| ianw | dodgy bt acquired ... http://paste.openstack.org/show/783505/ | 00:12 |
|---|---|---|
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Temporarily disable Fedora 30 build jobs https://review.opendev.org/688315 | 00:43 |
| clarkb | ianw ooh segfault in malloc | 00:58 |
| clarkb | are you able to install gdb and or valgrind? | 00:58 |
| ianw | there's a few bugs out there that all look similar-ish | 00:58 |
| clarkb | I guessyou can use rpm directly | 00:58 |
| ianw | yeah, but "dnf debuginfo-install" does all sorts of magic with repos etc that make it a pita to do manually | 00:59 |
| ianw | the debug packages install, but the dnf call segfaults ... i'm trying to make it ignore than, and move on to another dnf that fails under gdb and should then have symbols ... | 01:00 |
| ianw | 2019-10-14 01:00:11.636 | Completion plugin: Generating completion cache... | 01:00 |
| ianw | 2019-10-14 01:00:11.638 | free(): invalid pointer | 01:00 |
| ianw | ok, that's more like it ... | 01:00 |
| *** markvoelker has joined #openstack-infra | 01:32 | |
| *** markvoelker has quit IRC | 01:42 | |
| *** markvoelker has joined #openstack-infra | 01:43 | |
| *** markvoelker has quit IRC | 01:47 | |
| ianw | clarkb / pabelanger: filed https://bugzilla.redhat.com/show_bug.cgi?id=1761282 but pretty confident it's a dup of https://bugzilla.redhat.com/show_bug.cgi?id=1758737 | 01:57 |
| openstack | bugzilla.redhat.com bug 1761282 in dnf "dnf segfault after "Generating completion cache"; likely libdnf 0.35.5" [High,New] - Assigned to rpm-software-management | 01:57 |
| openstack | bugzilla.redhat.com bug 1758737 in libdnf "DNF completed with "Segmentation fault (core dumped)" message" [Unspecified,Post] - Assigned to rpm-software-management | 01:57 |
| ianw | the dependencies are too tangled to try and unwrap with hack installs of old versions i think. better to just wait for updated packages. | 01:58 |
| *** calbers has quit IRC | 02:06 | |
| *** calbers has joined #openstack-infra | 02:09 | |
| *** ramishra has joined #openstack-infra | 02:19 | |
| *** whoami-rajat has joined #openstack-infra | 03:14 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Temporarily disable Fedora 30 build jobs https://review.opendev.org/688315 | 03:17 |
| *** zhurong has joined #openstack-infra | 03:19 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: centos7; use numeric DIB_RELEASE https://review.opendev.org/688321 | 03:35 |
| *** psachin has joined #openstack-infra | 03:35 | |
| ianw | dtantsur|afk: ^ i think something like this will be the best way forward? | 03:36 |
| *** ykarel has joined #openstack-infra | 03:55 | |
| *** udesale has joined #openstack-infra | 04:07 | |
| openstackgerrit | Merged opendev/glean master: Do not bring up udev assigned interfaces https://review.opendev.org/688031 | 04:09 |
| *** ricolin has joined #openstack-infra | 04:12 | |
| *** ricolin_ has joined #openstack-infra | 04:15 | |
| *** dave-mccowan has quit IRC | 04:18 | |
| *** ricolin has quit IRC | 04:19 | |
| *** lpetrut has joined #openstack-infra | 04:42 | |
| *** odicha has joined #openstack-infra | 04:50 | |
| *** lpetrut has quit IRC | 04:52 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Remove RA solicit delay https://review.opendev.org/688218 | 04:52 |
| *** surpatil has joined #openstack-infra | 05:04 | |
| *** ykarel has quit IRC | 05:04 | |
| *** kjackal has joined #openstack-infra | 05:09 | |
| *** ykarel has joined #openstack-infra | 05:24 | |
| *** jbadiapa has joined #openstack-infra | 05:36 | |
| ianw | clarkb: ^ merged and release dib+glean; is rolled out to builders and i've scheduled builds of f28,29,30 + centos7/8 | 05:49 |
| ianw | i'm hoping that puts this problem behind us! working on a better fix | 05:49 |
| *** larainema has joined #openstack-infra | 05:51 | |
| *** AJaeger has quit IRC | 05:52 | |
| *** kopecmartin|off is now known as kopecmartin | 05:59 | |
| *** ccamacho has joined #openstack-infra | 06:10 | |
| *** AJaeger has joined #openstack-infra | 06:14 | |
| *** dpawlik has joined #openstack-infra | 06:23 | |
| *** pcaruana has joined #openstack-infra | 06:43 | |
| *** aluria has joined #openstack-infra | 06:57 | |
| *** apetrich has joined #openstack-infra | 07:04 | |
| *** slaweq has joined #openstack-infra | 07:04 | |
| *** tesseract has joined #openstack-infra | 07:04 | |
| *** rcernin has quit IRC | 07:11 | |
| *** raukadah is now known as chandankumar | 07:11 | |
| *** hwoarang has quit IRC | 07:30 | |
| *** hwoarang has joined #openstack-infra | 07:31 | |
| *** apetrich has quit IRC | 07:32 | |
| *** xenos76 has joined #openstack-infra | 07:32 | |
| *** lucasagomes has joined #openstack-infra | 07:34 | |
| *** gfidente has joined #openstack-infra | 07:37 | |
| *** rpittau|afk is now known as rpittau | 07:40 | |
| *** kjackal has quit IRC | 07:43 | |
| *** markvoelker has joined #openstack-infra | 07:45 | |
| *** jpena|off is now known as jpena | 07:46 | |
| *** kjackal has joined #openstack-infra | 07:46 | |
| *** ykarel is now known as ykarel|lunch | 07:48 | |
| *** markvoelker has quit IRC | 07:51 | |
| *** ralonsoh has joined #openstack-infra | 07:54 | |
| *** iurygregory has joined #openstack-infra | 07:58 | |
| *** gfidente has quit IRC | 07:59 | |
| *** dpawlik has quit IRC | 08:03 | |
| *** lpetrut has joined #openstack-infra | 08:13 | |
| *** dchen has quit IRC | 08:14 | |
| *** ykarel|lunch is now known as ykarel | 08:33 | |
| *** dtantsur|afk is now known as dtantsur | 08:34 | |
| *** derekh has joined #openstack-infra | 08:38 | |
| *** ricolin_ is now known as ricolin | 08:41 | |
| *** xek_ has joined #openstack-infra | 08:47 | |
| *** gfidente has joined #openstack-infra | 08:51 | |
| *** dpawlik has joined #openstack-infra | 09:13 | |
| *** dpawlik has quit IRC | 09:18 | |
| *** trident has quit IRC | 09:23 | |
| *** trident has joined #openstack-infra | 09:28 | |
| *** dpawlik has joined #openstack-infra | 09:30 | |
| *** brtknr_ has quit IRC | 09:31 | |
| *** brtknr has joined #openstack-infra | 09:32 | |
| *** apetrich has joined #openstack-infra | 09:39 | |
| *** jamesmcarthur has joined #openstack-infra | 09:40 | |
| *** ricolin has quit IRC | 09:41 | |
| *** pkopec has joined #openstack-infra | 09:43 | |
| *** jamesmcarthur has quit IRC | 09:45 | |
| *** ociuhandu has joined #openstack-infra | 09:53 | |
| *** ociuhandu has quit IRC | 10:24 | |
| *** surpatil has quit IRC | 10:25 | |
| *** surpatil has joined #openstack-infra | 10:25 | |
| *** ociuhandu has joined #openstack-infra | 10:29 | |
| *** prometheanfire has quit IRC | 10:40 | |
| *** prometheanfire has joined #openstack-infra | 10:41 | |
| *** surpatil has quit IRC | 10:44 | |
| *** surpatil has joined #openstack-infra | 10:46 | |
| *** elod has quit IRC | 10:51 | |
| *** elod has joined #openstack-infra | 10:52 | |
| *** ociuhandu has quit IRC | 10:53 | |
| *** dchen has joined #openstack-infra | 10:58 | |
| *** udesale has quit IRC | 11:13 | |
| *** dchen has quit IRC | 11:13 | |
| *** odicha has quit IRC | 11:24 | |
| *** ociuhandu has joined #openstack-infra | 11:28 | |
| *** jpena is now known as jpena|lunch | 11:31 | |
| *** whoami-rajat has quit IRC | 11:33 | |
| *** pcaruana has quit IRC | 11:34 | |
| *** rosmaita has joined #openstack-infra | 11:36 | |
| *** elod has quit IRC | 11:38 | |
| *** elod has joined #openstack-infra | 11:39 | |
| *** pcaruana has joined #openstack-infra | 11:41 | |
| *** surpatil has quit IRC | 11:56 | |
| *** surpatil has joined #openstack-infra | 11:57 | |
| *** FlorianFa has joined #openstack-infra | 12:05 | |
| *** rh-jelabarre has joined #openstack-infra | 12:05 | |
| *** rfolco has joined #openstack-infra | 12:08 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 12:10 |
| *** rfolco is now known as rfolco|ruck | 12:12 | |
| *** jroll has quit IRC | 12:14 | |
| *** roman_g has joined #openstack-infra | 12:14 | |
| *** jroll has joined #openstack-infra | 12:15 | |
| openstackgerrit | Albin Vass proposed zuul/zuul master: Label check should be case insensitive https://review.opendev.org/688393 | 12:18 |
| *** pgaxatte has joined #openstack-infra | 12:25 | |
| *** jpena|lunch is now known as jpena | 12:27 | |
| *** pgaxatte has quit IRC | 12:32 | |
| *** pgaxatte has joined #openstack-infra | 12:33 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 12:36 |
| *** larainema has quit IRC | 12:44 | |
| pabelanger | ianw: thanks for digging into it! | 12:51 |
| *** ociuhandu has quit IRC | 12:53 | |
| *** Guest31368 has joined #openstack-infra | 12:58 | |
| *** Guest31368 is now known as redrobot_ | 13:00 | |
| *** redrobot_ is now known as redrobot | 13:01 | |
| *** liuyulong has joined #openstack-infra | 13:04 | |
| *** surpatil has quit IRC | 13:05 | |
| *** Guest69423 has quit IRC | 13:06 | |
| *** pgaxatte has quit IRC | 13:09 | |
| *** eharney has joined #openstack-infra | 13:10 | |
| *** mriedem has joined #openstack-infra | 13:14 | |
| *** openstackstatus has quit IRC | 13:15 | |
| *** liuyulong has quit IRC | 13:16 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 13:19 |
| *** slaweq_ has joined #openstack-infra | 13:20 | |
| *** slaweq has quit IRC | 13:23 | |
| *** ociuhandu has joined #openstack-infra | 13:27 | |
| *** liuyulong has joined #openstack-infra | 13:28 | |
| *** surpatil has joined #openstack-infra | 13:28 | |
| *** bnemec has joined #openstack-infra | 13:31 | |
| *** ociuhandu has quit IRC | 13:32 | |
| *** slaweq_ is now known as slaweq | 13:35 | |
| *** pgaxatte has joined #openstack-infra | 13:40 | |
| *** lbragstad_ is now known as lbragstad | 13:44 | |
| *** rosmaita has quit IRC | 13:52 | |
| *** roman_g has quit IRC | 13:55 | |
| *** chandankumar is now known as raukadah | 13:56 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 14:01 |
| *** surpatil has quit IRC | 14:03 | |
| *** sean-k-mooney has joined #openstack-infra | 14:04 | |
| sean-k-mooney | cgoncalves: hi quick question how functional is https://review.opendev.org/#/c/684308/ there are no offical centos8 cloud images and i could not get cloud-init to work properly in the centos8 image for virt-builder | 14:05 |
