| *** tosky has quit IRC | 00:05 | |
| *** takamatsu has joined #openstack-infra | 00:16 | |
| *** takamatsu has quit IRC | 00:31 | |
| *** ricolin_ has joined #openstack-infra | 01:01 | |
| *** zhurong has quit IRC | 01:17 | |
| *** slaweq has joined #openstack-infra | 01:30 | |
| *** Lucas_Gray has quit IRC | 01:38 | |
| *** ramishra has joined #openstack-infra | 01:42 | |
| *** ricolin_ has quit IRC | 02:16 | |
| *** zhurong has joined #openstack-infra | 02:17 | |
| *** tkajinam has quit IRC | 02:34 | |
| *** ociuhandu has joined #openstack-infra | 02:37 | |
| *** dSrinivas has joined #openstack-infra | 02:40 | |
| *** ociuhandu has quit IRC | 02:41 | |
| dSrinivas | HI All, Devstack is not installing in Xenial and failing with this error http://paste.openstack.org/show/788064/ in Jenkins CI. Can anyone suggest me alternate way to install libvirt-clients package in Xenial | 02:43 |
|---|---|---|
| ianw_pto | dSrinivas: devstack master no longer supports Xenial as it's python is too old | 02:44 |
| ianw_pto | that package i think (from https://packages.ubuntu.com/search?keywords=libvirt-clients) is bionic only ... you must be using FORCE to get that far? | 02:44 |
| ianw_pto | you'll either need to use an older branch or use bionic | 02:45 |
| dSrinivas | ianw_pto Thank you. Our CI nodepool image is build on Xenial. So first we would like to make the CI up and running and then shift to bionic image. | 02:47 |
| dSrinivas | I have used FORCE in local.conf and started the process but it is failing to install libvirt-clients | 02:47 |
| dSrinivas | How Can i use older branch | 02:48 |
| ianw_pto | dSrinivas: you'll need to checkout on the stable/train branch | 02:49 |
| dSrinivas | ianw_pto, Here we have to make sure that our cinder driver code is not breaking when ever the patch is created for the cinder driver master repo. if i checkout to stable branch will it work as expected | 02:51 |
| dSrinivas | because we are installing devstack on stable branch instead of master branch | 02:51 |
| ianw_pto | dSrinivas: i do not think this will do what you want, because it will be installing cinder from branch | 02:53 |
| dSrinivas | ianw_pto, Any third party ppa to install libvirt-clients package in xenial | 02:55 |
| ianw_pto | no, but in general what you want just won't work because the master branch of devstack will do all sorts of other things that will fail on xenial | 02:56 |
| dSrinivas | ianw_pto, If a patch is created in cinder master branch and if i am installing the devstack with the stable/train. Is this correct scenario test or we have to use same master repo for both. | 03:01 |
| ianw_pto | dSrinivas: you really want to test everything on master, i can't see that mixing things will be helpful at all | 03:08 |
| dSrinivas | ianw_pto. Thank you | 03:09 |
| dSrinivas | for the information | 03:09 |
| dSrinivas | ianw_pto I have setup the jenkins CI 3 year old back and now i want to update the nodepool image to bionic. I know if i edit nodepool.yaml disk image release section to bionic will it work or nay other changes i have todo | 03:12 |
| ianw_pto | dSrinivas: it is quite likely it will "just work" ... it's the default path now | 03:16 |
| dSrinivas | @ianw | 03:18 |
| dSrinivas | ianw_pto I have old setup and packages are old, Will it work if i change the release to bionic or i have to setup new jenkins again | 03:19 |
| ianw_pto | i really could not say, sorry. if nothing has changed in 3 years, you may have issues with the version of diskimage-builder, etc. | 03:20 |
| dSrinivas | ianw_pto Ok thank you very much for the information | 03:23 |
| *** tkajinam has joined #openstack-infra | 03:41 | |
| *** psachin has joined #openstack-infra | 03:41 | |
| *** udesale has joined #openstack-infra | 03:55 | |
| *** ykarel|away has joined #openstack-infra | 04:37 | |
| *** chandankumar has joined #openstack-infra | 04:50 | |
| *** ykarel|away is now known as ykarel | 05:09 | |
| *** ociuhandu has joined #openstack-infra | 05:30 | |
| *** chandankumar has quit IRC | 05:32 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #openstack-infra | 05:33 | |
| *** dave-mccowan has quit IRC | 05:34 | |
| *** ociuhandu has quit IRC | 05:35 | |
| *** tkajinam has quit IRC | 05:52 | |
| openstackgerrit | Rico Lin proposed opendev/irc-meetings master: Combine auto-scaling and self-healing SIG meeting https://review.opendev.org/701137 | 05:53 |
| *** chandankumar has joined #openstack-infra | 05:53 | |
| *** tkajinam has joined #openstack-infra | 05:56 | |
| *** surpatil has joined #openstack-infra | 06:05 | |
| *** rcernin has quit IRC | 06:12 | |
| *** chandankumar has quit IRC | 06:14 | |
| *** SurajPatil has joined #openstack-infra | 06:16 | |
| *** chandankumar has joined #openstack-infra | 06:17 | |
| *** surpatil has quit IRC | 06:18 | |
| *** surpatil has joined #openstack-infra | 06:19 | |
| *** slaweq has quit IRC | 06:20 | |
| *** SurajPatil has quit IRC | 06:21 | |
| openstackgerrit | Rico Lin proposed opendev/irc-meetings master: Add initial Multi-Arch SIG meeting schedule https://review.opendev.org/701147 | 06:29 |
