenfin | dhellmann: Yeah, we don't seem to have any references to wheel outside of tests | 10:28 |
| stephenfin | like, any | 10:28 |
| stephenfin | wheel the library, that is. There are references to the 'bdist_wheel' section, funnily enough | 10:29 |
| stephenfin | AJaeger: np | 10:29 |
| stephenfin | dhellmann: https://github.com/pypa/wheel/commit/f7c9878712390414c03c64e9afa55ea4f30e965b | 10:31 |
| stephenfin | It's been there forever. I think a mass update is probably easiest, if you're doing it anyway | 10:31 |
| dhellmann | only a few patches in so far https://review.openstack.org/#/q/topic:python3-first+message:%22build+universal+wheels%22 | 10:33 |
| dhellmann | stephenfin : is this new version of wheel already in our CI system or are we managing it with constraints? | 10:33 |
| stephenfin | dhellmann: We're not managing it with constraints, no, but the version that introduces it is available all the way back in Trusty https://packages.ubuntu.com/trusty-updates/python-wheel | 10:34 |
| dhellmann | does it actually look in the new location yet? | 10:35 |
| dhellmann | ah, yeah, I guess by default | 10:36 |
| stephenfin | Yeah, the 'bdist_command' command is subclasses distutils.core.Command, so it'll read from the same setup.cfg|py section by default | 10:37 |
| stephenfin | s/is // | 10:37 |
| *** kopecmartin|ruck is now known as kopecmartin|scho | 10:38 | |
| *** slaweq_ has joined #openstack-infra | 10:38 | |
| *** udesale has quit IRC | 10:39 | |
| *** udesale has joined #openstack-infra | 10:40 | |
| *** slaweq has quit IRC | 10:40 | |
| dhellmann | AJaeger : what do you think about changing our packaging job to always build universal wheels by default? that would save a *lot* of patches to setup.cfg | 10:45 |
| *** yamamoto has quit IRC | 10:49 | |
| *** yamamoto has joined #openstack-infra | 10:49 | |
| AJaeger | would there be a way to disable these in case threre are problems? | 10:50 |
| AJaeger | otherwise, good idea... | 10:50 |
| dhellmann | AJaeger : the role that builds them is in zuul-jobs, so I was thinking of adding a variable like build_python_release_xargs and setting it to "" by default then overriding that in the definition of release-openstack-python3 in project-config | 10:51 |
| dhellmann | the only reason we'd have a problem is if someone writes a C extension | 10:52 |
| dhellmann | pure python packages can be universal with no problems afaik | 10:52 |
| AJaeger | ah, ok. Best discuss with rest of team, mordred, clarkb, and fungi should chime in. | 10:52 |
| *** yamamoto has quit IRC | 10:54 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/zuul-jobs master: support passing extra arguments to bdist_wheel in build-python-release https://review.openstack.org/607900 | 10:56 |
| *** timothyb89 has quit IRC | 10:56 | |
| *** slaweq__ has joined #openstack-infra | 10:56 | |
| *** slaweq_ has quit IRC | 10:57 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: always build universal wheels https://review.openstack.org/607902 | 10:57 |
| dhellmann | AJaeger, mordred , clarkb , fungi : ^^ | 10:58 |
| dhellmann | if a project doesn't want universal wheels, could they override that variable on release-openstack-python3 in their repo? | 11:01 |
| dhellmann | or I guess in project-config, since that's where we configure that job | 11:01 |
| AJaeger | dhellmann: not in their repo - but in project-config | 11:01 |
| *** dave-mccowan has joined #openstack-infra | 11:02 | |
| AJaeger | dhellmann: so, overwriting should work | 11:02 |
| *** pcaruana has quit IRC | 11:02 | |
| dhellmann | ok, good | 11:02 |
| *** pcaruana has joined #openstack-infra | 11:02 | |
| *** vivsoni has quit IRC | 11:03 | |
| *** jpena is now known as jpena|lunch | 11:03 | |
| AJaeger | dhellmann: FYI, I've been pushing charms python3-first changes, let's see whether tests pass and everything will work as expected... | 11:03 |
| *** amoralej is now known as amoralej|lunch | 11:03 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: always build universal wheels https://review.openstack.org/607902 | 11:04 |
| dhellmann | AJaeger : I'm starting to wonder if for some of these projects with broken stable branches we should just switch to no-op jobs so we can land the job settings | 11:04 |
| dhellmann | oh, though I guess the patch would be self-testing wouldn't it | 11:05 |
| dhellmann | so much for that early-morning idea | 11:05 |
| AJaeger | dhellmann: we would first need to remove jobs from projec-tconfig to allow doing noop-jobs. But yes, that's what I would propose for some of these... | 11:05 |
| *** janki has joined #openstack-infra | 11:06 | |
| dhellmann | I was confused and thought if project-config had noop-jobs we could land the patch to add the jobs, but of course zuul would run the tests on those jobs | 11:06 |
| AJaeger | dhellmann: ansible should be ready later today - do you have a script for these project-config rebases or do it manually? | 11:06 |
| dhellmann | I've been doing the rebases manually | 11:06 |
| AJaeger | dhellmann: yes, indeed - zuul would run the jobs | 11:06 |
| AJaeger | dhellmann: ok, then I'll do ansible rebase now | 11:06 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: remove job settings for OpenStackAnsible repositories https://review.openstack.org/597576 | 11:10 |
| AJaeger | dhellmann: want to WIP again? ^ | 11:10 |
| dhellmann | done | 11:11 |
| dhellmann | thanks for rebasing | 11:11 |
| AJaeger | np | 11:11 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/zuul-jobs master: support passing extra arguments to bdist_wheel in build-python-release https://review.openstack.org/607900 | 11:16 |
| panda | does import_playbook work on the executor ? | 11:30 |
| *** arxcruz has joined #openstack-infra | 11:38 | |
| *** yamamoto has joined #openstack-infra | 11:38 | |
| *** aidin has joined #openstack-infra | 11:42 | |
| *** tpsilva has joined #openstack-infra | 11:44 | |
| *** slagle has quit IRC | 11:51 | |
| AJaeger | dhellmann: https://review.openstack.org/607922 - 98k is pushing changes again ;( | 11:56 |
| dhellmann | sigh | 12:00 |
| AJaeger | dhellmann: you have permissions to abandon all of them? Nice ;) | 12:02 |
| dhellmann | apparently | 12:03 |
| dhellmann | maybe as a release manager? | 12:03 |
| AJaeger | yes! | 12:03 |
| *** rh-jelabarre has joined #openstack-infra | 12:04 | |
| *** jpena|lunch is now known as jpena | 12:04 | |
| *** jcoufal has joined #openstack-infra | 12:04 | |
| fungi | kashyap: as i understand, storyboard is already used by organizations other than the openstack community, but if, say, qemu wants to use it and has feature requests it would be helpful if they also had people interested in helping hack on realizing those | 12:06 |
| tosky | dhellmann, AJaeger: oh, can we ignore those "build universal wheel" changes for now? | 12:07 |
| *** vivsoni has joined #openstack-infra | 12:07 | |
| dhellmann | tosky : yes | 12:07 |
| tosky | thanks | 12:07 |