| openstackgerrit | Merged opendev/storyboard master: Imported Translations from Zanata https://review.opendev.org/684669 | 14:05 |
| *** beekneemech has joined #openstack-infra | 14:06 | |
| sean-k-mooney | cgoncalves: is the disk image build version functional? | 14:06 |
| *** bnemec has quit IRC | 14:06 | |
| *** yolanda__ has joined #openstack-infra | 14:06 | |
| sean-k-mooney | i.e. can i build it an play around with it to see if i can use it for devstack | 14:06 |
| *** dpawlik has quit IRC | 14:07 | |
| *** rosmaita has joined #openstack-infra | 14:07 | |
| cgoncalves | sean-k-mooney, aside from centos not providing yet a cloud image, I'd have expected it to be functional. I was even able to build an octavia amphora image and pass cloud-init on boot | 14:07 |
| cgoncalves | sure. do so and please report back your experience | 14:08 |
| cgoncalves | sean-k-mooney, FWIW I was also able to build a CentOS 8 image using the centos-minimal element | 14:09 |
| sean-k-mooney | cgoncalves: oh i was hoping you where building that yourself | 14:09 |
| *** yolanda has quit IRC | 14:09 | |
| cgoncalves | uhm? I'm lost, sorry :) | 14:09 |
| cgoncalves | so, I built a centos8 cloud image (generic one) myself, yes | 14:09 |
| sean-k-mooney | ya does the dib element not do that? | 14:10 |
| cgoncalves | I have a qcow2 and kickstart file I can share | 14:10 |
| sean-k-mooney | i had not checked that | 14:10 |
| cgoncalves | the centos-minimal element already does that, yes | 14:10 |
| sean-k-mooney | ah | 14:10 |
| cgoncalves | the centos element does not yet, hence my patch | 14:10 |
| sean-k-mooney | so centos-minimal will give me a qcow2 | 14:10 |
| sean-k-mooney | cool | 14:10 |
| cgoncalves | yeah. a minimal one :) | 14:10 |
| sean-k-mooney | the issue i had with virt-build was i coudl get it to install cloud-init | 14:11 |
| sean-k-mooney | but it was disabled | 14:11 |
| sean-k-mooney | and i could not enable the service | 14:11 |
| cgoncalves | really? its systemd service unit comes enabled by default | 14:11 |
| sean-k-mooney | i could run cloud init manully but even as root i was getting permission denised if i tried to enable the service | 14:11 |
| sean-k-mooney | that what i was expecting but nope | 14:12 |
| sean-k-mooney | i think the virt-builder template did somethign strange to block it | 14:12 |
| cgoncalves | I did not have to explicitly enable it, that I am sure | 14:12 |
| sean-k-mooney | anyway ill try out the minimal element for now | 14:12 |
| cgoncalves | nice | 14:12 |
| *** ricolin has joined #openstack-infra | 14:13 | |
| sean-k-mooney | did anywone reach out to the centos folks to see when an offical cloud image might get created | 14:13 |
| cgoncalves | FWIW, if someone wants to build a generic CentOS 8 cloud image while one is not available in cloud.centos.org: http://paste.openstack.org/show/783701/ | 14:13 |
| cgoncalves | yeah. they have it on QA for a while now... | 14:13 |
| *** jamesmcarthur has joined #openstack-infra | 14:13 | |
| *** jamesmcarthur has quit IRC | 14:14 | |
| cgoncalves | their brew system builds it and publishes the logs but not the qcow :/ | 14:14 |
| *** jamesmcarthur has joined #openstack-infra | 14:14 | |
| sean-k-mooney | they had to rewrite a lot of it with the chagne for modules | 14:15 |
| sean-k-mooney | so its proably an oversight | 14:15 |
| sean-k-mooney | infra does not publicly host the nodepool image anywhere right? | 14:16 |
| *** e0ne has joined #openstack-infra | 14:16 | |
| sean-k-mooney | actully | 14:16 |
| sean-k-mooney | that would not make sense | 14:16 |
| sean-k-mooney | becaue they have the proxy config | 14:16 |
| sean-k-mooney | and other things baked in for running in the gate | 14:16 |
| sean-k-mooney | anyway thanks. | 14:17 |
| mordred | sean-k-mooney: we do publish them though | 14:17 |
| mordred | sean-k-mooney: also - fwiw - we don't use cloud-init, we use glean | 14:17 |
| mordred | so the work we've been doing to get centos-8 images working properly for the gate likely won't help you if cloud-init is a goal | 14:18 |
| *** michael-beaver has joined #openstack-infra | 14:18 | |
| sean-k-mooney | mordred: yes i was aware ye use glean | 14:18 |
| sean-k-mooney | that was going to be my fall back | 14:18 |
| sean-k-mooney | if i could not get cloud init to run | 14:18 |
| mordred | I've been conferencing for 3 weeks so I don't know the status of our centos-8 - I thought I saw that ianw had finally had success? | 14:18 |
| sean-k-mooney | mordred: i just wanted a centos 8 image that i could boot on my openstack | 14:19 |
| cgoncalves | mordred, that is correct | 14:19 |
| mordred | nod | 14:19 |
| mordred | well - we don't actually bake proxy config in - we apply that in pre-run | 14:19 |
| mordred | the bits that would be annoying about re-using ours is that we do bake in copies of all of the openstack repos - so the images are a bit large | 14:19 |
| cgoncalves | sean-k-mooney, give my centos 8 qcow2 a try | 14:19 |
| *** priteau has joined #openstack-infra | 14:20 | |
| *** gagehugo has joined #openstack-infra | 14:20 | |
| sean-k-mooney | sure ill try http://paste.openstack.org/show/783701/ | 14:21 |
| mordred | infra-root: I'm travelling all day today - but I *should* have wifi all day and thus expect myself to be working | 14:21 |
| fungi | mordred: noted, thanks! | 14:21 |
| sean-k-mooney | mordred: having the repos backed in acatully would not be the end of the world. i will be suing it for openstack dev adn with qcow and shared backing files it would dedup well but i hold off on going that route unles i fail in building a useful image myself. | 14:23 |
| mordred | sean-k-mooney: ++ | 14:23 |
| fungi | sean-k-mooney: https://nb01.openstack.org/images/centos-8-0000000005.qcow2 | 14:24 |
| fungi | that's our latest build | 14:24 |
| sean-k-mooney | i have setup my cloud to give public ipv6 address to all my vms so i dont want ot hardcode password which is what i did over the weekend | 14:24 |
| fungi | 9.2gb | 14:24 |
| fungi | but yeah, you should just be able to use the latest state of the centos-minimal element | 14:25 |
| cgoncalves | wow, 9.2gb! actual data or zeroes? | 14:25 |
| mordred | actual data. the images have copies of every opendev git repo | 14:26 |
| sean-k-mooney | not all of the is the openstack data however | 14:26 |
| cgoncalves | ok, makes sense | 14:27 |
| sean-k-mooney | the debian images are smaller but still have the repos | 14:27 |
| *** efried_pto is now known as efried | 14:29 | |
| *** e0ne has quit IRC | 14:29 | |
| *** whoami-rajat has joined #openstack-infra | 14:32 | |
| fungi | yeah, centos seems to need several more gigabytes of stuff than debian | 14:32 |
| fungi | but then again, debian has as a focus keeping a strong separation between the absolute minimum base packages needed for a bootable system and things which are nice to have but not necessary | 14:33 |
| clarkb | wht is "funny" about that is if you build a really minimal image centos is much smaller | 14:33 |
| clarkb | once you add useful stuff to the images debian and ubuntu end up smaller | 14:33 |
| fungi | also depends on how you define "really minimal" | 14:33 |
| corvus | and "funny" | 14:34 |
| clarkb | ya in this case itsbare minimum to test glean | 14:34 |
| clarkb | as I've built all the images as small as possible for that use case recently | 14:34 |
| fungi | for example, debian separates the python interpreter and stdlib so you can have python scripts in the base system as long as they don't need stuff from the stdlib modules | 14:35 |
| mordred | fungi: similarly "funny" the "python:slim" base docker image is larger than the ubuntu docker image with python installed | 14:35 |
| fungi | hah | 14:35 |
| clarkb | ianw reports we should have working networking on FN nowish | 14:36 |
| fungi | also debian's package management has features to allow filtering certain files, so you can do things like automatically strip documentation/manpages | 14:36 |
| clarkb | weshould keep an eye out for those zuul test failures (they should go away) | 14:36 |
| clarkb | fungi: ya our images dont install man pages | 14:37 |
| cgoncalves | related: https://review.opendev.org/#/q/topic:yum-reduce-footprint | 14:37 |
| clarkb | which may explain why debuntu grow less than centos qhen we build our actual images | 14:38 |
| cgoncalves | kickstart has an option to exclude docs, --excludedocs. not sure there's an equivalent in yum/dnf | 14:39 |
| sean-k-mooney | im just waiting for someone to add alpine support to devstack | 14:39 |
| sean-k-mooney | then we can all use that :P | 14:40 |
| *** dklyle has quit IRC | 14:40 | |
| *** dtantsur is now known as dtantsur|brb | 14:40 | |
| *** dklyle has joined #openstack-infra | 14:40 | |
| mordred | sean-k-mooney: I actually was reading some stuff from a dude who has switched to alpine as his os for all of his servers and is very happy about having done so. I've been a fan of it as a base container os for a while, but really haven't considered it seriously for servers ... maybe I should think about it more at some point :) | 14:42 |
| sean-k-mooney | i reinstalled the server i use form my home cloud/dev and i was really considering it | 14:43 |
| corvus | mordred: weren't there some, er, concerns about the robustness and viability of the processes used to produce the core alpine system? | 14:43 |
| sean-k-mooney | not that i was aware of but maybe | 14:43 |
| mordred | (although that said we also saw memory leak issues in zuul when running in alpine containers that the theory is related to musldc) | 14:43 |
| clarkb | also its choice of libc made python unhappy iirc | 14:43 |
| sean-k-mooney | clarkb: that is one of the resaons i want to use it | 14:43 |
| mordred | corvus: yeah - SpamapS expressed concerned about the transparency of their dev process | 14:43 |
| sean-k-mooney | i would prefer to not have glibc on my system if i could avoid it | 14:44 |
| mordred | sean-k-mooney: I *think* muslc + python >= 3.6 + lots of dicts became quite displeased - but definitely zuul + alpine was not working for bmw and switching to debian base os fixed the issue | 14:44 |