| openstackgerrit | Rico Lin proposed opendev/irc-meetings master: Add initial Multi-Arch SIG meeting schedule https://review.opendev.org/701147 | 06:35 |
| *** chandankumar has quit IRC | 06:42 | |
| *** tkajinam has quit IRC | 06:52 | |
| openstackgerrit | Rico Lin proposed opendev/irc-meetings master: Combine auto-scaling and self-healing SIG meeting https://review.opendev.org/701137 | 07:04 |
| *** chandankumar has joined #openstack-infra | 07:05 | |
| *** tkajinam has joined #openstack-infra | 07:06 | |
| *** pgaxatte has joined #openstack-infra | 07:29 | |
| *** jtomasek has joined #openstack-infra | 07:37 | |
| *** ykarel is now known as ykarel|lunch | 07:40 | |
| *** ociuhandu has joined #openstack-infra | 07:43 | |
| *** ociuhandu has quit IRC | 07:47 | |
| *** dchen has quit IRC | 07:51 | |
| *** ktsuyuzaki is now known as kota_ | 08:10 | |
| *** tesseract has joined #openstack-infra | 08:11 | |
| *** dciabrin has joined #openstack-infra | 08:11 | |
| *** iurygregory has joined #openstack-infra | 08:16 | |
| *** tkajinam has quit IRC | 08:19 | |
| *** ociuhandu has joined #openstack-infra | 08:22 | |
| *** larainema has joined #openstack-infra | 08:25 | |
| *** iurygregory has quit IRC | 08:25 | |
| *** ociuhandu has quit IRC | 08:26 | |
| *** iurygregory has joined #openstack-infra | 08:27 | |
| *** ykarel|lunch is now known as ykarel | 08:31 | |
| *** ociuhandu has joined #openstack-infra | 08:34 | |
| *** ociuhandu has quit IRC | 08:35 | |
| *** ociuhandu has joined #openstack-infra | 08:36 | |
| openstackgerrit | Akihiro Motoki proposed openstack/openstack-zuul-jobs master: translation job: Fix horizon installation https://review.opendev.org/701170 | 08:38 |
| *** rpittau|afk is now known as rpittau | 08:39 | |
| amotoki | hi, the translation job are still run with python 2.7. horizon has dropped python 2.7 support completely (by dropping django 1.11 support) and it turns out the translation job in the master branch does not work. http://zuul.openstack.org/build/f5c3c186539b41f4b8e9d020de202f4f | 08:46 |
| *** lucasagomes has joined #openstack-infra | 08:50 | |
| *** ralonsoh has joined #openstack-infra | 08:52 | |
| *** priteau has joined #openstack-infra | 09:01 | |
| priteau | Hi. Is it possible to abort a Zuul gate job? I've noticed a small issue after approving a change. | 09:02 |
| amotoki | it looks like the most straightforward way is to specify virtualenv_python: python3 in ensure-babel role in zuul-jobs. | 09:03 |
| *** electrofelix has joined #openstack-infra | 09:03 | |
| amotoki | it affects all users who use zuul-jobs, so I am not sure this is a good way or not.... | 09:03 |
| amotoki | any advise would be appreaciated. | 09:04 |
| amotoki | priteau: AFAIK a workaround is to put a new patch set. | 09:04 |
| priteau | amotoki: Thanks, looks like it worked! | 09:06 |
| amotoki | priteau: :) | 09:06 |
| amotoki | AJaeger: any advise on my above question on translation jobs with py2? | 09:11 |
| *** ociuhandu has quit IRC | 09:17 | |
| *** ociuhandu has joined #openstack-infra | 09:24 | |
| *** derekh has joined #openstack-infra | 09:25 | |
| *** slaweq has joined #openstack-infra | 09:30 | |
| openstackgerrit | Akihiro Motoki proposed openstack/project-config master: Use python3 for translation jobs https://review.opendev.org/701181 | 09:36 |
| AJaeger | amotoki: oh fun ;( Do we need to duplicate the jobs so that older stable branches run python2 while newer python3? | 09:41 |
| amotoki | AJaeger: which stable branches are covered by the translation jobs? If stein and train, py3 would be enough. | 09:43 |
| openstackgerrit | Daniel Bengtsson proposed openstack/reviewstats master: Stop configuring install_command in tox. https://review.opendev.org/694506 | 09:44 |
| AJaeger | amotoki: stein and train only - so, I'm fine with trying it ;) | 09:44 |
| amotoki | AJaeger: sounds fine. "use python3 by default" was the community goal for Stein, so it should not affect projects who completed the goal. | 09:47 |
| AJaeger | config-core, please review https://review.opendev.org/#/c/701181 | 09:47 |
| AJaeger | amotoki: let's see whether it fails anywhere - and if, we can decide what to do there... | 09:47 |
| amotoki | AJaeger: 701181 is a fix for master branch. https://review.opendev.org/#/c/701170/ is a fix for stable branches. | 09:48 |
| *** SurajPatil has joined #openstack-infra | 09:49 | |
| *** ociuhandu has quit IRC | 09:49 | |
| AJaeger | thanks, amotoki - reviewing now | 09:49 |
| AJaeger | amotoki: does "pip install ." install test-requirements? | 09:50 |
| *** surpatil has quit IRC | 09:52 | |
| *** ociuhandu has joined #openstack-infra | 09:52 | |
| amotoki | AJaeger: "pip install ." does not install test-requirements. As I mentioned in the commit message, it is a good chance to include test-requirments. | 09:52 |
| amotoki | AJaeger: I would like to move django-babel to test-requirements soon :) | 09:52 |
| AJaeger | amotoki: I should have read that one ;) | 09:53 |
| *** dciabrin__ has quit IRC | 09:53 | |
| AJaeger | amotoki: Ok, I'm happy - +2s | 09:53 |
| *** pcaruana has joined #openstack-infra | 09:53 | |
| amotoki | thanks! | 09:54 |
| AJaeger | thanks for the fixes! | 09:54 |