| fungi | dhellmann: baking universal wheel building into our jobs instead of setting that in projects means that if someone, say, uses pip to install an sdist they'll end up with a non-universal intermediate wheel. or really any wheel building they do locally for that matter will require passing nondefault options | 12:08 |
| fungi | and also, yes, it can pose a problem for anyone who wants to use those jobs to build wheels for non-pure-python libraries | 12:09 |
| dhellmann | I'm ok making that trade-off in exchange for not having to produce another 400+ patches to update all of the setup.cfg files we have | 12:09 |
| *** annp has quit IRC | 12:09 | |
| dhellmann | the patches I wrote in zuul-jobs and project-config make it default but overridable | 12:09 |
| dhellmann | zuul-jobs just adds a pass-through variable | 12:09 |
| fungi | just wondering what precedent that sets for other standard project boilerplate... should we be seeking to centralize more things? | 12:10 |
| dhellmann | yes | 12:10 |
| dhellmann | if anything, the work this cycle convinces me that we need to think about how to build all of this in a way that doesn't *require* us to touch every repo | 12:10 |
| fungi | monorepo (only half joking) | 12:10 |
| dhellmann | I'm almost there | 12:10 |
| fungi | centralizing aspects of individual projects' repos creates central authorities who we have trouble scaling (or even staffing), while distributing that of course puts it in the control of people who may not care one way or the other. definitely interesting trade-offs | 12:12 |
| *** trown|outtypewww is now known as trown | 12:13 | |
| dhellmann | I think the answer is to provide lots of parameters with good defaults | 12:13 |
| fungi | if the problem is that teams take forever to merge the changes for such standardized content, it's also possible to consider administrative review overrides | 12:13 |
| *** boden has joined #openstack-infra | 12:13 | |
| dhellmann | the problem is that it's 1 patch, per branch, per repo | 12:14 |
| dhellmann | and someone has to write all of those | 12:14 |
| *** rfolco has joined #openstack-infra | 12:14 | |
| dhellmann | even scripting it, keeping up with them is just doing a bunch of extra work that takes up time we could spend doing more useful things | 12:14 |
| fungi | well, i'd argue we can ignore stable branches for something like universal wheels. we don't have to fix all mistakes in historical releases | 12:14 |
| dhellmann | nothing prevents teams from updating their setup.cfg, and we can document that they should, but we know we want universal wheels so we should just build them | 12:14 |
| dhellmann | if we ignore the stable branches, then any new releases would be creating python3 wheels, not python2 wheels like previous releases | 12:15 |
| dhellmann | even for branches that do not support python3 | 12:15 |
| fungi | we have those? | 12:15 |
| dhellmann | yes | 12:15 |
| fungi | but sure, in this particular case i don't feel too strongly about it | 12:15 |
| fungi | so, er, why would we want universal wheels for branches/projects that lack python 2 support? that seems broken/risky | 12:16 |
| fungi | those should produce python3-only wheels right? | 12:16 |
| *** rlandy has joined #openstack-infra | 12:18 | |
| dhellmann | either version of python can install a universal wheel | 12:19 |
| dhellmann | if we have only a py3 wheel, then a py2 user can't install that and goes through the sdist | 12:19 |
| dhellmann | in the future when we have completely dropped python 2 we can remove that --universal flag if we decide that's what we want to do | 12:19 |
| *** ssbarnea_ has joined #openstack-infra | 12:23 | |
| fungi | thinking through it, i suppose a universal wheel for a project which only supports python 3 isn't especially dangerous. though i do still think we should also authorize a central authority in the project with the ability to override team reviews on such changes to standard project configuration so that it doesn't sit around ignored in random corners of the community for ages | 12:24 |
| *** ssbarnea|bkp2 has quit IRC | 12:25 | |
| dhellmann | I'm not sure what that means | 12:25 |
| fungi | (says someone who still has a boilerplate change proposed to openstack/compass-specs ignored for nearly 4 years now) | 12:25 |
| dhellmann | that's impressive | 12:25 |
| *** amoralej|lunch is now known as amoralej | 12:26 | |
| fungi | that's just the oldest one. i have at least a few dozen that are 2-4 years old | 12:26 |
| dhellmann | are you suggesting that we give someone +2a rights in repos to push these sorts of changes through? or something else? | 12:26 |
| fungi | yes. when we were doing mass zuul changes during the v3 migration and had some things which needed to merge to random projects in a time-sensitive fashion, we were investigating just using an administrative override to approve them | 12:27 |
| fungi | for mechanically-generated changes which were passing tests | 12:27 |
| *** aidin has quit IRC | 12:27 | |
| dhellmann | ok, I can see that being useful. I still think we'll end up with a better system if we design it to not need to make those sorts of changes on a regular basis. | 12:28 |
| fungi | gertty query the topic with positive verify, process flag those and batch approve them | 12:28 |
| dhellmann | though there will be some cases where touching each repo is the right way, like adding new python3X jobs | 12:28 |
| dhellmann | yeah | 12:29 |
| *** Seb-Solon has quit IRC | 12:31 | |
| *** quiquell is now known as quiquell|lunch | 12:33 | |
| *** tdasilva has joined #openstack-infra | 12:34 | |
| *** tosky__ has joined #openstack-infra | 12:37 | |
| *** tosky is now known as Guest8399 | 12:37 | |
| *** tosky__ is now known as tosky | 12:37 | |
| *** aidin has joined #openstack-infra | 12:39 | |
| hwoarang | fungi: hello again. could you hold a node again for 'bifrost-integration-tinyipa-opensuse-150' on https://review.openstack.org/#/c/599521/ ? same ssh pubkey again. Thank you | 12:39 |
| *** Guest8399 has quit IRC | 12:39 | |
| fungi | hwoarang: done, i've set zuul autohold --tenant openstack --project openstack/bifrost --job bifrost-integration-tinyipa-opensuse-150 --change 599521 --reason "fungi/hwoarang debugging" --count 1 | 12:40 |
| *** agopi has quit IRC | 12:41 | |
| hwoarang | ok rechecking the job | 12:41 |
| hwoarang | thank you | 12:41 |
| fungi | feel free to recheck and i'll add your ssh key once that completes | 12:41 |
| *** agopi has joined #openstack-infra | 12:41 | |
| cmurphy | I was trying to do some debugging and thought I could run a temporary job to play with the environment but I seem to have missed some step to get the job to actually run: https://review.openstack.org/#/c/607048/6/.zuul.yaml what did i do wrong? | 12:44 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 12:45 |
| cmurphy | of course i just found it... | 12:45 |
| *** agopi has quit IRC | 12:45 | |