| sean-k-mooney | i woudl prefer to have everyting build with musl and clang but no one actuly does that end to end and i dont really want ot build my onw distro | 14:44 |
| sean-k-mooney | ya | 14:45 |
| sean-k-mooney | so for other reasons i decided not to make it may os on my server | 14:45 |
| mordred | yah | 14:45 |
| sean-k-mooney | but it does not stop me wanting to be able to use it for that | 14:45 |
| mordred | :) | 14:45 |
| sean-k-mooney | it would be less heretical then what im actully running | 14:46 |
| fungi | i'm just disappointed the emdebian effort folded ~6 years ago, but a lot of the work which came out of that regarding pressure to maintain minimal install base package sizes, optional file filtering at install time, support for compressed filesystems and so on got folded directly into the distro so it was sort of a success in that regard | 14:46 |
| mordred | \o/ unintended success | 14:46 |
| *** ociuhandu has joined #openstack-infra | 14:47 | |
| *** surpatil has joined #openstack-infra | 14:47 | |
| corvus | mordred: i'd like to start working on an integration job for zuul and the gerrit checks plugin -- how do you think i should go about that? should i run gerrit in a container image, or just javac it on the vm and run it in place? if container, should i base it on opendev or upstream dockerfile? | 14:49 |
| mordred | corvus: I think container - since you've already got the quickstart to base things on and we've already got the opendev container images ... | 14:49 |
| mordred | corvus: also - maybe that way we should make a new build job which builds images from master of gerrit with checks plugin added but is otherwise just like the other opendev images | 14:50 |
| *** ihti has quit IRC | 14:50 | |
| mordred | corvus: that way we're building towards having a checks-plugin enabled gerrit for opendev once it's ready | 14:50 |
| fungi | i like that idea | 14:50 |
| mordred | corvus: want me to make a gerrit-master job? | 14:50 |
| corvus | mordred: we'll want this job to be in the zuul tenant, and all the opendev gerrit stuff is currently in the openstack tenant | 14:50 |
| mordred | ah - hrm. interesting wrinkle | 14:51 |
| *** surpatil has quit IRC | 14:51 | |
| fungi | though we probably need an answer in our back pocket for when gerrit folks ask why we didn't just use their container image | 14:51 |
| corvus | mordred: so i guess we'll need to add the system-config repo to the zuul tenant to get that job... | 14:51 |
| mordred | fungi: we have one - theirs is aimed at a different use case | 14:51 |
| mordred | fungi: and we're building in plugins - it's all good on that front | 14:51 |
| fungi | i figured there was a straightforward answer | 14:51 |
| corvus | mordred: specifically -- i'm not sure upstream builds in the checks plugin | 14:52 |
| mordred | corvus: yeah. and we can have the opendev tenant publish master images | 14:52 |
| sean-k-mooney | oh speaking of building thigs form master is there a special process for adding non openstack git repos to upstream zuul. i want to add openvswitch and dpdk at some point in the nearish future | 14:52 |
| mordred | but the zuul tenant can just use the job for dependencies | 14:52 |
| mordred | corvus: I'm landing and have a 1 hour layer - but I can totally make some jobs as soon as I'm airborn again - but if that's a blocker for next steps for you I won't be offended if you do that too | 14:53 |
| fungi | sean-k-mooney: https://opendev.org/openstack/project-config/src/branch/master/zuul/main.yaml#L37-L45 | 14:53 |
| corvus | mordred: cool, i'll start work on getting all the repos added to the zuul tenant (and also moving that third-party-check pipeline from opendev (which has none of the projects we're interested in) to zuul | 14:53 |
| mordred | corvus: maybe make the system-config zuul tenant patch in the meantime? | 14:53 |
| mordred | yeah | 14:53 |
| corvus | mordred: ++ | 14:54 |
| sean-k-mooney | fungi: yep i was looking at that | 14:54 |
| *** liuyulong has quit IRC | 14:54 | |
| mordred | corvus: cool. and the checks-plugin repo I suppose :) | 14:54 |
| sean-k-mooney | fungi: just wondered if there was any paper work i needed to do | 14:54 |
| mordred | and then we'll get a master job added and that'll be cool | 14:54 |
| sean-k-mooney | fungi: or if i could just propose a patch when i needed to add it | 14:54 |
| fungi | sean-k-mooney: er, this section actually for the openstack tenant: https://opendev.org/openstack/project-config/src/branch/master/zuul/main.yaml#L1608-L1631 | 14:55 |
| fungi | sean-k-mooney: i think just explain in the commit message | 14:55 |
| sean-k-mooney | ok | 14:55 |
| sean-k-mooney | ill be working on an ovs-dpdk job later this week. | 14:55 |
| corvus | sean-k-mooney: https://docs.openstack.org/infra/manual/drivers.html#outbound-third-party-testing | 14:55 |
| sean-k-mooney | currently the devstack plugin i mainain just hits github directly | 14:56 |
| *** bdodd has joined #openstack-infra | 14:56 | |
| sean-k-mooney | it can use a local dir but to do that i need to have zuul clone it | 14:56 |
| fungi | ahh, yeah, i should have clarified whether you want to use those repos in jobs reported to gerrit, or actually report on pull requests/changes for those repos | 14:57 |
| fungi | i was assuming the former, corvus linked to documentation for the latter; | 14:57 |
| corvus | all the infos are there now :) | 14:57 |
| sean-k-mooney | ya i want to have the job voting on x/networking-ovs-dpdk and os-vif and maybe nova in the future | 14:57 |
| corvus | so probably fungi's thing, but "the more you know" | 14:58 |
| sean-k-mooney | basicaly replaceing the intel third party ci with a first party one | 14:58 |
| fungi | corvus: ahh, nice! | 14:58 |
| *** beekneemech is now known as bnemec | 15:01 | |
| openstackgerrit | James E. Blair proposed openstack/project-config master: Add gerrit-related projects to zuul tenant https://review.opendev.org/688423 | 15:08 |
| clarkb | infra-root I'm going to clean up all of my test resources in FN now (servers, images, and keys) | 15:10 |
| clarkb | * once I load my ssh keys | 15:10 |
| fungi | clarkb: if i still have a test instance in there, it can be deleted too | 15:11 |
| clarkb | k | 15:11 |
| fungi | can't remember if i cleaned it up from when i was doing packet captures there | 15:11 |
| *** udesale has joined #openstack-infra | 15:12 | |
| clarkb | fungi: I don't see it. Was it a nodepool held node or one you booted directly? If it wasn't a held node then I think it has been cleaned up already | 15:15 |
| clarkb | I only see 2 images and 2 nodes from ianw which I'll leave in place for now | 15:15 |
| clarkb | (I cleared out my test resources) | 15:15 |
| *** ttx has quit IRC | 15:18 | |
| fungi | it was one i booted directly while we were at ansiblefest | 15:19 |
| fungi | so i guess i deleted it after i stopped the packet capture | 15:19 |
| *** ykarel is now known as ykarel|afk | 15:21 | |
| *** mattw4 has joined #openstack-infra | 15:26 | |
| *** xek_ has quit IRC | 15:26 | |
| *** e0ne has joined #openstack-infra | 15:28 | |
| *** ttx has joined #openstack-infra | 15:29 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 15:31 |
| openstackgerrit | James E. Blair proposed opendev/project-config master: Remove third-party-check pipeline https://review.opendev.org/688427 | 15:31 |
| *** elod has quit IRC | 15:31 | |
| *** e0ne has quit IRC | 15:34 | |
| zbr | ianw: my configure mirrors patch keeps growing as testing it uncovers other bugs. | 15:36 |
| *** jtomasek has quit IRC | 15:38 | |
| *** elod has joined #openstack-infra | 15:42 | |
| *** jtomasek_ has joined #openstack-infra | 15:43 | |
| *** lucasagomes has quit IRC | 15:45 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 15:46 |
| *** ccamacho has quit IRC | 15:47 | |
| clarkb | zbr: can you see comments on ps13 and ps18? | 15:48 |
| zbr | clarkb: yep. and yes, if you use omit, "var is defined" will be true! in fact it would be an omit__.... string. | 15:49 |
| clarkb | TIL | 15:49 |
| zbr | (obviously that i plan to split this in smaller bits, once I get it working) | 15:49 |
| *** mattw4 has quit IRC | 15:52 | |
| *** mattw4 has joined #openstack-infra | 15:56 | |
| corvus | Shrews: https://review.opendev.org/686761 worked as expected | 15:57 |
| corvus | Shrews: or, rather, the change we merged before it | 15:57 |
| corvus | (the one that set diskimages to []) | 15:57 |
| fungi | #status log the trove instance for review-dev-MySQL was rebooted around 04:00 utc by the provider for cold migration due to impending host failure, trouble ticket #191014-ord-0000099 | 16:00 |
| * fungi taps foot | 16:00 | |
| * fungi twiddles thumbs | 16:00 | |
| * fungi goes looking for openstackstatus | 16:00 | |
| corvus | does anyone know why debian-buster and opensuse-423 have not been building (in 41 and 188 days, respectively)? | 16:01 |
| clarkb | corvus: buster because our mirror is broken | 16:02 |
| clarkb | reprepro is not updating the package index when it updates the packages and no one knows why | 16:02 |
| clarkb | suse 423 I don't know | 16:02 |
| *** rpioso is now known as rpioso|afk | 16:02 | |
| *** openstackstatus has joined #openstack-infra | 16:03 | |
| *** ChanServ sets mode: +v openstackstatus | 16:03 | |
| fungi | #status log statusbot restarted following its disappearance associated with a ping timeout at 13:15 utc, no indication of problems in its debug log to explain | 16:03 |
| openstackstatus | fungi: finished logging | 16:03 |
| fungi | #status log the trove instance for review-dev-MySQL was rebooted around 04:00 utc by the provider for cold migration due to impending host failure, trouble ticket #191014-ord-0000099 | 16:03 |
| openstackstatus | fungi: finished logging | 16:03 |
| *** e0ne has joined #openstack-infra | 16:03 | |
| corvus | should we remove those images from our system? | 16:04 |
| openstackgerrit | Merged openstack/project-config master: Remove arm64ci-nrt1 (2/2) https://review.opendev.org/686761 | 16:05 |
| fungi | i didn't realize the reprepro problem was breaking image builds, i'll take a quick look at the log | 16:05 |