| AJaeger | config-core, please review https://review.opendev.org/#/c/701170/ as well! 701181 and 701170 are needed to fix the translation jobs. | 09:54 |
| *** ociuhandu has quit IRC | 09:56 | |
| *** ociuhandu has joined #openstack-infra | 09:59 | |
| *** surpatil has joined #openstack-infra | 10:01 | |
| *** SurajPatil has quit IRC | 10:04 | |
| *** pcaruana has quit IRC | 10:11 | |
| *** pcaruana has joined #openstack-infra | 10:12 | |
| *** SurajPatil has joined #openstack-infra | 10:26 | |
| *** surpatil has quit IRC | 10:28 | |
| *** ociuhandu has quit IRC | 10:29 | |
| *** slaweq has quit IRC | 10:31 | |
| *** aedc has joined #openstack-infra | 10:32 | |
| *** ociuhandu has joined #openstack-infra | 10:33 | |
| *** ramishra has quit IRC | 10:33 | |
| *** priteau has quit IRC | 10:33 | |
| *** ociuhandu has quit IRC | 10:34 | |
| *** ociuhandu has joined #openstack-infra | 10:34 | |
| *** ramishra has joined #openstack-infra | 10:37 | |
| *** surpatil has joined #openstack-infra | 10:53 | |
| *** SurajPatil has quit IRC | 10:56 | |
| *** surpatil is now known as surpatil|afk | 11:11 | |
| *** udesale has quit IRC | 11:16 | |
| *** ociuhandu has quit IRC | 11:25 | |
| *** ociuhandu has joined #openstack-infra | 11:27 | |
| *** ociuhandu has quit IRC | 11:28 | |
| *** ociuhandu has joined #openstack-infra | 11:29 | |
| openstackgerrit | Akihiro Motoki proposed zuul/zuul-jobs master: Allow to use ensure-babel without ensure-sphinx https://review.opendev.org/701209 | 11:35 |
| *** surpatil|afk has quit IRC | 11:36 | |
| *** Lucas_Gray has joined #openstack-infra | 11:38 | |
| *** slaweq has joined #openstack-infra | 11:41 | |
| *** ociuhandu has quit IRC | 11:42 | |
| *** ociuhandu has joined #openstack-infra | 11:48 | |
| *** slaweq has quit IRC | 11:57 | |
| *** surpatil|afk has joined #openstack-infra | 11:58 | |
| *** ociuhandu has quit IRC | 11:58 | |
| *** ociuhandu has joined #openstack-infra | 11:59 | |
| *** rfolco has joined #openstack-infra | 12:10 | |
| *** verdurin has quit IRC | 12:15 | |
| *** verdurin has joined #openstack-infra | 12:17 | |
| *** beagles is now known as beagles|afk | 12:27 | |
| *** surpatil|afk is now known as surpatil | 12:31 | |
| *** xek_ has joined #openstack-infra | 12:56 | |
| *** rlandy has joined #openstack-infra | 12:58 | |
| *** SurajPatil has joined #openstack-infra | 13:01 | |
| *** surpatil has quit IRC | 13:04 | |
| *** rh-jelabarre has joined #openstack-infra | 13:09 | |
| *** ociuhandu has quit IRC | 13:12 | |
| *** dSrinivas has quit IRC | 13:21 | |
| *** surpatil has joined #openstack-infra | 13:29 | |
| *** SurajPatil has quit IRC | 13:32 | |
| *** beagles|afk is now known as beagles | 13:43 | |
| *** ahosam has joined #openstack-infra | 13:46 | |
| *** ykarel has quit IRC | 13:47 | |
| *** hrw has joined #openstack-infra | 13:48 | |
| hrw | morning | 13:48 |
| hrw | reprepro config for ubuntu-cloud-archive needs changes - linaro-london is arm64 not amd64 and therefore kolla-ubuntu-source-aarch64 job in https://review.opendev.org/701121 fails | 13:50 |
| *** Lucas_Gray has quit IRC | 13:53 | |
| *** goldyfruit_ has quit IRC | 13:54 | |
| *** Lucas_Gray has joined #openstack-infra | 13:55 | |
| hrw | I do not know syntax used for generation of reprepro config so the only patch I could write is adding arm64 to list of architectures. but that would enlarge mirror space on all boxes without any gain for !linaro-london | 13:56 |
| hrw | but as it is shared for all machines... | 13:58 |
| openstackgerrit | Marcin Juszkiewicz proposed opendev/system-config master: reprepro: sync arm64 packages from Ubuntu cloud archive https://review.opendev.org/701227 | 14:00 |
| *** tosky has joined #openstack-infra | 14:01 | |
| *** petevg has joined #openstack-infra | 14:02 | |
| *** Goneri has quit IRC | 14:03 | |
| *** nhicher has joined #openstack-infra | 14:05 | |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: doc: add note for speculative queue attribute https://review.opendev.org/701230 | 14:06 |
| *** eharney has joined #openstack-infra | 14:08 | |
| *** ykarel has joined #openstack-infra | 14:08 | |
| *** Lucas_Gray has quit IRC | 14:10 | |
| *** jtomasek has quit IRC | 14:13 | |
| *** goldyfruit_ has joined #openstack-infra | 14:13 | |
| *** jtomasek has joined #openstack-infra | 14:16 | |
| *** ahosam has quit IRC | 14:20 | |
| *** bnemec has joined #openstack-infra | 14:28 | |
| *** surpatil has quit IRC | 14:34 | |
| openstackgerrit | Merged zuul/nodepool master: Fix resource warnings when running tests https://review.opendev.org/700275 | 14:41 |
| *** ykarel is now known as ykarel|away | 14:46 | |
| openstackgerrit | Merged zuul/nodepool master: Parallelize test_ec2_machine https://review.opendev.org/700276 | 14:47 |
| *** aedc has quit IRC | 14:51 | |
| *** aedc has joined #openstack-infra | 14:51 | |
| *** ociuhandu has joined #openstack-infra | 14:56 | |
| mordred | AJaeger, amotoki: also - we should probably update zuul-jobs things like ensure-babel to default to python3 instead, since python2 is EOL - but we should probably go through an annoucement period to do that | 14:57 |
| *** goldyfruit_ has quit IRC | 14:59 | |