| *** agopi has joined #openstack-infra | 12:45 | |
| *** kopecmartin|scho is now known as kopecmartin|ruck | 12:46 | |
| *** agopi has quit IRC | 12:47 | |
| *** agopi has joined #openstack-infra | 12:47 | |
| *** niedbalski has joined #openstack-infra | 12:47 | |
| *** slagle has joined #openstack-infra | 12:47 | |
| *** agopi has quit IRC | 12:48 | |
| *** agopi has joined #openstack-infra | 12:48 | |
| *** carl_cai has joined #openstack-infra | 12:49 | |
| Tengu | hello there! I'm wondering... the change 607130 is in the Gate for over 12h nows, it seems to start, does some of the validations while others are queued, then start all of them, and bam, back to "queued" for ALL the validations. | 12:49 |
| Tengu | is it an intended behavior? | 12:49 |
| AJaeger | Tengu: it is in the middle of a queue of changes. We test the complete set of repos in a well-defined state together. | 12:55 |
| Tengu | AJaeger: oh, ok. so it will restart until it's at the top? | 12:55 |
| AJaeger | If the changes in front of your change fail, we need to restart the testing for your change... | 12:55 |
| Tengu | :'( | 12:55 |
| AJaeger | Tengu: it will only restart if chanes in front of it fail - and those the "global state" changes | 12:55 |
| Tengu | ok | 12:56 |
| Tengu | well. eventually it will succeed. if no time_out | 12:56 |
| AJaeger | Tengu: we try to test many in parallel - but if then something fails, we have to start... | 12:56 |
| AJaeger | Exactly - so, please have patience - and help getting stable code in to fix those races that lead to failures and timeouts right now | 12:56 |
| *** sthussey has joined #openstack-infra | 12:56 | |
| *** bobh has joined #openstack-infra | 12:59 | |
| *** ansmith has joined #openstack-infra | 13:00 | |
| Tengu | I'm trying to help finding the cause, but CI isn't my "thing" :( | 13:00 |
| *** quiquell|lunch is now known as quiquell | 13:05 | |
| *** aidin has quit IRC | 13:11 | |
| AJaeger | fungi, config-core, please review https://review.openstack.org/#/c/607582/ https://review.openstack.org/#/c/607584/ and https://review.openstack.org/#/c/607348/ - all small cleanups. | 13:20 |
| *** slagle has quit IRC | 13:22 | |
| *** kgiusti has joined #openstack-infra | 13:22 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: remove job settings for OpenStack Charms repositories https://review.openstack.org/601688 | 13:27 |
| AJaeger | dhellmann: want to WIP? ^ | 13:29 |
| dhellmann | done | 13:29 |
| AJaeger | thanks. charms and ansible might make it today - let's see... | 13:30 |
| lucasagomes | dhellmann, thanks for rebasing https://review.openstack.org/#/c/602609/ onto the validation patch! It seems that it's working now | 13:34 |
| *** Emine has joined #openstack-infra | 13:34 | |
| *** ykarel_ has joined #openstack-infra | 13:36 | |
| *** slaweq__ is now known as slaweq | 13:36 | |
| *** ykarel has quit IRC | 13:38 | |
| *** slagle has joined #openstack-infra | 13:41 | |
| *** eharney has joined #openstack-infra | 13:43 | |
| *** Emine has quit IRC | 13:44 | |
| *** agopi has quit IRC | 13:44 | |
| *** dpawlik has quit IRC | 13:47 | |
| *** pcaruana has quit IRC | 13:50 | |
| *** ykarel_ is now known as ykarel | 13:50 | |
| *** oanson has quit IRC | 13:51 | |
| *** fuentess has joined #openstack-infra | 13:57 | |
| *** priteau has quit IRC | 13:58 | |
| *** priteau has joined #openstack-infra | 13:59 | |
| AJaeger | config-core, please review also: two new GPU nodes: https://review.openstack.org/607686 ; moving of roles https://review.openstack.org/525760 and neutron grafana update: https://review.openstack.org/607586 | 14:00 |
| *** agopi has joined #openstack-infra | 14:02 | |
| *** njohnston has quit IRC | 14:02 | |
| *** njohnston has joined #openstack-infra | 14:03 | |
| openstackgerrit | Merged openstack-infra/project-config master: Remove noop jobs for ara-{plugins,clients,server},os-log-merger https://review.openstack.org/607584 | 14:03 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: Handle all schedule in a single table https://review.openstack.org/607307 | 14:07 |
| openstackgerrit | Thierry Carrez proposed openstack/ptgbot master: No longer load schedule from config https://review.openstack.org/607308 | 14:07 |
| *** hamzy has quit IRC | 14:10 | |
| panda | does import_playbook work on the executor ? | 14:14 |
| clarkb | panda: I'm not sure. You should be able to try it and it will tell you. Otherwise you'll need to check zuul docs problably | 14:15 |
| *** jaosorior has quit IRC | 14:16 | |
| fuentess | clarkb: hi, could you help me hold a Fedora 28 machine? I want to check if I can reproduce the issues that I see in the logs | 14:20 |
| clarkb | fuentess: ya | 14:20 |
| fuentess | clarkb: thanks | 14:21 |
| clarkb | fuentess: the next kata-runsh-fedora-28 job that fails and runs against the proxy repo should be held | 14:22 |
| fuentess | cool, let me trigger a job | 14:23 |
| *** shachar has joined #openstack-infra | 14:27 | |
| *** tosky__ has joined #openstack-infra | 14:27 | |
| *** tosky has quit IRC | 14:28 | |
| *** tosky__ is now known as tosky | 14:28 | |
| *** Bhujay has joined #openstack-infra | 14:28 | |
| *** snapiri has quit IRC | 14:29 | |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Remove some legacy-manila jobs https://review.openstack.org/607348 | 14:29 |
| *** kukacz has quit IRC | 14:30 | |
| *** kukacz has joined #openstack-infra | 14:30 | |
| *** ykarel is now known as ykarel|afk | 14:32 | |
| openstackgerrit | Merged openstack-infra/project-config master: Add GPU instances to CI infrastructure https://review.openstack.org/607686 | 14:33 |
| openstackgerrit | Merged openstack-infra/project-config master: Move docs/cover job for monasca-{transform,ui} in-repo https://review.openstack.org/607582 | 14:33 |
| ajo | AJaeger: https://tarballs.openstack.org/os-log-merger/ it worked, this publishes the branch, | 14:38 |
| ajo | AJaeger: how can I publish the tags ? | 14:38 |
| AJaeger | ajo https://docs.openstack.org/infra/manual/drivers.html#tagging-a-release | 14:39 |
| ajo | AJaeger: but that will trigger a tar.gz release? | 14:39 |
| ajo | let me try, may be last failed because tox -evenv was not working | 14:39 |
| AJaeger | ajo: wait... | 14:39 |
| AJaeger | ajo: what do you mean with "publish the tags" | 14:39 |
| ajo | AJaeger: I mean, have a .tar.gz for the -x.x.x version | 14:40 |
| ajo | when I push a tag | 14:40 |
| ajo | I see that other projects have that | 14:40 |
| ajo | AJaeger: for example: https://tarballs.openstack.org/os-testr/ | 14:40 |
| AJaeger | ajo: those get pushed to pypi... | 14:40 |
| AJaeger | ajo: need to dig into jobs to see whether we still push those tarballs there in addition... | 14:41 |
| ajo | AJaeger: my purpose is making the tar.gzs available over tarballs.openstack.org so packagers can pick it up | 14:41 |
| ajo | AJaeger: let me try to do a new tag | 14:41 |
| *** jamesmcarthur has quit IRC | 14:41 | |
| *** jamesmcarthur has joined #openstack-infra | 14:42 | |