| fungi | 2019-10-14 14:38:05.465 | W: Couldn't download package base-files (ver 10.3 arch amd64) at http://mirror.dfw.rax.openstack.org/debian/pool/main/b/base-files/base-files_10.3_amd64.deb | 16:06 |
| fungi | yeah, looks like it probably is | 16:06 |
| clarkb | the opensuse 15 images should've replaced the 423 image. I know AJaeger and dirk were working on uplifting jobs | 16:06 |
| corvus | clarkb: yeah, that's what i'm getting from looking at comments in git repos | 16:06 |
| *** iurygregory has quit IRC | 16:06 | |
| *** rpittau is now known as rpittau|afk | 16:06 | |
| *** ociuhandu_ has joined #openstack-infra | 16:07 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 16:07 |
| corvus | clarkb: i think they may have completed that work -- short of actually removing it. i will do so. | 16:07 |
| openstackgerrit | Jan Gutter proposed zuul/nodepool master: Fix typo in port-cleanup-interval description https://review.opendev.org/688440 | 16:07 |
| clarkb | corvus: ok | 16:07 |
| fungi | i fully support removing the debian-buster images from our config until someone is able to figure out what's wrong with reprepro there | 16:07 |
| *** dtantsur|brb is now known as dtantsur | 16:08 | |
| openstackgerrit | James E. Blair proposed openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 16:09 |
| openstackgerrit | James E. Blair proposed opendev/base-jobs master: Remove opensuse-423 nodeset https://review.opendev.org/688442 | 16:10 |
| *** ociuhandu has quit IRC | 16:10 | |
| clarkb | corvus: zuul managed to catch at least one user of that, I assume we can clean it up thoug (not sure why bifrost would need to keep running on really old suse) | 16:10 |
| *** ociuhandu_ has quit IRC | 16:14 | |
| AJaeger | corvus, clarkb https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 is the list of open changes for removing 42.3 from jobs | 16:16 |
| AJaeger | I fear a few more are missing - we hit a lot of broken branches ;( | 16:18 |
| corvus | shall i start force-merging those? | 16:18 |
| AJaeger | I was just going to raise that question ;) | 16:18 |
| corvus | or should we just force-merge the project-config change and leave the projects broken | 16:18 |
| AJaeger | I'm happy to provide all the missing changes on all branches, if we force merge them... | 16:18 |
| AJaeger | I would rather force-merge into projects and have Zuul not complain... | 16:19 |
| AJaeger | dirk1, do you agree? ^ | 16:19 |
| openstackgerrit | James E. Blair proposed openstack/project-config master: Remove references to OpenSUSE 423 https://review.opendev.org/688443 | 16:20 |
| openstackgerrit | James E. Blair proposed openstack/project-config master: Remove opensuse-423 from nodepool https://review.opendev.org/688444 | 16:20 |
| *** igordc has joined #openstack-infra | 16:21 | |
| *** ykarel|afk has quit IRC | 16:21 | |
| *** kjackal has quit IRC | 16:21 | |
| corvus | AJaeger: i think at this point i'm confident that all the non-broken images have been updated and we can merge https://review.opendev.org/680696 | 16:21 |
| corvus | AJaeger: (the only out-of-date images now are opensuse-423 and debian-buster) | 16:22 |
| clarkb | re buster, fungi do you think we can rm (or mv) the broken index files and force reprepro to rebuild them that way? | 16:22 |
| corvus | AJaeger: and i think we're considering them broken enough not to worry | 16:22 |
| AJaeger | corvus: agreed | 16:23 |
| AJaeger | corvus: go ahead and change your WIP to a +A ;) | 16:23 |
| dirk1 | Ajaeger: sure, better get rid of this sooner rather than later | 16:23 |
| corvus | AJaeger: i changed it to a +0 -- i'll give clarkb and fungi a quick chance to halt it in case they want to hold it for buster: https://review.opendev.org/680696 | 16:24 |
| AJaeger | mnaser: are you fine to have corvus force merge these open changes https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 ? | 16:25 |
| AJaeger | dirk1: thanks | 16:25 |
| clarkb | I'm not aware of working jobs on buster | 16:25 |
| fungi | clarkb: i'm less confused the indices are broken, and starting to suspect that we simply aren't including buster-updates in the sources list | 16:25 |
| clarkb | (due to the mirror trouble) | 16:25 |
| mnaser | AJaeger: except for bifrost, i am not opposed | 16:25 |
| fungi | s/confused/convinced/ | 16:25 |
| mnaser | thats not my say :) | 16:25 |
| fungi | i'm looking now to see if i can figure out where the sources list is defined for these image builds | 16:25 |
| AJaeger | mnaser: sure -thanks | 16:26 |
| fungi | doesn't seem to be in diskimage-builder or project-config but maybe i'm not grepping for the right thing | 16:26 |
| AJaeger | corvus: want to force merge the openstack-ansible ones? | 16:26 |
| clarkb | fungi: /var/log/reprepro/debian-buster-updates.log says we do | 16:26 |
| corvus | AJaeger: will do | 16:27 |
| clarkb | fungi: I know that I compared indice to package dir and the files did not match | 16:27 |
| clarkb | fungi: we also confirmed that the broken jobs that were running had updates enabled | 16:27 |
| clarkb | I don't think this is a missing updates repo on end nodes problem | 16:27 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Added Docker image builds https://review.opendev.org/688446 | 16:27 |
| *** eharney has quit IRC | 16:28 | |
| corvus | AJaeger: The opensuse-423 nodeset is being removed. This change is being force-merged to correct the resulting configuration error. | 16:28 |
| corvus | AJaeger: that message look good? | 16:28 |
| AJaeger | yes, it does. | 16:28 |
| AJaeger | corvus: let me iterate on https://review.opendev.org/#/c/688441 and figure out what else is missing - I'll add depends-on to it... | 16:29 |
| fungi | clarkb: taking the first example error from the image build log, it's trying to install base-files_10.3_amd64.deb but reprepro replaced that with base-files_10.3+deb10u1_amd64.deb on 2019-09-07 | 16:31 |
| clarkb | fungi: ya and the reason it fails is because base-files_10.3+deb10u1_amd64.deb is what we have on disk if this is the same behavior we have observed before | 16:31 |
| clarkb | but the index still points to the old file | 16:31 |
| clarkb | the old file no longer exists | 16:31 |
| fungi | however, i think base-files_10.3+deb10u1_amd64.deb is likely in the buster-updates index and not in the original release index | 16:32 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 16:32 |
| corvus | AJaeger: https://review.opendev.org/#/c/677415/2 depends on a change i don't want to force-merge | 16:32 |
| corvus | AJaeger: i'll see about rebasing it | 16:34 |
| clarkb | fungi: 2019-10-14 14:37:38.826 | updates:deb http://mirror.dfw.rax.openstack.org/debian buster-updates main it seems to try and use that as part of its apt config | 16:34 |
| *** udesale has quit IRC | 16:34 | |
| corvus | AJaeger: down to 1: https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 | 16:35 |
| *** pgaxatte has quit IRC | 16:35 | |
| clarkb | fungi: also http://mirror.dfw.rax.openstack.org/debian/dists/buster-updates/main/binary-amd64/Packages has no base-files in it | 16:36 |
| fungi | clarkb: yeah, i find it here: https://cdn-aws.deb.debian.org/debian/dists/buster/main/binary-amd64/Packages.xz | 16:36 |
| fungi | so not in buster-updates but in the main buster index | 16:36 |
| *** ramishra has quit IRC | 16:36 | |
| AJaeger | yeah - will work on the remaining bifrost ones in the next hours | 16:36 |
| fungi | Version: 10.3+deb10u1 | 16:36 |
| fungi | so i agree, something is causing it to not write that index | 16:37 |
| clarkb | which is not what we have in our index | 16:37 |
| AJaeger | corvus: I'll tell you once I'm done... | 16:37 |
| corvus | AJaeger, dirk1: i'm uncomfortable force-merging https://review.opendev.org/#/c/677181/ -- i only want to force-merge changes that simply remove jobs. so either we should get a change that fixes jobs and merge it in the regular way, or make a new change that simply removes 423 without changing anything else, and i can force-merge that. | 16:37 |
| *** michael-beaver has quit IRC | 16:37 | |
| clarkb | fungi: best I can tell is reprepro is downloading the pcakge files but not making the corresponding updates to the repo indexes. I've read through manpages and a couple howtos for reprepro and from everythign I can see reprepro is supposed to be designed to avoid thsi exact problem | 16:38 |
| clarkb | (it builds indexes based on what it actually has) | 16:38 |
| clarkb | leaves me super confused as to how it gets into this state | 16:38 |
| dirk1 | corvus: agreed. Let me take a look | 16:38 |
| dirk1 | Fyi, openSUSE-150 is also about to be end of life so we definitely need to speed in fixing things | 16:39 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Added Docker image builds https://review.opendev.org/688446 | 16:40 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Fix documentation builds https://review.opendev.org/688450 | 16:41 |
| openstackgerrit | James E. Blair proposed openstack/project-config master: Move gear to opendev tenant https://review.opendev.org/688451 | 16:42 |
| *** xek_ has joined #openstack-infra | 16:42 | |
| *** odicha has joined #openstack-infra | 16:42 | |
| *** xek_ has quit IRC | 16:42 | |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 16:43 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 16:43 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Added Docker image builds https://review.opendev.org/688446 | 16:43 |
| *** odicha has quit IRC | 16:43 | |
| mordred | corvus: back online - I miss anything good/ | 16:43 |
| mordred | ? | 16:43 |
| *** xek has joined #openstack-infra | 16:44 | |
| AJaeger | corvus: 51 config-errors - a few more changes to merge than expected ;( http://zuul.opendev.org/t/openstack/config-errors | 16:45 |
| clarkb | mordred: any idea why reprepro wouldn't update the buster index file(s) after downloading newer packages (and deleting the ones the index it has points to)? | 16:45 |
| fungi | clarkb: i'm kinda tempted to wipe the current buster mirrors and let them get rebuilt by reprepro, in case it's gotten itself in some weird state. but regardless i expect we ought to pause or remove the buster images from nodepool in the meantime | 16:45 |