| openstackgerrit | Merged zuul/nodepool master: Support userdata for instances in aws https://review.opendev.org/632712 | 14:59 |
| *** goldyfruit_ has joined #openstack-infra | 14:59 | |
| mordred | AJaeger: also +A on those two | 14:59 |
| *** diablo_rojo has joined #openstack-infra | 15:00 | |
| AJaeger | thanks, mordred | 15:01 |
| AJaeger | Let me propose a change for python2->3 | 15:01 |
| mordred | fungi, clarkb, Shrews: whence you awaken, https://review.opendev.org/#/c/690504 is a little long in the tooth and could use a +3 | 15:02 |
| *** diablo_rojo has quit IRC | 15:02 | |
| AJaeger | mordred: I leave the python2 default change for another time | 15:04 |
| mordred | AJaeger: kk | 15:04 |
| AJaeger | but if anybody else wants to tackle it... | 15:05 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Bump gitea version to 1.10.2 https://review.opendev.org/699490 | 15:05 |
| mordred | config-core, infra-root: TIL: "git config --global diff.noprefix true" which turns off the a/ and b/ prefixes in the diff headers with git. thought I'd pass along the hint | 15:06 |
| *** xek_ has quit IRC | 15:06 | |
| mordred | (makes it much easier to copy/paste a path from a git show or git diff output) | 15:07 |
| AJaeger | nice, thanks | 15:07 |
| *** ociuhandu has quit IRC | 15:08 | |
| openstackgerrit | Merged openstack/project-config master: Use python3 for translation jobs https://review.opendev.org/701181 | 15:08 |
| *** aedc_ has joined #openstack-infra | 15:08 | |
| *** ociuhandu has joined #openstack-infra | 15:09 | |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: translation job: Fix horizon installation https://review.opendev.org/701170 | 15:09 |
| *** priteau has joined #openstack-infra | 15:10 | |
| *** Goneri has joined #openstack-infra | 15:10 | |
| AJaeger | frickler: we cannot switch translation jobs to bionic since the translations jobs need Java8 which is not available on Bionic ;( | 15:11 |
| mordred | AJaeger: "yay" | 15:12 |
| *** aedc has quit IRC | 15:12 | |
| Shrews | mordred: taking a look, but first remind me... what is this "gerrit" thing again???? | 15:13 |
| mordred | AJaeger: maybe we should make a docker image with the java8 CLI translations tools and update the translation jobs to use that image for the translations commands | 15:13 |
| AJaeger | mordred: nice idea | 15:14 |
| mordred | Shrews: right? I'm pretty sure I've used a computer before, but I'm not 100% sure that true | 15:14 |
| Shrews | also, it seems reeeally slow for me rn | 15:15 |
| *** dave-mccowan has joined #openstack-infra | 15:16 | |
| openstackgerrit | Merged zuul/nodepool master: Make public ip configurable in aws https://review.opendev.org/632715 | 15:16 |
| *** goldyfruit_ has quit IRC | 15:16 | |
| *** ociuhandu has quit IRC | 15:20 | |
| *** chandankumar has quit IRC | 15:27 | |
| *** iurygregory has quit IRC | 15:35 | |
| *** iurygregory has joined #openstack-infra | 15:36 | |
| openstackgerrit | David Shrewsbury proposed zuul/zuul-jobs master: ensure-tox: Output tox version https://review.opendev.org/701236 | 15:36 |
| *** psachin has quit IRC | 15:37 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: DNM: testing https://review.opendev.org/701237 | 15:37 |
| AJaeger | amotoki: any ideas why this fails now: http://zuul.opendev.org/t/openstack/build/b250fe8ba5754fbfbeaf5d6a3df196f7/log/job-output.txt#679-687 | 15:38 |
| AJaeger | amotoki: it works fine with python36 locally for me ;( | 15:38 |
| openstackgerrit | Merged opendev/system-config master: Base gitea-init on opendevorg/python-base https://review.opendev.org/690504 | 15:41 |
| *** stewie925 has joined #openstack-infra | 15:41 | |
| AJaeger | amotoki: ah, binary file - change to json.loads in 3.6 | 15:41 |
| stewie925 | hello gm - would like to request to look into the https://review.opendev.org/#/c/685395/ Zuul issue? thank you | 15:42 |
| *** lucasagomes has quit IRC | 15:43 | |
| AJaeger | stewie925: why is that a Zuul issue? Looking at https://zuul.opendev.org/t/openstack/build/49e5c4b6ad10497d8d6f53dda13ff475/log/job-output.txt#3851-3855 I see an error | 15:44 |
| stewie925 | AJaeger: oh thanks | 15:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Update Zanata script for Python 3.5 https://review.opendev.org/701240 | 15:49 |
| *** iokiwi0 has joined #openstack-infra | 15:50 | |
| AJaeger | config-core, amotoki, here's the fix for the failure above, reviews welcome to fix translation jobs, please ^ | 15:50 |
| *** iokiwi has quit IRC | 15:51 | |
| *** iokiwi0 is now known as iokiwi | 15:51 | |
| AJaeger | config-core, please review https://review.opendev.org/700913 and https://review.opendev.org/700909 and https://review.opendev.org/700452 | 15:55 |
| *** lucasagomes has joined #openstack-infra | 15:56 | |
| *** electrofelix has quit IRC | 15:56 | |
| *** chandankumar has joined #openstack-infra | 16:02 | |
| *** ociuhandu has joined #openstack-infra | 16:05 | |
| *** dayou has quit IRC | 16:07 | |
| clarkb | corvus: zuul-ci.org's ssl cert expires in 2 weeks. Did you want to try the delegate from zuul-ci.org into acme.opendev.org plan for static.openstack.org migrations with zuul-ci.org too in order to address that? | 16:08 |