| *** armax has joined #openstack-infra | 14:43 | |
| AJaeger | ajo: Looks like we publish to both pypi and tarballs. | 14:44 |
| AJaeger | ajo: Remember: A tag once published cannot be deleted... | 14:44 |
| *** hamzy has joined #openstack-infra | 14:45 | |
| ajo | AJaeger: it triggered this: | 14:46 |
| ajo | openstack/os-log-merger | 14:46 |
| ajo | b4553ba | 14:46 |
| ajo | unknown | 14:46 |
| ajo | 1 min | 14:46 |
| ajo | release-openstack-pythonqueued | 14:46 |
| ajo | announce-releasequeued | 14:46 |
| ajo | propose-update-constraintsqueued | 14:46 |
| AJaeger | ;) | 14:46 |
| *** kopecmartin|ruck is now known as kopecmartin|scho | 14:47 | |
| AJaeger | ajo: if it fails, a mail is sent to http://lists.openstack.org/cgi-bin/mailman/listinfo/release-job-failures | 14:47 |
| ajo | ack thanks AJaeger | 14:47 |
| *** jchhatbar has joined #openstack-infra | 14:50 | |
| *** janki has quit IRC | 14:53 | |
| fuentess | clarkb: I see that the job already finished, do you still have my public key? or want me to send it? | 14:55 |
| clarkb | fuentess: I'll need it again | 14:55 |
| fuentess | clarkb: sent it privately | 14:56 |
| *** shachar is now known as snapiri | 14:57 | |
| clarkb | fuentess: root@38.108.68.10 | 14:57 |
| clarkb | fuentess: let me know if it doesn't work | 14:57 |
| fuentess | clarkb: thanks, I'm already in it | 14:57 |
| *** bobh has quit IRC | 14:58 | |
| *** e0ne has quit IRC | 14:58 | |
| *** _erlon_ has joined #openstack-infra | 14:59 | |
| *** Swami has joined #openstack-infra | 15:00 | |
| openstackgerrit | Merged openstack-infra/project-config master: Align neutron grafana with actual jobs being run https://review.openstack.org/607586 | 15:03 |
| *** bobh has joined #openstack-infra | 15:04 | |
| *** olivierb_ has quit IRC | 15:04 | |
| fungi | hwoarang: ssh root@104.239.140.126 | 15:05 |
| *** olivierb_ has joined #openstack-infra | 15:05 | |
| clarkb | fungi: we're the ssh connection dealers | 15:05 |
| hwoarang | thank you fungi | 15:08 |
| *** bobh has quit IRC | 15:10 | |
| *** e0ne has joined #openstack-infra | 15:13 | |
| fungi | indeed | 15:16 |
| *** quiquell is now known as quiquell|off | 15:18 | |
| *** sshnaidm is now known as sshnaidm|afk | 15:20 | |
| *** slagle has quit IRC | 15:25 | |
| *** olivierb_ has quit IRC | 15:29 | |
| *** olivierb_ has joined #openstack-infra | 15:29 | |
| openstackgerrit | Merged openstack-infra/openstack-zuul-jobs master: Move prepare-zanata-client to o-z-j https://review.openstack.org/525760 | 15:29 |
| *** agopi is now known as agopi|long_lunch | 15:31 | |
| *** sshnaidm|afk is now known as sshnaidm | 15:36 | |
| *** udesale has quit IRC | 15:36 | |
| openstackgerrit | Stephen Finucane proposed openstack-dev/pbr master: tox: Suppress output https://review.openstack.org/607895 | 15:36 |
| *** bobh has joined #openstack-infra | 15:37 | |
| *** hamzy_ has joined #openstack-infra | 15:42 | |
| *** bobh has quit IRC | 15:44 | |
| *** hamzy has quit IRC | 15:44 | |
| panda | clarkb: at this point I'm not sure about import_playbook, but include_role is giving problems for sure. using role: is still more reliable. What I've notice, it's that include_role return no output if some of the modules on the role fail. | 15:46 |
| *** jchhatbar has quit IRC | 15:46 | |
| panda | in this test, the change uses role: for a pre-run playbook, and include_role for the run playbook. http://logs.openstack.org/88/607288/2/check/tripleo-ci-centos-7-containers-multinode/d3f871e/job-output.txt.gz | 15:47 |
| *** e0ne has quit IRC | 15:47 | |
| panda | There is not output in the console about the role included with include_role | 15:47 |
| panda | execution stops, job fails. | 15:47 |
| openstackgerrit | Dean Troyer proposed openstack-infra/project-config master: Add publish jobs for stx-governance and per-project https://review.openstack.org/608003 | 15:47 |
| roman_g | Hello team. Which ansible version can I expect on Zuul nodes (assuming I use empty "nodes: []" set)? | 15:48 |
| panda | http://logs.openstack.org/88/607288/2/check/tripleo-ci-centos-7-containers-multinode/d3f871e/job-output.txt.gz#_2018-10-03_13_10_01_522132 | 15:48 |
| clarkb | roman_g: 2.5 | 15:49 |
| roman_g | clarkb: that's great. thank you! | 15:49 |
| *** ykarel|afk is now known as ykarel | 15:52 | |
| njohnston | So I noticed that some job data I expected to see in a graphite dashboard was not getting populated. So I decided to go check graphite to see if I could see the data there. It turns out that I am seeing almost no data points - not just for lines missing from grafana but also for lines present and populated in grafana. Is graphite.o.o still the right place to go to validate data? | 15:54 |
| *** jamespage has joined #openstack-infra | 15:54 | |
| njohnston | Example: neutron-functional job. Graphite shows this: https://i.postimg.cc/T2cwJTmj/Screen_Shot_2018-10-04_at_11.43.24_AM.png Grafana has data: http://grafana.openstack.org/d/Hj5IHcSmz/neutron-failure-rate?orgId=1&panelId=22&fullscreen | 15:54 |
| jamespage | dhellmann: hi - I just landed the last of the project-config migrations for the openstack charms so I think we're ready for the project-config change now | 15:54 |
| njohnston | srry that first sentence there should have said "expected to see in a *grafana* dashboard" | 15:55 |
| *** olivierb_ has quit IRC | 15:55 | |
| *** Bhujay has quit IRC | 15:55 | |
| njohnston | second link updated: http://grafana.openstack.org/d/Hj5IHcSmz/neutron-failure-rate?orgId=1&panelId=24&fullscreen | 15:57 |
| clarkb | njohnston: graphite.openstack.org runs the statsd daemon and stores all the data. It also runs graphite as a web api to that data, grafana is an alternate api to that data | 15:58 |
| clarkb | njohnston: the data source should be the same for both of them | 15:58 |
| *** gyee has joined #openstack-infra | 16:00 | |
| clarkb | njohnston: on the graphite side I think you may need to use a function like sum to create aggregated data like grafana does | 16:01 |
| clarkb | njohnston: hard to tell in the graphite screenshot but I am guessing it is plotting points along the x axis | 16:01 |
| openstackgerrit | Merged openstack-infra/project-config master: Add release-zuul-python job that deals with javascript https://review.openstack.org/607637 | 16:02 |
| njohnston | ok, I'll plot some using sum functions to show the discrepancy | 16:02 |
| *** trown is now known as trown|lunch | 16:04 | |
| njohnston | clarkb: OK, here is a comparison using openstack-neutron job openstack-tox-py27. Grafana run counts per day: http://grafana.openstack.org/d/Hj5IHcSmz/neutron-failure-rate?orgId=1&panelId=27&fullscreen shows counts per day for the last week summed by day | 16:07 |
| njohnston | clarkb: Here is the same data in graphite, not summed per day but there is no way it could come anywhere close to the grafana values: https://i.postimg.cc/Zq8qCmxM/Screen_Shot_2018-10-04_at_12.04.28_PM.png | 16:07 |