| mordred | clarkb: nope. | 16:46 |
| clarkb | fungi: it might be easier to test if things are working if we keep it in nodepool? | 16:46 |
| mordred | also - I kinda think trying fungi's thing might not be terrible | 16:46 |
| *** kjackal has joined #openstack-infra | 16:47 | |
| fungi | clarkb: do we have dib jobs which try to build a buster image? | 16:47 |
| * fungi checks | 16:48 | |
| fungi | if so that should tell us whether it works regardless of whether there's an image available in opendev | 16:48 |
| clarkb | no we don't appear to | 16:48 |
| clarkb | at least they don't run regularly on changes to dib | 16:48 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Fix documentation builds https://review.opendev.org/688450 | 16:48 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Added Docker image builds https://review.opendev.org/688446 | 16:48 |
| ralonsoh | hi folks, is there a way to download a log directory recursively. E.g: storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_847/681893/1/gate/neutron-fullstack/847f4d9/controller/logs/dsvm-fullstack-logs/ | 16:49 |
| *** xek has quit IRC | 16:49 | |
| AJaeger | TheJulia, dtantsur: are you fine with force merging a couple of job removals for openSUSE 42.3 for bifrost? See https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 | 16:50 |
| dtantsur | AJaeger: if it's urgent - yes. if not, I'd rather wait for the CI (and force merge if it's broken). | 16:50 |
| openstackgerrit | James E. Blair proposed opendev/gear master: Add in-repo zuul config https://review.opendev.org/688455 | 16:50 |
| fungi | clarkb: agreed, not finding one | 16:50 |
| clarkb | ralonsoh: wget -r and associated options can do it | 16:50 |
| corvus | AJaeger: i'm not going to force-merge 51 changes. | 16:51 |
| ralonsoh | clarkb, I tried "wget -r -l 0 --no-parent --reject "index.html*" -e robots=off" | 16:51 |
| ralonsoh | clarkb, no luck | 16:51 |
| clarkb | ralonsoh: how does it fail? | 16:51 |
| ralonsoh | clarkb, it retrieves only.. | 16:51 |
| ralonsoh | storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f725 100%[========================================================================================================================================>] 5.18K --.-KB/s in 0s | 16:51 |
| ralonsoh | this file | 16:51 |
| ralonsoh | Saving to: ‘storage.gra1.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_847/681893/1/gate/neutron-fullstack/847f4d9/controller/logs/dsvm-fullstack-logs/index.html.tmp’ | 16:52 |
| AJaeger | corvus: how do you want to handle those then? I know it's a mess ;( | 16:53 |
| clarkb | ralonsoh: reading the man page I have -l 0 means depth of 0 | 16:53 |
| clarkb | -l inf is infinite | 16:53 |
| clarkb | maybe try set that to 3 and rerun? | 16:53 |
| ralonsoh | the same | 16:54 |
| AJaeger | corvus: let's see what just passes... | 16:54 |
| *** jpena is now known as jpena|off | 16:55 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: add support for centos-8 https://review.opendev.org/688118 | 16:56 |
| clarkb | ralonsoh: I think the issue is they are compressed | 16:56 |
| openstackgerrit | Merged openstack/project-config master: Add gerrit-related projects to zuul tenant https://review.opendev.org/688423 | 16:56 |
| ralonsoh | clarkb, but I don't understand, should be the same for downloading, shouldn't it? | 16:57 |
| *** dirk1 is now known as dirk | 16:57 | |
| *** jtomasek_ has quit IRC | 16:58 | |
| clarkb | ralonsoh: in order to crawl the html wget must be able to find urls in the returned content. If that content is compressed and wget doesn't know how to decompress it then it cannot find those urls | 16:59 |
| clarkb | ralonsoh: supposedly new wget knows how to decompress, but I'm not having any luck getting it to work yet | 16:59 |
| *** jtomasek has joined #openstack-infra | 17:00 | |
| *** e0ne has quit IRC | 17:00 | |
| *** derekh has quit IRC | 17:00 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: configure-mirrors: fix mirror_fqdn variable https://review.opendev.org/688468 | 17:02 |
| clarkb | ralonsoh: ah I see that is an older job so everything is deflate encoded and wget supposedly only knows gzip | 17:02 |
| clarkb | let me try it on a more recent job | 17:02 |
| AJaeger | corvus: sorry, wasn't aware that pike and queens were not done ;( | 17:02 |
| clarkb | ralonsoh: ya that works | 17:03 |
| ralonsoh | clarkb, so I can't download all jobs | 17:03 |
| clarkb | ralonsoh: wget --compression=gzip --mirror $URL | 17:03 |
| ralonsoh | *old | 17:03 |
| clarkb | ralonsoh: we updated the compression type last friday | 17:03 |
| ralonsoh | ahhhh ok | 17:03 |
| *** ykarel|afk has joined #openstack-infra | 17:04 | |
| clarkb | jobs that have run since then should work if you use the above command and have wget 1.19.2 and newer (I think) | 17:04 |
| openstackgerrit | Mohammed Naser proposed opendev/gear master: Add BSD/Darwin support. https://review.opendev.org/671674 | 17:04 |
| clarkb | mnaser: fyi rebasing isn't necessary to get the doc build fixes | 17:05 |
| clarkb | mnaser: once the doc build fixes merge you can simply recheck older changes | 17:06 |
| mnaser | clarkb: yeah i figured i'd do that earlier because that patch is +W'd anyways | 17:06 |
| mnaser | just to get a result faster, unless that has some other implication i dont know about :) | 17:06 |
| clarkb | rebasing like that removes all votes | 17:06 |
| clarkb | it isn't +W'd | 17:07 |
| mnaser | uh thats' weird | 17:07 |
| clarkb | it also dirties the review history (inter patchset diffs in particular) | 17:07 |
| mnaser | i could swear i've rebased without these disappearing | 17:07 |
| mnaser | sorry about that then :) i'll avoid that in the future | 17:07 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul-jobs master: config-mirrors: Consolidate dnf/yum handler https://review.opendev.org/688479 | 17:09 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Expose volumes for the gerrit hooks, static and tmp dirs https://review.opendev.org/678875 | 17:10 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add jobs to build gerrit master branch https://review.opendev.org/688482 | 17:10 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add checks plugin to gerrit master build https://review.opendev.org/688483 | 17:10 |
| *** priteau has quit IRC | 17:11 | |
| fungi | #status log moved /afs/.openstack.org/mirror/debian/dists/buster* to /root/tmp/ on mirror-update01.openstack.org and invoked reprepro-mirror-update for debian in a root screen session | 17:12 |
| openstackstatus | fungi: finished logging | 17:12 |
| clarkb | fungi: did you remove the vos release? | 17:13 |
| clarkb | (good chance that will timeout) | 17:13 |
| openstackgerrit | Merged opendev/gear master: Fix documentation builds https://review.opendev.org/688450 | 17:13 |
| mordred | corvus, clarkb, fungi: there's patches to get us master builds of gerrit, as well as adding checks plugins. I stacked them on top of two other gerrit images patches that are ready to go - mostly as a motivation to get those landed :) | 17:13 |
| fungi | clarkb: i didn't remove packages, just indices | 17:13 |
| fungi | so hopefully it won't take long | 17:13 |
| clarkb | fungi: ah | 17:13 |
| fungi | i want to see if it generates all new indices, and how they compare to the ones i moved out of the way | 17:14 |
| clarkb | fungi: ++ | 17:14 |
| *** e0ne has joined #openstack-infra | 17:16 | |
| fungi | clarkb: so this is interesting... | 17:20 |
| fungi | it finished, but created no dists directories at all for buster: http://mirror.dfw.rax.openstack.org/debian/dists/ | 17:20 |
| mordred | maybe buster doesn't actually exist | 17:21 |
| *** e0ne has quit IRC | 17:22 | |
| *** signed8bit has joined #openstack-infra | 17:22 | |
| clarkb | that contiues to match previously observed behavior (no package index updates) | 17:22 |
| clarkb | does it exit 0 ? | 17:22 |
| AJaeger | mnaser: I pushed all the other opensuse 42.3 changes, could you approve (or ask the ansible team to) those that pass, please? See https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 | 17:23 |
| jrosser | AJaeger: I can keep an eye on those | 17:24 |
| AJaeger | jrosser: thanks! | 17:24 |
| AJaeger | jrosser: mnaser started already, so better coordinate. Thanks to both of you! | 17:24 |
| fungi | clarkb: yep, and proceeds with the vos release | 17:24 |
| *** gmann is now known as gmann_afk | 17:25 | |
| *** tesseract has quit IRC | 17:25 | |
| mnaser | AJaeger: im totally ok with force merging those before ci runs | 17:27 |
| mnaser | i am doing a second eye on them | 17:27 |
| AJaeger | mnaser: thanks - all depends on whether an infra-root has time for force merge the rest | 17:28 |
| *** lpetrut has quit IRC | 17:28 | |
| AJaeger | mnaser: I would merge those without force that pass... | 17:28 |
| openstackgerrit | Merged openstack/project-config master: Move gear to opendev tenant https://review.opendev.org/688451 | 17:29 |
| fungi | clarkb: i've tried moving the base-files buster packages out of the pool now to see if reprepro will replace them | 17:31 |
| fungi | during the "Checking state of mirror" phase it spots them as absent | 17:33 |
| *** ricolin has quit IRC | 17:33 | |
| clarkb | how would it know they aren't absent without indexes? | 17:33 |
| clarkb | is it doing that check against upstream's package list? | 17:33 |
| fungi | i think so | 17:33 |
| *** dtantsur is now known as dtantsur|afk | 17:36 | |
| fungi | i must really not understand how reprepro is supposed to work. it didn't replace the missing files | 17:39 |
| jrosser | fungi: if you have an example command I have storage and bandwidth to try it out in a different environment | 17:46 |
| fungi | jrosser: well, we have a wrapper script and a pile of conffiles to go with them, so i'm still teasing all that apart | 17:47 |
| fungi | but i appreciate the offer, will let you know if it's worth trying | 17:47 |
| jrosser | sure, no problem | 17:47 |
| AJaeger | jrosser: could you review https://review.opendev.org/#/c/688519/ , please? That one is still open... | 17:48 |