| *** bnemec has quit IRC | 16:10 | |
| *** arxcruz is now known as arxcruz|off | 16:11 | |
| *** iurygregory has quit IRC | 16:11 | |
| *** ociuhandu has quit IRC | 16:11 | |
| AJaeger | is log streaming broken? I get on http://zuul.opendev.org/t/openstack/stream/f9b921f9e4ae4d8fb5e7cc19e7dd6c13?logfile=console.log just "End of stream" - but works on a few others | 16:11 |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: Update Zanata script for Python 3.5 https://review.opendev.org/701240 | 16:13 |
| *** ociuhandu has joined #openstack-infra | 16:14 | |
| *** gyee has joined #openstack-infra | 16:15 | |
| *** larainema has quit IRC | 16:16 | |
| *** ociuhandu has quit IRC | 16:19 | |
| *** ociuhandu has joined #openstack-infra | 16:20 | |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Add ozj to docs site https://review.opendev.org/701244 | 16:20 |
| corvus | clarkb: right now it's a manual cert, yeah? | 16:23 |
| clarkb | corvus: yes, I think you provisioned the first cert 2 years ago. not sure if someone else did the other last year. But it is manual | 16:23 |
| corvus | clarkb: have we done the first acme -> afs site yet? | 16:24 |
| clarkb | corvus: not yet | 16:24 |
| *** pgaxatte has quit IRC | 16:25 | |
| corvus | zuul-ci.org is currently an A record for files02.o.o | 16:25 |
| clarkb | yup it would be very similar in set up to the static.o.o offload plan | 16:25 |
| *** ociuhandu has quit IRC | 16:25 | |
| *** psachin has joined #openstack-infra | 16:27 | |
| corvus | clarkb: was there a writeup of that somewhere? | 16:27 |
| clarkb | corvus: yes there is a spec. Let me find a link | 16:28 |
| corvus | https://docs.opendev.org/opendev/infra-specs/latest/specs/retire-static.html ? | 16:29 |
| openstackgerrit | David Shrewsbury proposed zuul/zuul-jobs master: ensure-tox: Output tox version https://review.opendev.org/701236 | 16:29 |
| openstackgerrit | Rico Lin proposed opendev/irc-meetings master: Combine auto-scaling and self-healing SIG meeting https://review.opendev.org/701137 | 16:29 |
| clarkb | corvus: yup | 16:30 |
| corvus | clarkb: there doesn't seem to be a static01.opendev.org host | 16:30 |
| clarkb | corvus: the chagnes to provision that host are in review | 16:30 |
| fungi | clarkb: i was the one who provisioned the replacement last year (we may have missed catching the expiration at the time due to lack of sni support in sslcheck tool) | 16:30 |
| clarkb | corvus: I don't think we need to switch zuul-ci.org to that to make this happen though | 16:30 |
| fungi | s/provisioned/purchased/ | 16:30 |
| clarkb | corvus: for zuul-ci.org we'd do the dns, LE, and apache config update | 16:31 |
| corvus | clarkb: you mean manually put the output of LE into hostvars? | 16:31 |
| clarkb | corvus: no, update the apache vhost to read the output of LE and add a handler to reload apache when LE updates | 16:32 |
| corvus | or do you mean run LE on files01.openstack.org and use that instead of the hostvars? | 16:32 |
| clarkb | yup that | 16:32 |
| corvus | (or rather files02) | 16:32 |
| corvus | k, i'll take a quick look at that | 16:32 |
| *** ociuhandu has joined #openstack-infra | 16:32 | |
| fungi | clarkb: also, i expect you saw, but zuul.opendev.org will need a new cert too | 16:33 |
| clarkb | fungi: yup | 16:33 |
| corvus | can someone else look into that one? :) | 16:33 |
| clarkb | fungi: that one I should be able to switch over to LE pretty easily since it is already in the opdnev.org domain so the acme delegation is "normal" | 16:33 |
| *** mattw4 has joined #openstack-infra | 16:34 | |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-zuul-jobs master: Fix get-modulename.py for python3 https://review.opendev.org/701248 | 16:36 |
| AJaeger | config-core, another change needed for translations under python3, please ^ | 16:36 |
| openstackgerrit | Merged zuul/zuul-jobs master: Add pypi_fqdn to differentiate it package mirrors https://review.opendev.org/696337 | 16:36 |
| fungi | clarkb: do we already have any puppet-configured services using le, or are they all configured by ansible up to this point? | 16:37 |
| *** ociuhandu has quit IRC | 16:37 | |
| clarkb | fungi: we do. Basically we update the puppet managed vhost template to point at the LE results, then ansible also triggers the service reload | 16:37 |
| clarkb | all of the config remains in puppet but ansible can update the cert data under it and trigger service reloads if necessary | 16:38 |
| fungi | okay, just wasn't sure which service(s) we'd done that way so far | 16:38 |
| fungi | and yeah, that solution sounds safe enough | 16:38 |
| openstackgerrit | James E. Blair proposed opendev/zone-zuul-ci.org master: Update gitreview https://review.opendev.org/701249 | 16:46 |
| openstackgerrit | James E. Blair proposed opendev/zone-zuul-ci.org master: Delegate acme challenges to opendev https://review.opendev.org/701250 | 16:46 |
| corvus | clarkb, fungi: ^ step 1 | 16:46 |
| corvus | (well, steps 1a and 1b :) | 16:46 |
| *** ahosam has joined #openstack-infra | 16:46 | |
| *** tosky has quit IRC | 16:46 | |