| *** aidin has joined #openstack-infra | 16:10 | |
| clarkb | njohnston: all of those 1.00 values along the x axis could add upquickly | 16:10 |
| clarkb | it is really hard to tell from that particular graph | 16:10 |
| *** dtantsur is now known as dtantsur|afk | 16:10 | |
| clarkb | njohnston: you can get the raw data from grpahite by using the json data type iirc. Then you could sum them yourself | 16:11 |
| clarkb | I forget what the parameter to set that is | 16:11 |
| njohnston | clarkb: You know what, I don't have a high enough degree of confidence I don't have user error; I'll try that json method | 16:11 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 16:15 |
| *** ykarel_ has joined #openstack-infra | 16:18 | |
| *** ykarel_ is now known as ykarel|away | 16:20 | |
| *** ykarel has quit IRC | 16:20 | |
| *** slagle has joined #openstack-infra | 16:20 | |
| hwoarang | fungi: do you mind if i keep the reserved node overnight? problem is more complicated than i thought so i need more time | 16:21 |
| fungi | hwoarang: feel free, we don't generally delete them until someone says they're done investigating | 16:21 |
| hwoarang | ok thank you | 16:22 |
| fungi | hwoarang: it's part of why we've started putting people's irc nicks in the hold reason text, so we can check back in with them | 16:22 |
| hwoarang | :) | 16:23 |
| dhellmann | jamespage : ack, you're right | 16:23 |
| dhellmann | lucasagomes : \o/ | 16:24 |
| *** jlibosva has quit IRC | 16:27 | |
| *** aidin has quit IRC | 16:31 | |
| *** Bhujay has joined #openstack-infra | 16:34 | |
| *** aojea has quit IRC | 16:37 | |
| *** dosaboy has quit IRC | 16:39 | |
| mwhahaha | hey openstackgerrit is missing from #tripleo is the config broken or does the bot need a restart? It disappeared two days ago | 16:43 |
| *** kjackal has quit IRC | 16:44 | |
| fungi | mwhahaha: openstackgerrit opportunistically parts and joins channels when it has messages to convey, so as to be able to operate at or under freenode's 120 cnocurrent channels limit | 16:45 |
| *** dosaboy has joined #openstack-infra | 16:45 | |
| mwhahaha | fungi: that doesn't seem to be working, makes sense though | 16:45 |
| fungi | i'm checking its logs now | 16:46 |
| fungi | because presumably it's had events to send to #tripleo in the past two days and it's doing so in other channels successfully | 16:46 |
| fungi | gerritbot last restarted at 18:35 on tuesday | 16:48 |
| mwhahaha | yea thats when it stopped | 16:49 |
| mwhahaha | i just looked last time i saw it was when it quit at 12:35 (my time) | 16:50 |
| fungi | 2018-10-04 16:40:54,470 INFO gerritbot: Sending "Merged openstack/tripleo-heat-templates stable/queens: Add networks to IronicConductor role. https://review.openstack.org/607121" to #tripleo | 16:51 |
| fungi | so it seems to think it's in that channel | 16:51 |
| mwhahaha | silly bot | 16:52 |
| fungi | here's the first one after the restart: | 16:53 |
| fungi | 2018-10-02 18:35:31,022 INFO gerritbot: Sending "wes hayutin proposed openstack/tripleo-quickstart-extras master: by default collect all files in /var/lib/mistral https://review.openstack.org/607340" to #tripleo | 16:54 |
| fungi | 2018-10-02 18:35:31,023 INFO gerritbot: Joined channel #tripleo | 16:54 |
| mwhahaha | http://eavesdrop.openstack.org/irclogs/%23tripleo/%23tripleo.2018-10-02.log.html#t2018-10-02T18:35:26 yea it didn't join | 16:54 |
| fungi | no errors or tracebacks associated with that | 16:54 |
| fungi | bizarre | 16:54 |
| *** dosaboy has quit IRC | 16:55 | |
| *** diablo_rojo has joined #openstack-infra | 16:55 | |
| *** ykarel|away has quit IRC | 16:58 | |
| fungi | unfortunately the tendency for gerrit events to contain text of review comments that include python tracebacks is making it exceedingly hard to search for actual tracebacks in the gerritbot service log | 16:58 |
| *** Bhujay has quit IRC | 17:01 | |
| *** dosaboy has joined #openstack-infra | 17:01 | |
| *** shardy has quit IRC | 17:04 | |
| *** derekh has quit IRC | 17:04 | |
| *** trown|lunch is now known as trown | 17:04 | |
| *** ramishra has quit IRC | 17:09 | |
| fungi | mwhahaha: i'm stumped. my best guess is that there was some unreported error when the bot tried to /join #tripleo and so it continued on the assumption it had joined thereafter. since it only does a /part from the least active channels when it has a message for a channel it's not currently in, it's never thought it needed to rejoin #tripleo over the past two days | 17:14 |
| fungi | i have a feeling it's an authentication race | 17:14 |
| mwhahaha | :( | 17:14 |
| fungi | since we're back to requiring nick registration, it's entirely possible the /join happened before nickserv confirmed its /identify | 17:14 |
| fungi | we merged some related patches to meetbot, ptgbot and statusbot | 17:15 |
| fungi | so i'll see if there's anything we should be doing with gerritbot to avoid such races | 17:15 |
| clarkb | change the auth method? | 17:16 |
| fungi | right that's one of the options we've used in some of the bots | 17:16 |
| fungi | ptgbot and statusbot we stitched to sasl auth | 17:16 |
| fungi | er, switched | 17:16 |
| fungi | meetbot had a configuration option to delay /join until after /identify succeeded | 17:16 |
| clarkb | ah | 17:16 |
| fungi | depends mostly on which framework gerritbot is using | 17:17 |
| fungi | and how easy/hard various options end up being | 17:17 |
| *** pcaruana has joined #openstack-infra | 17:17 | |
| fungi | adapting the irc module for python (what statusbot and ptgbot use) required some moderate degree of patching and a new library | 17:17 |
| clarkb | I want to say it is using irc module too? | 17:18 |
| fungi | gerritbot looks like it might be a candidate for that option, yes | 17:19 |
| fungi | https://review.openstack.org/599049 is my statusbot fix if anyone wants to review it ;) | 17:20 |
| clarkb | oh good I've already done so | 17:20 |
| fungi | if there are no volunteers to try porting that to gerritbot, i'll add it to my priority to do list | 17:20 |
| *** jpena is now known as jpena|off | 17:21 | |
| clarkb | I'm not in a great spot right now. Currently in the austin office and not long before I have to work my way toward the airport | 17:22 |
| *** gfidente is now known as gfidente|afk | 17:28 | |
| *** Bhujay has joined #openstack-infra | 17:30 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/elastic-recheck master: Generic OOM killer query https://review.openstack.org/608038 | 17:32 |
| clarkb | mordred: mriedem ^ thats a thin gto help debugging in the futuer | 17:33 |
| *** Swami has quit IRC | 17:36 | |
| *** amoralej is now known as amoralej|off | 17:37 | |
| *** bobh has joined #openstack-infra | 17:45 | |
| AJaeger | clarkb, fungi, could you review https://review.openstack.org/601688 (python3-first charms jobs removal) and https://review.openstack.org/#/c/607501/ (change of release jobs), please? | 17:47 |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Update stackviz tarball location https://review.openstack.org/608044 | 17:47 |