| fungi | the reprepro manpage mentions a lockfile and a database, but doesn't indicate where it stores them by default (and we don't seem to override it as far as i can see) so i'm trying to track those down now | 17:48 |
| jrosser | AJaeger: done | 17:50 |
| fungi | aha, they're written to the base path for the target mirror | 17:50 |
| fungi | so /afs/.openstack.org/mirror/debian/db | 17:50 |
| *** verdurin has quit IRC | 17:51 | |
| *** verdurin has joined #openstack-infra | 17:51 | |
| fungi | ooh, i just noticed that our reprepro config has both stretch and buster listed as "Suite: stable" when the former should be "Suite: oldstable" now | 17:53 |
| *** e0ne has joined #openstack-infra | 17:54 | |
| fungi | trying a quick manual edit to see if that gets things moving again | 17:55 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Expose volumes for the gerrit hooks, static and tmp dirs https://review.opendev.org/678875 | 17:57 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add jobs to build gerrit master branch https://review.opendev.org/688482 | 17:57 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add checks plugin to gerrit master build https://review.opendev.org/688483 | 17:57 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Use bazel 1.0.0 https://review.opendev.org/688523 | 17:57 |
| *** e0ne has quit IRC | 17:59 | |
| mordred | corvus: bazel 1.0.0 is out now ^^ | 17:59 |
| corvus | mordred: is the bazel release why the jobs failed? | 17:59 |
| mordred | corvus: yup! | 18:00 |
| corvus | cool, was just about to start digging into that, glad i don't have to :) | 18:00 |
| * mordred has provided valkue | 18:00 | |
| corvus | mordred: qq on 678875 -- we don't need *volume* stanzas to bind-mount stuff in -- so why should we make volumes there? | 18:01 |
| corvus | mordred: i grok tmp -- that's for gerrit to write semi-persistent state to (weirdly named though it is) | 18:01 |
| *** psachin has quit IRC | 18:05 | |
| *** ralonsoh has quit IRC | 18:12 | |
| corvus | mordred: and 'stable-master' does not exist :) <insert project-is-not-gated-by-zuul joke here> | 18:12 |
| fungi | #status log moved /afs/.openstack.org/mirror/debian/db to /root/tmp/ on mirror-update01.openstack.org and invoked reprepro-mirror-update for debian in a root screen session | 18:15 |
| openstackstatus | fungi: finished logging | 18:15 |
| *** rpioso|afk is now known as rpioso | 18:15 | |
| clarkb | fungi: that is with the stable vs oldstable fix? | 18:16 |
| fungi | clarkb: yeah, just tweaking the suite name doesn't seem to have made any difference, but now that i know it keeps state in a database i'm going to see what it does when there's no previous database | 18:17 |
| clarkb | ah | 18:17 |
| fungi | seems the reason the check subcommand reports the package missing when i remove it is that it has record of already downloading it | 18:18 |
| AJaeger | thanks, jrosser | 18:19 |
| fungi | so doesn't redownload it, just complains | 18:19 |
| clarkb | fungi: db may be the culprit then | 18:23 |
| fungi | that's what i'm hoping anyway | 18:25 |
| * SpamapS would like to reiterate that alpine's dev process ~18 months ago was absolutely terrifying opaque, and I don't think I'll ever trust them to fix it. | 18:27 | |
| SpamapS | terrifyingly I guess | 18:27 |
| AJaeger | Zuul is complaining in openstack tenant about " Unknown project opendev/gear" - did we change config? | 18:28 |
| SpamapS | IMO, they do not know how to scale that community, they don't know how to be transparent about how to help, and while I can understand maybe there's some issues with glibc, musl isn't enough reason to just start blindly trusting a bunch of peeps who have no idea how to vet devs on the internet. | 18:28 |
| *** e0ne has joined #openstack-infra | 18:28 | |
| clarkb | AJaeger: https://review.opendev.org/688451 | 18:29 |
| clarkb | AJaeger: I'm guessing some other config was missed (maybe required-projects listing?) | 18:29 |
| AJaeger | thanks, clarkb ! corvus, will you add gear back to openstack so that requireed-project works? ^ | 18:29 |
| AJaeger | clarkb: agreed! | 18:29 |
| AJaeger | corvus: or shall I? | 18:29 |
| AJaeger | corvus: I'll do it... | 18:30 |
| *** e0ne has quit IRC | 18:30 | |
| corvus | AJaeger: thanks -- we should just add it with "include: []" | 18:31 |
| corvus | pabelanger: you may want to move windmill to the opendev tenant :) | 18:32 |
| AJaeger | corvus: agreed | 18:32 |
| AJaeger | corvus, clarkb ^ | 18:32 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Add gear to openstack for required-projects https://review.opendev.org/688530 | 18:32 |
| AJaeger | too fast ^ | 18:32 |
| corvus | AJaeger: +3 | 18:33 |
| AJaeger | thanks | 18:34 |
| openstackgerrit | Matt Riedemann proposed opendev/elastic-recheck master: Add query for cirros guest dhcp lease fail bug 1848078 https://review.opendev.org/688532 | 18:39 |
| openstack | bug 1848078 in OpenStack-Gate "ssh fails in tempest runs when guest dhcp lease fails" [Undecided,Confirmed] https://launchpad.net/bugs/1848078 | 18:39 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Don't install zuul in the bindep tox testenv https://review.opendev.org/652251 | 18:43 |
| *** factor has quit IRC | 18:50 | |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Add nodesets API route https://review.opendev.org/605877 | 18:51 |
| openstackgerrit | Matt Riedemann proposed opendev/elastic-recheck master: Remove old queries: 2019-10-14 https://review.opendev.org/688535 | 18:54 |
| *** kopecmartin is now known as kopecmartin|off | 19:00 | |
| openstackgerrit | Merged openstack/project-config master: Add gear to openstack for required-projects https://review.opendev.org/688530 | 19:07 |
| clarkb | fungi: is it still going? I Guess it has to redo everything without its db? | 19:08 |
| fungi | yep | 19:08 |
| fungi | still going | 19:08 |
| openstackgerrit | Merged opendev/elastic-recheck master: Add query for cirros guest dhcp lease fail bug 1848078 https://review.opendev.org/688532 | 19:10 |
| openstack | bug 1848078 in OpenStack-Gate "ssh fails in tempest runs when guest dhcp lease fails" [Undecided,Confirmed] https://launchpad.net/bugs/1848078 | 19:10 |
| openstackgerrit | Tobias Henkel proposed zuul/zuul master: Extract an abstract base Parser class https://review.opendev.org/545610 | 19:10 |
| *** ykarel|afk has quit IRC | 19:14 | |
| AJaeger | I'm still seeing the windmill errors on gear after 688530 merged - was my change wrong? http://zuul.opendev.org/t/openstack/config-errors | 19:14 |
| AJaeger | or does it take that much longer for Zuul to figure it out? | 19:15 |
| *** e0ne has joined #openstack-infra | 19:16 | |
| fungi | AJaeger: changes to tenant configuration don't take effect immediately on merge | 19:18 |
| AJaeger | fungi: ah! thanks | 19:19 |
| fungi | requires a full reconfigure triggered by configuration management | 19:19 |
| *** dave-mccowan has joined #openstack-infra | 19:19 | |
| AJaeger | that explains it | 19:19 |
| AJaeger | corvus: looks like around 20 changes might need force merge, see https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 - some still in the queue... | 19:20 |
| *** bnemec has quit IRC | 19:22 | |
| openstackgerrit | Merged zuul/nodepool master: Fix typo in port-cleanup-interval description https://review.opendev.org/688440 | 19:23 |
| *** Jeffrey4l has quit IRC | 19:23 | |
| *** Jeffrey4l has joined #openstack-infra | 19:24 | |
| *** dave-mccowan has quit IRC | 19:24 | |
| *** whoami-rajat has quit IRC | 19:41 | |
| AJaeger | corvus: 22 config errors... | 19:46 |
| AJaeger | fungi: reconfigured got triggered | 19:46 |
| corvus | AJaeger: we can revert the change that removed the osa job defs, or force-merge removals | 19:48 |
| AJaeger | corvus: so out of those 50 changes, more than half passed and merged, so I propose to force move the rest... | 19:51 |
| AJaeger | ... passed and merged | 19:51 |
| corvus | AJaeger: so you want to force-merge all of these? https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 | 19:52 |
| AJaeger | corvus: there're a few (have no zuul votes) that are in gate right now, so I would not force-merge those. But yes, for the rest. Hope that is it (I fear it's not) | 19:53 |
| corvus | hrm. i didn't really mean to sign up for this | 19:54 |
| AJaeger | there're two in the gate | 19:54 |
| corvus | i'd rather just revert the change that broke the config | 19:54 |
| AJaeger | any other infra-root that can help? Or what should be done? | 19:54 |
| corvus | i believe this revert will fix the config errors: https://review.opendev.org/#/c/688550/ | 19:55 |
| corvus | i plan to force-merge that unless another infra-root wants to take over pulling on this thread | 19:55 |
| *** bnemec has joined #openstack-infra | 19:56 | |
| clarkb | the pulling required is force merging updates to osa on stable branches? | 19:56 |
| corvus | clarkb: that's the current assumption -- see "Hope that is it (I fear it's not)" from AJaeger earlier | 19:56 |
| corvus | clarkb: also, some bifrost | 19:56 |
| *** dpawlik has joined #openstack-infra | 19:56 | |
| clarkb | I can help after lunch | 19:57 |
| *** gfidente has quit IRC | 19:57 | |
| clarkb | I expect if we dont force merge we'll nevermerge the changes (and be stuck with 423?) | 19:58 |
| corvus | honestly, i think even if someone wants to continue pulling on the thread, we should revert the change that removed the osa jobs -- we should be able to verify it's ready | 19:58 |
| corvus | so i'm going to do that | 19:58 |
| corvus | AJaeger, clarkb: i merged https://review.opendev.org/#/c/688550/ -- that should reduce the list of zuul errors, then someone can propose a revert-revert of that to see if we've caught all the uses. | 19:59 |
| AJaeger | corvus: so, revert-revert with depends-on? | 20:00 |
| corvus | AJaeger: when the ones in gate have merged, why don't you let us know, maybe clarkb and i can split the force-merging. | 20:00 |
| AJaeger | corvus: will do, thanks | 20:00 |
| corvus | AJaeger: maybe don't set a depends-on if we're going to force merge. that's probably how we got into this in the first place. | 20:00 |