| clarkb | note that repo has a zone-zuul-ci-core group with no one in it | 16:47 |
| clarkb | corvus: should I seed that with another group or? | 16:47 |
| corvus | clarkb: i guess infra-core for now? | 16:48 |
| clarkb | done | 16:48 |
| *** lucasagomes has quit IRC | 16:49 | |
| clarkb | (also if it wasn't clear I'll work on zuul.opendev.org cert renewal. This one is slightly less urgent as it expires in 23 days instead of 14) | 16:49 |
| corvus | thx | 16:49 |
| openstackgerrit | Merged opendev/irc-meetings master: Add initial Multi-Arch SIG meeting schedule https://review.opendev.org/701147 | 16:50 |
| clarkb | it is officially on the "back from holidays" list after a few other things like responding to roman_q and helping AJaeger withe fuel cleanups and beginning of the year paperwork | 16:50 |
| clarkb | but first breakfast | 16:50 |
| openstackgerrit | Ghanshyam Mann proposed opendev/irc-meetings master: Update QA office hour time https://review.opendev.org/701251 | 16:50 |
| fungi | i guess le issues certs for www.@ as well as @ without needing a separate _acme-challenge.www.@ record? | 16:51 |
| corvus | fungi: that is unclear to me, but i'm planning 3 changes in series and we'll get the certs on disk in change 2, but won't use them until change 3, so we can verify | 16:52 |
| fungi | wfm | 16:52 |
| fungi | also i'm assuming the mixed spaces and tabs in the zonefile indenting will be fine, if memory serves that's valid | 16:53 |
| clarkb | in the ansible le config you list out the names in each cert. you should be able to add www to the list | 16:53 |
| fungi | k | 16:53 |
| clarkb | however I think you have to delegate www separately | 16:54 |
| corvus | yeah, i don't see a www.tarballs.opendev.org SAN entry in the cert for tarballs.opendev.org | 16:54 |
| corvus | so it didn't give it to us automatically | 16:54 |
| openstackgerrit | Merged zuul/zuul master: doc: add note for speculative queue attribute https://review.opendev.org/701230 | 16:56 |
| openstackgerrit | Paul Belanger proposed zuul/zuul master: Use -N '' for ssh-keygen https://review.opendev.org/701252 | 16:56 |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Get letsencrypt certs for zuul-ci.org https://review.opendev.org/701253 | 16:57 |
| corvus | clarkb, fungi: ^ i think that's step 2 | 16:57 |
| *** rpittau is now known as rpittau|afk | 16:58 | |
| openstackgerrit | James E. Blair proposed opendev/zone-zuul-ci.org master: Use tabs instead of spaces https://review.opendev.org/701254 | 16:59 |
| corvus | fungi: ^ | 16:59 |
| openstackgerrit | Merged opendev/zone-zuul-ci.org master: Update gitreview https://review.opendev.org/701249 | 17:01 |
| openstackgerrit | Merged opendev/zone-zuul-ci.org master: Delegate acme challenges to opendev https://review.opendev.org/701250 | 17:01 |
| *** chkumar246 has joined #openstack-infra | 17:01 | |
| *** chandankumar has quit IRC | 17:04 | |
| *** ociuhandu has joined #openstack-infra | 17:06 | |
| *** chkumar246 has quit IRC | 17:07 | |
| clarkb | corvus: I think you also need to add files to the letsencrypt group in inventory/groups.yaml | 17:10 |
| clarkb | corvus: and looking at it again I'm pretty sure you need to do the _acme_challange.www delegation explicity I think | 17:10 |
| clarkb | otherwise this looks good to provision the certs but not consume them then we can do the apache update in a followup when happy with the resulting certs | 17:10 |
| *** ykarel|away has quit IRC | 17:11 | |
| *** ociuhandu has quit IRC | 17:11 | |
| *** nicolasbock has joined #openstack-infra | 17:11 | |
| openstackgerrit | Merged opendev/zone-zuul-ci.org master: Use tabs instead of spaces https://review.opendev.org/701254 | 17:14 |
| *** ociuhandu has joined #openstack-infra | 17:15 | |
| *** ociuhandu has quit IRC | 17:16 | |
| *** priteau has quit IRC | 17:16 | |
| AJaeger | config-core, please review https://review.opendev.org/701248 as next fix for python3 in translation jobs | 17:16 |
| corvus | clarkb: files02.opendev.org should already be in groups.ymal | 17:16 |
| *** ociuhandu has joined #openstack-infra | 17:16 | |
| *** ociuhandu has quit IRC | 17:16 | |
| clarkb | corvus: oh! sorry | 17:16 |
| corvus | clarkb: (note also that i'm adding new certs to an existing LE var for files | 17:17 |
| corvus | clarkb: (it's already getting tarballs.opendev.org) | 17:17 |
| clarkb | yup changed to a +2 | 17:17 |
| corvus | clarkb: should we go ahead and add the www then? or let this run to confirm? | 17:18 |
| clarkb | I think we should add www because the provision will fail without the delegation (as we are asking for www) | 17:19 |
| *** dayou has joined #openstack-infra | 17:20 | |
| openstackgerrit | James E. Blair proposed opendev/zone-zuul-ci.org master: Also delegate www. acme record to opendev https://review.opendev.org/701257 | 17:20 |
| corvus | clarkb, fungi: ^ | 17:20 |
| clarkb | lgtm +2 | 17:21 |
| openstackgerrit | Merged zuul/zuul-jobs master: Ensure python interpreter is installed for tox-py3X https://review.opendev.org/700899 | 17:22 |
| *** sshnaidm is now known as sshnaidm|afk | 17:31 | |
| *** ociuhandu has joined #openstack-infra | 17:31 | |