| clarkb | mtreinish: fungi ^ yall should double check me on that but I think that is necessary to get the python3 fixes onto the test nodes | 17:48 |
| *** Emine has joined #openstack-infra | 17:51 | |
| *** yamamoto has quit IRC | 17:51 | |
| *** yamamoto has joined #openstack-infra | 17:52 | |
| mtreinish | clarkb: well that path resolves for me. But the version downloaded is stackviz 0.0.1.dev276 but doing 'git rev-list --all --count' on current master returns 301 | 17:55 |
| mtreinish | so is that the right version? or is there fuzzy math for that dev number? | 17:55 |
| clarkb | mtreinish: what does git describe say? | 17:55 |
| clarkb | hrm no tags so describe doesn't work | 17:56 |
| clarkb | let me double check if the python3 fix is in the newer tarball | 17:56 |
| *** yamamoto has quit IRC | 17:57 | |
| clarkb | mtreinish: the newer tarball url has the python3 fix. I'm guessing there is something about how pbr calculates that 276 nymber | 17:57 |
| mtreinish | ok then the patch lgtm. Just wanted to make sure | 17:58 |
| clarkb | mtreinish: yup, I asked for you to double check me :) | 17:58 |
| clarkb | thank you for checking | 17:58 |
| *** ykarel|away has joined #openstack-infra | 17:59 | |
| mtreinish | clarkb: looking at the git log, it might be 276 commits since setup.py was added to the repo | 18:00 |
| clarkb | ah | 18:00 |
| mtreinish | hmm, that doesn't add up either | 18:01 |
| *** olivierb has joined #openstack-infra | 18:01 | |
| fungi | git log --oneline origin/master|wc -l | 18:01 |
| fungi | 277 | 18:01 |
| mtreinish | ah git log --oneline | wc -l is 277, while git rev list was 301 | 18:01 |
| *** olivierb has quit IRC | 18:01 | |
| *** pcaruana has quit IRC | 18:01 | |
| clarkb | I also think there are some compiled "artifacts" in that sdist | 18:01 |
| clarkb | but the file size matches up closely with the existing tarball so I think we are still compiling those in | 18:01 |
| fungi | yeah, the dist tarball is more than an order of magnitude larger than the regular tarball | 18:02 |
| openstackgerrit | Merged openstack-infra/project-config master: Update release jobs for driverlog and tempest https://review.openstack.org/607501 | 18:03 |
| fungi | 745k reported for it vs 32k for the sdist tarball and 23k for the wheel | 18:03 |
| clarkb | yup | 18:03 |
| *** Swami has joined #openstack-infra | 18:04 | |
| *** ykarel_ has joined #openstack-infra | 18:04 | |
| openstackgerrit | Merged openstack-infra/project-config master: remove job settings for OpenStack Charms repositories https://review.openstack.org/601688 | 18:05 |
| *** ykarel|away has quit IRC | 18:07 | |
| weshay | any chance tripleo could get https://review.openstack.org/#/c/607952/ moved to the top of the gate queue? | 18:07 |
| fungi | weshay: sure, just a sec | 18:08 |
| weshay | thank you! | 18:08 |
| fungi | weshay: done | 18:08 |
| weshay | woot | 18:08 |
| *** timrc has joined #openstack-infra | 18:10 | |
| clarkb | fungi: thanks | 18:14 |
| *** Bhujay has quit IRC | 18:20 | |
| *** imacdonn has quit IRC | 18:22 | |
| *** imacdonn has joined #openstack-infra | 18:22 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/zuul-jobs master: support passing extra arguments to bdist_wheel in build-python-release https://review.openstack.org/607900 | 18:27 |
| *** yamamoto has joined #openstack-infra | 18:30 | |
| *** e0ne has joined #openstack-infra | 18:31 | |
| *** e0ne has quit IRC | 18:33 | |
| fungi | "We will also have a question and answer session at the end of the presentation. | 18:33 |
| fungi | s/presentation./presentation."/ | 18:33 |
| fungi | nice | 18:33 |
| fungi | er, wrong channel | 18:34 |
| *** mriedem has joined #openstack-infra | 18:34 | |
| timrc | fungi: Now you're obligated to link us to the presentation. | 18:36 |
| fungi | timrc: gladly, once the announcement goes out later today | 18:36 |
| *** kjackal has joined #openstack-infra | 18:44 | |
| *** eernst has joined #openstack-infra | 18:45 | |
| *** eernst has joined #openstack-infra | 18:45 | |
| *** electrofelix has quit IRC | 18:51 | |
| *** eernst has quit IRC | 18:51 | |
| *** eernst has joined #openstack-infra | 18:51 | |
| *** agopi|long_lunch is now known as agopi | 18:54 | |
| clarkb | fungi: off the top of your head do you know if the pipeline precedence update made it in? | 18:55 |
| *** jcoufal has quit IRC | 18:57 | |
| openstackgerrit | John Studarus proposed openstack-infra/project-config master: upgrade complete https://review.openstack.org/608059 | 18:57 |
| *** studarus_ has joined #openstack-infra | 18:59 | |
| fungi | clarkb: it did a couple days ago | 19:01 |
| clarkb | neat | 19:01 |
| AJaeger | clarkb, fungi, want to enqueue 608059? It will give us an extra 75 nodes ;) | 19:02 |
| clarkb | I've approved the change. Can enqueue in a moment | 19:03 |
| AJaeger | thanks, clarkb | 19:03 |
| kmalloc | well.. that is unfortunate. | 19:03 |
| clarkb | kmalloc: ? | 19:03 |
| kmalloc | clarkb, fungi, mordred: am I crazy... ubuntu one is OpenID not OpenID Connect? | 19:03 |
| clarkb | oui | 19:03 |
| kmalloc | which means dex is off the table without a bunch of go code. | 19:04 |
| kmalloc | and keycloak may also not be on the table. | 19:04 |
| clarkb | openstackid does both | 19:04 |
| clarkb | iirc | 19:04 |
| kmalloc | right | 19:04 |
| kmalloc | but we want to support both things for a period of time | 19:04 |
| clarkb | AJaeger: enqueued | 19:04 |
| clarkb | kmalloc: ya | 19:04 |
| AJaeger | great | 19:05 |
| clarkb | (just pointing out the other system should be ok) | 19:05 |
| kmalloc | meaning, we need to have a OID aware thing not just OIDC aware | 19:05 |
| fungi | kmalloc: that is my understanding as well, yes | 19:05 |
| kmalloc | yeah, so i think this means i need to write a proxy since the proxies that exist is OIDC only | 19:05 |
| kmalloc | not a huge deal, but the lack of Oauth data being passed back breaks the assumptions | 19:05 |
| kmalloc | nbd. I can do a stop-gap OID proxy and i know keystone (long term) could easily do both | 19:06 |
| kmalloc | but i am not keen on a bunch of golang code to wedge it into dex (or java for keycloak) if it's going to be a bunch of time sunk into it. | 19:06 |
| kmalloc | if that makes sense. | 19:07 |
| kmalloc | i'm still checking out keycloak's support (it might work) | 19:07 |
| kmalloc | but dex looks very very OIDC specific | 19:07 |
| clarkb | kmalloc: on the java side you may be able to take gerrits implementation for some of it? | 19:07 |
| kmalloc | yeah possibly | 19:07 |
| kmalloc | though keycloak may already do this ok | 19:08 |
| kmalloc | so i'm digging into it | 19:08 |
| mrhillsman | so i have this strange issue hoping to get some help on | 19:10 |
| mrhillsman | all of our jobs are "fake" running | 19:10 |
| *** eernst has quit IRC | 19:10 | |
| mrhillsman | http://status.openlabtesting.org/t/openlab/status.html | 19:10 |