| AJaeger | corvus: then I won't be able to see whether all is in, or will I? | 20:01 |
| *** gmann_afk is now known as gmann | 20:01 | |
| corvus | AJaeger: depends on what you want to do. at any rate, we should not force-merge a change like https://review.opendev.org/#/c/664658/ because the depends-on masks errors | 20:02 |
| corvus | AJaeger: so use depends-on if you want to check that you've got everything, but remove them if you want us to force-merge the change itself. better yet is if we don't need to force-merge that one. | 20:03 |
| AJaeger | corvus: ok | 20:03 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 20:11 |
| AJaeger | corvus: https://review.opendev.org/#/c/688553/ looks good now - and 688441 is the second sanity-check ;) | 20:12 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add jobs to build gerrit master branch https://review.opendev.org/688482 | 20:12 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add checks plugin to gerrit master build https://review.opendev.org/688483 | 20:12 |
| mordred | are our puppet ci jobs just broken? | 20:14 |
| mordred | all of the docker patches are failing them | 20:15 |
| mordred | "Gem::InstallError: dry-inflector requires Ruby version >= 2.4." | 20:15 |
| AJaeger | mmh, "This change depends on a change with an invalid configuration." - I don't understand why in 688553 ;( | 20:15 |
| *** slaweq_ has joined #openstack-infra | 20:15 | |
| clarkb | mordred: you'll need to pin that gem in the puppet helper lib | 20:15 |
| mordred | clarkb: "neat" | 20:16 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 20:17 |
| AJaeger | corvus: I removed the depends-on and did not include the change in the goodbye_42.3 list | 20:17 |
| *** slaweq has quit IRC | 20:18 | |
| *** slaweq__ has joined #openstack-infra | 20:19 | |
| AJaeger | clarkb, corvus, so https://review.opendev.org/#/q/status:open+topic:goodbye_42.3 looks good to force-merge now, changes are either merged or failed without a depends-on. | 20:19 |
| fungi | mordred: i just did something similar for another lib last week, in infra-spec-helper | 20:19 |
| clarkb | AJaeger: k, I'm finished with lunch now and will start reviewing and merging them | 20:20 |
| fungi | mordred: stuff is increasingly adding deps on newer ruby than centos-7 provides | 20:20 |
| AJaeger | thanks, clarkb | 20:20 |
| fungi | mordred: that said, i wonder whether we even need to run centos-7 jobs for our puppet modules now. we stopped using it to run servers some time ago | 20:21 |
| *** jamesmcarthur has quit IRC | 20:21 | |
| *** slaweq_ has quit IRC | 20:21 | |
| openstackgerrit | Monty Taylor proposed opendev/puppet-openstack_infra_spec_helper master: Pin dry-inflector to <0.2 https://review.opendev.org/688555 | 20:24 |
| mordred | fungi: I think dropping centos-7 testing is a great idea | 20:24 |
| mordred | I am fairly certain we're not going to deploy another centos-7 machine | 20:24 |
| clarkb | ++ | 20:24 |
| *** pcaruana has quit IRC | 20:26 | |
| *** armax has joined #openstack-infra | 20:27 | |
| *** benj has joined #openstack-infra | 20:28 | |
| clarkb | ok I did all the osa changes | 20:28 |
| clarkb | bifrost remains. Did we give a heads up to TheJulia and dtantsur|afk ? | 20:28 |
| *** benj is now known as Guest39910 | 20:28 | |
| clarkb | and https://review.opendev.org/#/c/677181/2 does more than just remove 423 | 20:29 |
| clarkb | I think we should update ^ to only remove 423 | 20:29 |
| clarkb | but those jobs were never voting anyway so we can force merge them without much consultation I don't think | 20:30 |
| *** jamesmcarthur has joined #openstack-infra | 20:30 | |
| AJaeger | yes - and dtantsur|afk approved the one that passed and left comment in https://review.opendev.org/#/c/688449/ | 20:30 |
| mordred | fungi: we don't seem to be running these on centos7 | 20:30 |
| clarkb | AJaeger: oh perfect | 20:30 |
| fungi | mordred: oh, what job was it? | 20:30 |
| AJaeger | clarkb: shall I split 677181 ? | 20:30 |
| *** dpawlik has quit IRC | 20:31 | |
| clarkb | AJaeger: please | 20:31 |
| AJaeger | on it | 20:31 |
| mordred | fungi: puppet-beaker-rspec-puppet-4-infra-system-config and openstackci-beaker-puppet-4 | 20:31 |
| mordred | fungi: on xenial | 20:31 |
| fungi | ahh | 20:32 |
| mordred | so I think the pin is the answer here now | 20:32 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Use bazel 1.0.0 https://review.opendev.org/688523 | 20:32 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Expose volumes for the gerrit hooks, static and tmp dirs https://review.opendev.org/678875 | 20:32 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add jobs to build gerrit master branch https://review.opendev.org/688482 | 20:32 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add checks plugin to gerrit master build https://review.opendev.org/688483 | 20:32 |
| fungi | mordred: got it. in that case https://review.opendev.org/687884 is the one i did last week | 20:33 |
| *** ihti has joined #openstack-infra | 20:33 | |
| fungi | the failures i was seeing were in puppet-beaker-rspec-puppet-4-centos-7-infra and legacy-puppet-syntax-4-centos-7 | 20:34 |
| fungi | maybe we're just missing some cleanup of those jobs in various places | 20:34 |
| mordred | yeah | 20:34 |
| clarkb | heh shade apparently runs one of these 423 jobs | 20:36 |
| *** ihti has quit IRC | 20:37 | |
| AJaeger | yeah, next one... | 20:37 |
| clarkb | https://opendev.org/openstack/shade/src/branch/stable/rocky/.zuul.yaml#L231-L232 | 20:37 |
| clarkb | https://opendev.org/openstack/shade/src/branch/stable/stein/.zuul.yaml#L232-L233 needs to be updated to use 15 instead of 150 | 20:38 |
| clarkb | but I think that is the only 423 case in shade | 20:38 |
| clarkb | mordred: is shade even active anymore? | 20:38 |
| AJaeger | mordred: https://review.opendev.org/688559 is the shade change, please review | 20:38 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Stop running centos-7 puppet jobs https://review.opendev.org/688560 | 20:39 |
| AJaeger | clarkb: I leave 150->150 for another time | 20:39 |
| clarkb | AJaeger: wfm | 20:40 |
| clarkb | mordred: left a comment on the puppet jobs change | 20:40 |
| clarkb | mordred: and for the shade change should I be force merging that one too? | 20:40 |
| mordred | clarkb: no - I think we should be able to just merge it | 20:41 |
| mordred | +A on the shade change | 20:41 |
| clarkb | I'm curious why shade would care about that job | 20:41 |
| clarkb | if bifrost works on suse or centos or ubuntu shouldn't really matter to shade | 20:42 |
| mordred | clarkb: because bifrost uses ansible modules which use shade ... oh - hrm, *shrug*? | 20:42 |
| *** dpawlik has joined #openstack-infra | 20:42 | |
| AJaeger | clarkb: let's do the rest tomorrow if needed and first finish the runs... | 20:42 |
| clarkb | right but that isn't platform specific | 20:43 |
| AJaeger | jrosser: could you +2A https://review.opendev.org/688558 as well, please? | 20:43 |
| AJaeger | clarkb, thanks a lot for the force-merges. | 20:43 |
| mordred | clarkb: why not remove centos-7 from puppet-yum? | 20:44 |
| clarkb | mordred: because its the only platform that puppet-yum makes sense on? | 20:44 |
| mordred | just because it's more general purpose? | 20:44 |
| mordred | clarkb: maybe we should just retire puppet-yum then? | 20:44 |
| clarkb | that may be the correct thing to do | 20:44 |
| mordred | I'll make a patch | 20:44 |
| openstackgerrit | Merged zuul/zuul master: Don't install zuul in the bindep tox testenv https://review.opendev.org/652251 | 20:45 |
| AJaeger | wow, https://review.opendev.org/688441 is passing - clarkb, corvus, one step forward ;) | 20:46 |
| clarkb | mordred: asking about the shade + platform thing because maybe for suse150 we just remove that job entirely | 20:46 |
| *** dpawlik has quit IRC | 20:46 | |
| clarkb | then let the generic tinyipa job catch it (and that job can update its platform ovr time automagically (from shade's perspective) | 20:47 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Retire puppet-yum - remove gate jobs https://review.opendev.org/688563 | 20:47 |
| AJaeger | clarkb: like https://review.opendev.org/688441 ? | 20:48 |
| AJaeger | mordred: ^ | 20:48 |
| AJaeger | if that's the case, we can remove from master as well... | 20:48 |
| AJaeger | or keep bifrost-integration-tinyipa-opensuse-15 there... | 20:48 |
| clarkb | AJaeger: I think that was a mispaste | 20:49 |
| openstackgerrit | Monty Taylor proposed opendev/puppet-yum master: Retire puppet-yum https://review.opendev.org/688565 | 20:49 |
| AJaeger | clarkb: https://review.opendev.org/688558 needs force-merge as well | 20:49 |
| AJaeger | clarkb: https://review.opendev.org/#/c/688564/ is the one I wanted | 20:49 |
| mordred | clarkb: wfm re: shade+platform | 20:50 |
| clarkb | 558 is done | 20:50 |
| *** slaweq has joined #openstack-infra | 20:50 | |
| AJaeger | thanks | 20:50 |
| corvus | AJaeger, clarkb: clarkb seems to have gotten the force-merges -- need anything from me? | 20:50 |
| *** slaweq__ has quit IRC | 20:51 | |
| clarkb | corvus: at this point I think we are just waiting for shade to merge it fix, then we can figure out landing the bifrost fix | 20:52 |
| clarkb | then we can recheck the chagne to remove the nodeset | 20:52 |
| *** pkopec has quit IRC | 20:53 | |
| AJaeger | corvus: clarkb did a great job ;) | 20:53 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Finish retiring puppet-yum https://review.opendev.org/688566 | 20:54 |
| AJaeger | so, https://review.opendev.org/688553 looks fine so far in the queue - if Zuul throws no config error, it can merge as well... | 20:54 |
| AJaeger | (that's the revert-revert of the ansible-test change) | 20:54 |
| mordred | clarkb, AJaeger: I think that ^^ is th sequence to retire puppet-yum | 20:54 |
| clarkb | mordred: as a sanity check yum::mirror and yum::repo don't appear to be used anywhere | 20:55 |
| AJaeger | mordred: will check tomorrow - calling it a day in a few minutes... | 20:55 |
| clarkb | but codesearching for the yum class is a bit more difficult :) | 20:55 |