| *** mattw4 has quit IRC | 17:31 | |
| *** mattw4 has joined #openstack-infra | 17:32 | |
| *** evrardjp has quit IRC | 17:33 | |
| *** evrardjp has joined #openstack-infra | 17:33 | |
| *** ociuhandu has quit IRC | 17:36 | |
| openstackgerrit | Merged opendev/zone-zuul-ci.org master: Also delegate www. acme record to opendev https://review.opendev.org/701257 | 17:36 |
| clarkb | I've responded to romang re airship CI things and suggested thursday at 16:00UTC as a good time to call and sync up more synchronously | 17:38 |
| clarkb | I'm not sure if that works for others, and can probably move that around a bit to accomodate others if they are interested and can't make that time | 17:38 |
| *** ahosam has quit IRC | 17:39 | |
| fungi | there's an openstack release team meeting i'll be trying to pay attention in, but i can probably do both | 17:45 |
| clarkb | I suggested pbx.o.o and if that doesn't work: meet.jit.si instead | 17:46 |
| *** derekh has quit IRC | 18:00 | |
| corvus | fungi: clarkb and i are pretty sure we do need https://review.opendev.org/701253 | 18:01 |
| corvus | fungi: oh derp, sorry wrong link | 18:01 |
| corvus | fungi: i meant to highlight https://review.opendev.org/701257 which you already +3d | 18:01 |
| corvus | then i guess 701253 is ready to go (dig -t txt _acme-challenge.www.zuul-ci.org works) | 18:02 |
| *** pcaruana has quit IRC | 18:07 | |
| *** psachin has quit IRC | 18:12 | |
| *** nicolasbock has quit IRC | 18:15 | |
| *** pcaruana has joined #openstack-infra | 18:16 | |
| *** lbragstad_ has joined #openstack-infra | 18:21 | |
| *** pcaruana has quit IRC | 18:21 | |
| *** lbragstad has quit IRC | 18:23 | |
| *** ralonsoh has quit IRC | 18:29 | |
| openstackgerrit | Merged zuul/zuul master: Use -N '' for ssh-keygen https://review.opendev.org/701252 | 18:32 |
| openstackgerrit | Merged opendev/system-config master: Get letsencrypt certs for zuul-ci.org https://review.opendev.org/701253 | 18:33 |
| *** efried has joined #openstack-infra | 18:35 | |
| efried | o/ infra | 18:35 |
| efried | Does someone here have the power to remove -2s from a user who has left OpenStack? | 18:35 |
| *** tesseract has quit IRC | 18:38 | |
| corvus | efried: yes, any infra-root | 18:40 |
| corvus | efried: however | 18:40 |
| corvus | efried: the "best" way to do that might be to remove them from the -core group | 18:40 |
| corvus | efried: that will better reflect reality, and it will downgrade their -2 to a -1 | 18:40 |
| efried | corvus: oh, cool, good to know. However, I don't want to remove mriedem's core powers just yet, as there are a couple of things I'm hoping he comes in and polishes up in the near future. But in the meantime, would you mind taking his -2 off of https://review.opendev.org/#/c/694033/ ? I've added mine (it's a procedural hold pending blueprint approval) to replace it. | 18:41 |
| efried | corvus: ...and https://review.opendev.org/#/c/688932/ ditto? | 18:42 |
| corvus | efried: that seems non-controversial to me; infra-root: any thoughts? should i go ahead and remove those 2 +2s? | 18:45 |
| fungi | fine by me | 18:45 |
| corvus | done; history appears in the change log | 18:46 |
| efried | corvus: thanks! | 18:46 |
| AJaeger | config-core, please review https://review.opendev.org/701248 as next fix for python3 in translation jobs, they're all broken otherwise | 18:48 |
| clarkb | looking | 18:50 |
| corvus | +2 clarkb see parent as well | 18:50 |
| *** eharney has quit IRC | 18:50 | |
| clarkb | corvus: yup I got the parent earlier | 18:50 |
| *** efried has quit IRC | 18:51 | |
| AJaeger | thanks! | 18:51 |
| hrw | morning. Can I get someone to look at https://review.opendev.org/701227 patch? It adds mirroring of arm64 packages for ubuntu cloud archive | 18:51 |
| *** efried has joined #openstack-infra | 18:51 | |
| AJaeger | smcginnis: your reviewstats stack is now passing - see https://review.opendev.org/#/c/699683 . Reviews welcome for some cleanups... | 18:52 |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: Fix get-modulename.py for python3 https://review.opendev.org/701248 | 18:57 |
| efried | corvus: hm, Ken'ichi was removed from core a while back but https://review.opendev.org/#/c/383049/ still carries his -2 | 18:59 |
| efried | In this case that's okay I think -- I'm going to abandon that change for lack of movement -- just wondering why that goes against what you said earlier? Voted in an old version of gerrit or something? | 19:00 |
| efried | oh, I can't abandon that, it's in tempest. | 19:00 |
| *** eharney has joined #openstack-infra | 19:03 | |
| corvus | efried: ken'ichi is still in tempest-core | 19:05 |
| efried | okay, that makes sense | 19:05 |
| fungi | also you'll find old -2 votes which were there at the time the change merged stick around forever, gerrit only filters them out for changes which are still open | 19:07 |
| fungi | (which makes sense, at least in my opinion) | 19:07 |
| efried | agree | 19:08 |
| efried | I noticed one of those ones from Matt earlier so I ran a report and that one from Ken'ichi surfaced. | 19:09 |
| efried | btw, I couldn't figure out a way to do a gerrit search on "-2 by specific user" | 19:09 |