| kmalloc | clarkb: well color me disappointed in Ubuntu One. :P | 19:10 |
| clarkb | mrhillsman: check that you have nodepool nodes being booted and assigned | 19:12 |
| clarkb | I think that they go into that state after git things are sorted out and before the node is assigned | 19:12 |
| clarkb | and I see at least one node failure | 19:12 |
| *** ykarel__ has joined #openstack-infra | 19:13 | |
| openstackgerrit | Merged openstack-infra/project-config master: upgrade complete https://review.openstack.org/608059 | 19:13 |
| fungi | kmalloc: keep in mind that canonical pulled most of their devs off that as a priority long before the oidc spec emerged | 19:13 |
| *** ykarel__ has quit IRC | 19:14 | |
| mrhillsman | ok thx clarkb | 19:14 |
| *** ykarel__ has joined #openstack-infra | 19:14 | |
| kmalloc | fungi: thats fair. i think we're going to need some custom code to make this work. | 19:15 |
| *** ykarel_ has quit IRC | 19:15 | |
| *** mriedem has quit IRC | 19:15 | |
| kmalloc | fungi: rather than just lean on a service. Though I am pretty sure I can make this ultimately work. | 19:15 |
| *** ykarel_ has joined #openstack-infra | 19:17 | |
| *** _erlon_ has quit IRC | 19:19 | |
| *** ykarel__ has quit IRC | 19:20 | |
| *** ykarel__ has joined #openstack-infra | 19:21 | |
| *** ykarel_ has quit IRC | 19:23 | |
| *** mriedem has joined #openstack-infra | 19:24 | |
| *** ykarel__ has quit IRC | 19:26 | |
| *** aidin has joined #openstack-infra | 19:31 | |
| fuentess | clarkb: hey, I reproduced the issues I saw, and then when running the tests, the VM got hanged... seems like the same issue we have hit before. You can dispose the VM, but I'm not sure if you will be able to | 19:34 |
| *** aidin has quit IRC | 19:35 | |
| fungi | fuentess: we just delete the vm anyway, so unless it's somehow hung the hypervisor... | 19:38 |
| fuentess | fungi: thanks | 19:38 |
| AJaeger | fungi, I think dtroyer would love to move forward with sterling publishing - want to review https://review.openstack.org/#/c/608003/ ? | 19:38 |
| dtroyer | AJaeger: heh, I'm working on the next batch already for releasenotes and api-ref… am I heading in the right direction? | 19:42 |
| AJaeger | dtroyer: yes, looks that way ;) | 19:43 |
| *** gfidente|afk has quit IRC | 19:43 | |
| *** slagle has quit IRC | 19:43 | |
| AJaeger | dtroyer: but those are changes that need to be tested and you cannot test them, so we need to merge, merge a change that publishes - and go from there... | 19:43 |
| dtroyer | cool, thanks. right, why it's being batched | 19:44 |
| dtroyer | is publish-openstack-releasenotes-base a good model for our version of that? | 19:44 |
| AJaeger | team, FYI, the lines in project-config/zuul.d/projects.yaml have decreased from 35000 to 10000 since the time we removed the Zuul v2 config.... | 19:44 |
| AJaeger | dtroyer: that jobs calls sphinx directly - if you want to use tox -e releasenotes instead, base it on the tox-docs job. | 19:46 |
| AJaeger | dtroyer: your call... | 19:46 |
| dtroyer | I'd prefer the tox interface, ok, thanks. so many ways to do things it's hard to know the tradeoffs sometimes | 19:46 |
| *** eharney has quit IRC | 19:47 | |
| AJaeger | dtroyer: some is history - we're moving away from tox interface, so today we would do releasenotes differently... | 19:47 |
| dtroyer | oh? I thought PTI pushed us toward tox? | 19:47 |
| *** jcoufal has joined #openstack-infra | 19:48 | |
| AJaeger | dtroyer: PTI for docs. releasenotes is a different beast, we didn't update PTI for that yet AFAIK | 19:49 |
| dtroyer | ah, ok. yeah it is different… I'll stick with tox for now then since we have that set up and I want to run that locally too | 19:50 |
| dtroyer | but will try to follow along too... | 19:50 |
| openstackgerrit | Merged openstack-infra/project-config master: Add publish jobs for stx-governance and per-project https://review.openstack.org/608003 | 19:54 |
| fungi | the original argument for not doing docs builds with tox is that most other sphinx-using projects "just call sphinx-build" (plus tox is seen as being python-specific). what we discovered the deeper we went is that most sphinx-using projects also simply shrug off the developer experience for setting up a sufficient dev environment to be _able_ to call sphinx-build | 19:55 |
| fungi | so we're "just calling pshinx-build" after setting up a suitable environment (by using tox) | 19:56 |
| fungi | and so might as well invoke sphinx-build with tox while we're at it | 19:56 |
| *** timothyb89 has joined #openstack-infra | 19:57 | |
| dtroyer | ah… I like the self-contained-ness of the tox setup for local use, in fact have started using an tox env 'newnote' for creating new release notes using the exiting releasenote env | 19:57 |
| fungi | yup, my personal projects support `tox -e dist` for building sdists and wheels and testing them for sanity | 19:58 |
| * dtroyer may or may not have Makefiles in work repos for this still dating back a bit | 19:58 | |
| fungi | also `tox -e debug` for service projects which sets up a venv and starts the service, then exits so it can be interacted with manually | 19:58 |
| *** jcoufal_ has joined #openstack-infra | 20:03 | |
| *** fuentess has quit IRC | 20:04 | |
| *** studarus_ has quit IRC | 20:04 | |
| *** jcoufal has quit IRC | 20:05 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Use prepare-zanata-client role https://review.openstack.org/608069 | 20:07 |
| *** e0ne has joined #openstack-infra | 20:08 | |
| clarkb | I'm doing the metal tube in the sky thing again now.likely afk until sometime tomorrow | 20:09 |
| AJaeger | clarkb: safe travels! | 20:11 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-pip master: [DNM] debugging https://review.openstack.org/607048 | 20:13 |
| *** hashar has joined #openstack-infra | 20:13 | |
| *** e0ne has quit IRC | 20:14 | |
| *** sthussey has quit IRC | 20:15 | |
| *** priteau has quit IRC | 20:15 | |
| AJaeger | dhellmann: enjoy https://review.openstack.org/#/q/topic:python3-first+is:open+message:%22import+zuul+job+settings+from+project-config%22 ;) | 20:18 |
| *** slagle has joined #openstack-infra | 20:19 | |
| *** hashar has quit IRC | 20:22 | |
| *** diablo_rojo has quit IRC | 20:24 | |
| *** diablo_rojo has joined #openstack-infra | 20:25 | |
| *** e0ne has joined #openstack-infra | 20:30 | |
| openstackgerrit | Merged openstack-infra/project-config master: Update stackviz tarball location https://review.openstack.org/608044 | 20:33 |
| *** kgiusti has left #openstack-infra | 20:33 | |
| *** priteau has joined #openstack-infra | 20:35 | |
| *** devananda has quit IRC | 20:38 | |
| *** slagle has quit IRC | 20:39 | |
| *** jtomasek has quit IRC | 20:39 | |
| AJaeger | dhellmann: could you rebase https://review.openstack.org/#/c/601583 , please? trove is done thanks to smcginnis' reviews... | 20:40 |
| * AJaeger waves good night | 20:40 | |
| *** ansmith has quit IRC | 20:41 | |