| openstackgerrit | Monty Taylor proposed openstack/openstack-zuul-jobs master: Stop running legacy puppet4 centos-7 jobs https://review.opendev.org/688567 | 20:56 |
| clarkb | http://codesearch.openstack.org/?q=yum&i=nope&files=.*%5C.pp&repos= looks ok though so I am pretty sure we can clean it up | 20:56 |
| * AJaeger waves good night | 20:56 | |
| *** diablo_rojo has joined #openstack-infra | 20:56 | |
| mordred | clarkb: nope - not used | 20:57 |
| *** rcernin has joined #openstack-infra | 20:57 | |
| AJaeger | clarkb: https://review.opendev.org/#/c/688443 could merge as well | 20:57 |
| clarkb | AJaeger: should we approve that one after we remove the nodeset successfully? | 20:58 |
| mordred | AJaeger, clarkb: https://review.opendev.org/#/q/status:open+topic:infra-centos-7 | 20:59 |
| mordred | forwhen you have a moment | 20:59 |
| mordred | good night AJaeger! | 20:59 |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: Remove legacy-opensuse-423 nodeset https://review.opendev.org/688441 | 21:01 |
| *** mattw4 has quit IRC | 21:02 | |
| *** mattw4 has joined #openstack-infra | 21:02 | |
| * clarkb is very carefully applying votes to changes that shouldn't be force merged | 21:06 | |
| clarkb | maybe I should just remove my perms for now | 21:06 |
| clarkb | mordred: see comment from me (and I assume that is why ci failed too) on https://review.opendev.org/#/c/688566/1 | 21:09 |
| openstackgerrit | Merged opendev/project-config master: Remove third-party-check pipeline https://review.opendev.org/688427 | 21:19 |
| clarkb | corvus: ^ fyi | 21:19 |
| *** goldyfruit has joined #openstack-infra | 21:20 | |
| corvus | cool, i think that means we're ready to run the job once we get that sorted | 21:20 |
| openstackgerrit | James E. Blair proposed opendev/gear master: Add in-repo zuul config https://review.opendev.org/688455 | 21:21 |
| *** weifan has joined #openstack-infra | 21:21 | |
| *** mattw4 has quit IRC | 21:25 | |
| *** mattw4 has joined #openstack-infra | 21:25 | |
| openstackgerrit | Merged openstack/diskimage-builder master: centos7; use numeric DIB_RELEASE https://review.opendev.org/688321 | 21:31 |
| *** xenos76 has quit IRC | 21:36 | |
| *** prometheanfire has quit IRC | 21:36 | |
| *** prometheanfire has joined #openstack-infra | 21:37 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: install-kubernetes: add support for cri-o runtime https://review.opendev.org/688578 | 21:47 |
| *** jistr has quit IRC | 21:52 | |
| *** vesper11- has joined #openstack-infra | 21:53 | |
| *** vesper11 has quit IRC | 21:53 | |
| *** bnemec has quit IRC | 21:55 | |
| openstackgerrit | James E. Blair proposed opendev/gear master: Add in-repo zuul config https://review.opendev.org/688455 | 21:55 |
| *** rcernin has quit IRC | 21:59 | |
| *** jistr has joined #openstack-infra | 22:01 | |
| *** jistr has quit IRC | 22:02 | |
| *** jistr has joined #openstack-infra | 22:03 | |
| *** trident has quit IRC | 22:04 | |
| *** trident has joined #openstack-infra | 22:07 | |
| *** slaweq has quit IRC | 22:14 | |
| *** gagehugo has quit IRC | 22:18 | |
| *** diablo_rojo has quit IRC | 22:19 | |
| corvus | mordred: i was thinking -- since we want to test changes to the checks plugin, maybe we don't want to build a whole new gerrit master image with the checks plugin installed -- maybe we want to take the most recent gerrit master image, whether that's opendev or upstream, and then build only the checks plugin and install it. that way the job is faster | 22:19 |
| corvus | mordred: (it takes like 5 seconds to build the checks plugin standalone) | 22:19 |
| corvus | mordred: also https://review.opendev.org/688455 could use a +3 when you get a sec (cc mnaser -- looks like you can rebase on that and we're gtg) | 22:21 |
| mordred | corvus: that seems reasonable - but don't you need the built source for gerrit to build an out of tree plugin? | 22:21 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: [dnm] testing dnf copr repo for f30 build https://review.opendev.org/688580 | 22:21 |
| corvus | mordred: hrm... i think you need the source -- does it need to be built? | 22:21 |
| mordred | corvus: I do not know the answer to that | 22:22 |
| mordred | corvus: one of the reasons I opted for the in-tree build is because I found out of tree build instructions confuing :) | 22:22 |
| corvus | mordred: i want to say when i tested it locally, i built only the checks plugin and splatted it into the upstream container image | 22:23 |
| mordred | corvus: cool. | 22:23 |
| corvus | mordred: no, i think i'm wrong... i think i built the whole thing | 22:24 |
| corvus | it's confusing... i am scrolling through about 300 bazel commands in 'history' | 22:24 |
| mordred | corvus: I mean - we could *also* tag the builder image we use to build gerrit | 22:25 |
| corvus | lines 509--667 in my history are "attempt to build the checks plugin" | 22:25 |
| mordred | corvus: and use that as the base to build a plugin | 22:25 |
| corvus | mordred: i'm going to go with "unclear" as the answer to whether the full built source is necessary :) | 22:26 |
| mordred | :) | 22:26 |
| mordred | k - well - I'll play with some options | 22:26 |
| corvus | mordred: k -- if it is necessary, i think i like "tag the builder image" as a solution -- because i think the goal of "build the plugin quickly" is a good one | 22:26 |
| mordred | ++ | 22:27 |
| mordred | and tag the builder image isn't that terrible - and doesn't impact our final images | 22:27 |
| corvus | mordred: i'll also see if i can't figure out somethig better than "unclear" to the question above | 22:27 |
| openstackgerrit | Merged opendev/gear master: Add in-repo zuul config https://review.opendev.org/688455 | 22:27 |
| fungi | heads up, pip 19.3 release announcement just hit distutils-sig: https://mail.python.org/archives/list/distutils-sig@python.org/thread/LOAMKIHSQ2UFRSTNKV2YVPTM7VAV7DRH/ | 22:27 |
| fungi | doesn't look especially hairy though | 22:27 |
| *** e0ne has quit IRC | 22:29 | |
| *** weifan has quit IRC | 22:30 | |
| *** e0ne has joined #openstack-infra | 22:31 | |
| fungi | clarkb: i've learned about the NO_TIMEOUT envvar for our reprepro wrapper. didn't notice there was a baked-in 90-minute timeout | 22:31 |
| fungi | but at least we provided a toggle | 22:32 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Finish retiring puppet-yum https://review.opendev.org/688566 | 22:32 |
| *** e0ne has quit IRC | 22:38 | |
| corvus | mordred: i think it might work with only a source checkout. maybe it's worth making a job that inherits from gerrit-master, adds the checks plugin source, then runs "bazel build plugins/checks:checks" to see? | 22:41 |
| *** weifan has joined #openstack-infra | 22:46 | |
| *** diablo_rojo has joined #openstack-infra | 22:46 | |
| *** slaweq has joined #openstack-infra | 22:52 | |
| *** slaweq has quit IRC | 22:56 | |
| *** e0ne has joined #openstack-infra | 22:57 | |
| clarkb | corvus: https://review.opendev.org/#/c/688553/4 should be safe to merge now right? | 23:01 |
| clarkb | it actually runs tests now instead of the zuul complaints | 23:01 |
| *** e0ne has quit IRC | 23:02 | |
| corvus | clarkb: force-merge? yes, i think so. | 23:02 |
| clarkb | ya | 23:02 |
| clarkb | I'll do that now. bifrost is close to getting there too | 23:02 |
| clarkb | then we can rever revert the ndoeset removal | 23:02 |
| clarkb | do you know if that change exists yet? | 23:03 |
| corvus | clarkb: did we merge that one? | 23:03 |
| clarkb | (or maybe we never merged the first removal?) | 23:03 |
| clarkb | ah ok probably never merged the removal in the first place then | 23:03 |
| corvus | yeah, it was the change you're asking about that's the revert-revert of the thing that broke lots of config | 23:03 |
| clarkb | ok that force merge is done. I'll force merge the bifrost change as soon as it reports back to gerrit (so that we have a record of jobs not working on stable/stein there) | 23:04 |
| *** e0ne has joined #openstack-infra | 23:04 | |
| clarkb | then we should be able to land the nodeset removal normall | 23:05 |
| *** weifan has quit IRC | 23:05 | |
| *** rcernin has joined #openstack-infra | 23:07 | |
| *** e0ne has quit IRC | 23:14 | |
| *** tkajinam has joined #openstack-infra | 23:17 | |
| *** jamesmcarthur has quit IRC | 23:17 | |
| *** weifan has joined #openstack-infra | 23:17 | |
| *** tkajinam_ has joined #openstack-infra | 23:17 | |
| *** jamesmcarthur has joined #openstack-infra | 23:19 | |
| *** e0ne has joined #openstack-infra | 23:19 | |
| *** tkajinam has quit IRC | 23:21 | |
| *** e0ne has quit IRC | 23:24 | |
| *** e0ne has joined #openstack-infra | 23:25 | |
| clarkb | bifrost should be done now and I've rechecked the nodeset removal change | 23:26 |
| clarkb | https://review.opendev.org/688582 needs to be merged (another OSA project/branch with 423 jobs) | 23:30 |
| clarkb | corvus: fungi ^ maybe one of you want to force merge that if it fails testing since I pushed it up | 23:30 |
| *** jamesmcarthur has quit IRC | 23:30 | |
| *** hwoarang has quit IRC | 23:30 | |
| *** e0ne has quit IRC | 23:31 | |
| *** dchen has joined #openstack-infra | 23:34 | |
| *** e0ne has joined #openstack-infra | 23:35 | |
| *** jamesmcarthur has joined #openstack-infra | 23:35 | |
| *** dychen has joined #openstack-infra | 23:38 | |
| *** e0ne has quit IRC | 23:39 | |
| *** hwoarang has joined #openstack-infra | 23:40 | |
| clarkb | oh I needed to restart nodepool launchers to pick up that DELETED status handling /me makes note to do that tomorrow | 23:42 |
| *** e0ne has joined #openstack-infra | 23:46 | |
| *** mriedem has quit IRC | 23:48 | |
| *** jamesmcarthur has quit IRC | 23:48 | |
| *** mattw4 has quit IRC | 23:50 | |
| *** e0ne has quit IRC | 23:53 | |
| *** e0ne has joined #openstack-infra | 23:54 | |
| *** jamesmcarthur has joined #openstack-infra | 23:55 | |
| clarkb | mnaser: jrosser https://review.opendev.org/#/c/688582/ passes testing. If we could get that reviewed and approved through normal channels that would be great | 23:58 |
| *** e0ne has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!