| *** kjackal has quit IRC | 20:42 | |
| *** tosky has quit IRC | 20:42 | |
| *** tosky has joined #openstack-infra | 20:43 | |
| *** ijw has quit IRC | 20:43 | |
| *** hamzy_ has quit IRC | 20:44 | |
| bkero | How many logs (in terms of disk space) does openstack-infra generate per day? | 20:48 |
| *** priteau has quit IRC | 20:48 | |
| mtreinish | bkero: compressed or raw? | 20:51 |
| fungi | bkero: compressed we have graphs | 20:51 |
| bkero | Is that in cacti? | 20:51 |
| bkero | Well at the end I was thinking about line entries in kibana, which I guess is a subset of logs. But that got me curious about the actual disk space use of all the logs collected. | 20:53 |
| bkero | So uncompressed would be a better metrick | 20:53 |
| bkero | s/metric/ | 20:53 |
| fungi | bkero: yeah, http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=717&rra_id=all | 20:53 |
| fungi | if you take the retention timeframe into account and make some educated guesses about average compression ratio, you can get a ballpark figure | 20:54 |
| bkero | What's the retention time? | 20:54 |
| *** trown is now known as trown|outtypewww | 20:54 | |
| bkero | I guess it might be different depending on log type | 20:55 |
| fungi | nope, flat 30 days across the board | 20:55 |
| bkero | Oh that's convenient and easy to understand. Thanks! | 20:55 |
| fungi | https://git.openstack.org/cgit/openstack-infra/puppet-openstackci/tree/files/log_archive_maintenance.sh is what does our current log expiration | 20:55 |
| bkero | That's a hell of a find statement, heh | 20:56 |
| fungi | so if we assume we're compressing files on average to ~1/3 of their original size it's probably 1tb/day raw | 20:57 |
| fungi | adjust compression ratio to fit your definition of likely | 20:58 |
| bkero | I'd expect a bit more than that usually. Maybe 33% is safe though. | 20:59 |
| fungi | 1/3 is a generous estimate, it's probably better, yes | 20:59 |
| bkero | You could achieve some real mean compression if you had shared keys across multiple files/jobs. | 20:59 |
| fungi | easily 1tb raw daily at least | 20:59 |
| fungi | bkero: i can only imagine what the performance would be like to stream files compressed that way | 21:00 |
| *** e0ne has quit IRC | 21:00 | |
| bkero | fungi: Certainly worse, but maybe acceptably worse if those files are rarely/never accessed. | 21:00 |
| fungi | more concerned with how slow it would be to access that shared compression token index | 21:01 |
| fungi | since the decompressor would need it | 21:02 |
| bkero | It would, and it would need to append to it when introducing new logs. | 21:03 |
| bkero | It's just trading CPU and memory use for disk space. | 21:03 |
| fungi | or bandwidth, depending on where the decompression happens | 21:03 |
| bkero | I guess in my conceptualization I thought it would live in situ with the data | 21:05 |
| *** mriedem has quit IRC | 21:05 | |
| bkero | Although it could also live on a SSD adjacent to the spinning rust if the token index needed random read perf | 21:06 |
| fungi | i meant if a client wanted to retrieve compressed data and then decompress it locally | 21:08 |
| fungi | but sure, those too | 21:08 |
| bkero | I can't find much evidence of folks exploring recompressing like that. I guess that's what tar was originally for. | 21:12 |
| *** agopi has quit IRC | 21:12 | |
| fungi | people do use wget recursive and similar to retrieve full sets of logs from specific builds for local inspection | 21:20 |
| bkero | Oh certainly, there are plenty of use cases for that. | 21:23 |
| bkero | I wonder if there are apache modules for doing indexes inside of gzipped tarballs | 21:24 |
| *** devananda has joined #openstack-infra | 21:25 | |
| fungi | timrc: my earlier misdirected comment was in regards to a draft i was proofreading for http://lists.openstack.org/pipermail/openstack-dev/2018-October/135441.html | 21:30 |
| *** kjackal has joined #openstack-infra | 21:30 | |
| *** panda is now known as panda|off | 21:36 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: remove job settings for trove repositories https://review.openstack.org/601583 | 21:36 |
| dhellmann | config-core: the trove zuul cleanup patch is ready for review: https://review.openstack.org/#/c/601583 | 21:46 |
| *** ijw has joined #openstack-infra | 21:48 | |
| *** weshay is now known as weshay_pto | 21:51 | |
| *** bobh has quit IRC | 21:58 | |
| *** jamesmcarthur has quit IRC | 22:05 | |
| *** jcoufal_ has quit IRC | 22:07 | |
| *** kjackal has quit IRC | 22:10 | |
| *** jamesmcarthur has joined #openstack-infra | 22:17 | |
| *** jamesmcarthur has quit IRC | 22:22 | |
| *** bdodd has quit IRC | 22:25 | |
| *** rcernin has joined #openstack-infra | 22:27 | |
| *** jamesmcarthur has joined #openstack-infra | 22:31 | |
| *** tpsilva has quit IRC | 22:40 | |
| *** bdodd has joined #openstack-infra | 22:42 | |
| *** lifeless has joined #openstack-infra | 22:44 | |
| lifeless | hmm, dunno how I lost this channel ... | 22:44 |
| lifeless | fungi: should we merge that patch? | 22:44 |
| *** jamesmcarthur has quit IRC | 22:48 | |
| fungi | lifeless: if you mean 607894 i'm okay waiting for an additional reviewer since dhellmann seems to have some concerns over it, but we do need that or something like it merged to un-block pbr testing | 22:49 |
| lifeless | It seemed like dhellmann mainly didn't want to have to shepard hundreds of patches through; and its as compat back to Trusty, I think its fine. | 22:51 |
| lifeless | anyhow, you have my +2 now I'm unconfused :) | 22:51 |
| lifeless | so I'll leave it with you | 22:51 |
| fungi | thanks! i *think* dhellmann didn't actually have concerns with 607894 itself and was merely replying to AJaeger's question about whether we need to go update setup.cfg similarly in other projects | 22:52 |
| fungi | but i want to be sure | 22:53 |
| fungi | an alternative, i think, is that stephenfin could split the setup.cfg diff to a separate change since that seems like it may only be tangentially-related cleanup to silence a deprecation warning | 22:54 |
| *** harlowja has quit IRC | 22:54 | |
| *** harlowja has joined #openstack-infra | 22:55 | |
| *** boden has quit IRC | 23:03 | |
| *** pbourke has quit IRC | 23:03 | |
| *** jamesmcarthur has joined #openstack-infra | 23:04 | |
| *** pbourke has joined #openstack-infra | 23:04 | |
| *** eernst has joined #openstack-infra | 23:05 | |
| *** eernst has quit IRC | 23:07 | |
| *** ijw has quit IRC | 23:14 | |
| *** ijw_ has joined #openstack-infra | 23:14 | |
| *** tosky has quit IRC | 23:17 | |
| *** jamesmcarthur has quit IRC | 23:24 | |
| *** slagle has joined #openstack-infra | 23:35 | |
| *** edmondsw has quit IRC | 23:38 | |
| *** jamesmcarthur has joined #openstack-infra | 23:43 | |
| *** jamesmcarthur has quit IRC | 23:49 | |
| *** jamesmcarthur has joined #openstack-infra | 23:50 | |
| *** ijw_ has quit IRC | 23:53 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