| efried | the best I could do was label:Code-Review=-2 and then use the 'reviewer' key. Which obv gets more results than the ones I was looking for. | 19:10 |
| efried | It was a manageable list, anyway. | 19:10 |
| *** cloudnull has quit IRC | 19:16 | |
| *** cloudnull has joined #openstack-infra | 19:17 | |
| openstackgerrit | Clark Boylan proposed opendev/puppet-zuul master: Add option for LE provisioned certs to zuul::web https://review.opendev.org/701265 | 19:19 |
| clarkb | infra-root ^ I think that is the first step in using LE certs for zuul.opendev.org | 19:19 |
| clarkb | once we are happy with that I can push the dns and system-config changes to provision the cert and take advantage of that puppetry | 19:20 |
| fungi | lookin' | 19:20 |
| clarkb | efried: https://review.opendev.org/Documentation/user-search.html#labels you would do label:Code-Review=-1,mriedem or similar | 19:23 |
| smcginnis | Thanks AJaeger | 19:24 |
| *** adriant has quit IRC | 19:24 | |
| *** iokiwi has quit IRC | 19:24 | |
| *** lifeless has quit IRC | 19:24 | |
| *** lifeless has joined #openstack-infra | 19:25 | |
| *** dayou has quit IRC | 19:27 | |
| *** dayou has joined #openstack-infra | 19:28 | |
| *** iokiwi has joined #openstack-infra | 19:29 | |
| *** adriant has joined #openstack-infra | 19:29 | |
| corvus | clarkb: i'm just gonna trust you on that if ! false nil thing. | 19:31 |
| clarkb | corvus: I found it in another erb via codesearch. I think cmorpheus was the originator of that pattern | 19:32 |
| corvus | then it is certainly correct | 19:32 |
| clarkb | also I think testing covers that? I'll try to confirn this | 19:32 |
| *** cmorpheus is now known as cmurphy | 19:33 | |
| cmurphy | i think it's right | 19:33 |
| corvus | i think that translates as "look up the variable vhosts_flags_int, which should be a dictionary, look up the key in the vhost_name variable, which should produce a dictionary, look up the key 'use_le' in that dictionary, then see if that value is one of false, the empty string, or nil. if that is not the case, include the le cert" | 19:34 |
| cmurphy | i think you could simplify it now that there's no puppet 3 left | 19:35 |
| corvus | or, canceling out the double negatives, "see if the use_le flag is true" | 19:35 |
| corvus | it's just weirdly hard to see if something is true i guess | 19:35 |
| fungi | because puppet | 19:36 |
| clarkb | corvus: ya truthy value ambiguity | 19:36 |
| corvus | as mordred would say, "it's not NOT true" | 19:36 |
| fungi | because some puppet releases treat the empty string as boolean truthy | 19:36 |
| cmurphy | i'm 95% sure on puppet 4 you could just 'if ! scope.lookupvar(...)' | 19:37 |
| cmurphy | or something | 19:37 |
| corvus | clarkb: +2 in case you wanted to hold for test results | 19:37 |
| clarkb | thanks | 19:37 |
| AJaeger | clarkb: regarding retirement of last round of fuel repos, I put up as you suggested an etherpad with what needs to be done, see https://etherpad.openstack.org/p/retire-x-fuel | 19:39 |
| clarkb | thanks | 19:39 |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Re-elect James Blair as project lead https://review.opendev.org/701267 | 19:41 |
| *** rlandy is now known as rlandy|afk | 19:43 | |
| *** slaweq has joined #openstack-infra | 19:44 | |
| *** bnemec has joined #openstack-infra | 19:47 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: DNM: testing https://review.opendev.org/701237 | 20:00 |
| AJaeger | yeah, first translation jobs work now with python3 - amotoki, swift just passed | 20:00 |
| *** ociuhandu has joined #openstack-infra | 20:04 | |
| *** ociuhandu has quit IRC | 20:04 | |
| *** ianychoi has quit IRC | 20:04 | |
| *** ociuhandu has joined #openstack-infra | 20:05 | |
| *** ociuhandu has quit IRC | 20:06 | |
| *** ociuhandu has joined #openstack-infra | 20:07 | |
| *** dave-mccowan has quit IRC | 20:12 | |
| *** igordc has joined #openstack-infra | 20:13 | |
| *** igordc has quit IRC | 20:15 | |
| *** dave-mccowan has joined #openstack-infra | 20:16 | |
| *** ociuhandu has quit IRC | 20:18 | |
| *** ianychoi has joined #openstack-infra | 20:19 | |
| *** Lucas_Gray has joined #openstack-infra | 20:21 | |
| *** rfolco has quit IRC | 20:34 | |
| *** ociuhandu has joined #openstack-infra | 20:41 | |
| *** ociuhandu has quit IRC | 20:44 | |
| fungi | infra-root: looks like https://review.opendev.org/689271 is a simple mailing list addition and has fallen through the cracks for a few months if anyone else has a moment for a second review | 20:49 |
| *** tosky has joined #openstack-infra | 20:55 | |
| efried | corvus: aha! Didn't see that, thanks. Confirms what I pared down manually earlier, just these three stable reviews remaining. https://review.opendev.org/#/q/status:open+label:Code-Review%253D-2%252Cuser%253Dmriedem | 20:55 |
| efried | (which ought to wait for another stable core, as noted) | 20:55 |
| openstackgerrit | Clark Boylan proposed opendev/puppet-zuul master: Add option for LE provisioned certs to zuul::web https://review.opendev.org/701265 | 20:56 |
| openstackgerrit | Clark Boylan proposed opendev/puppet-zuul master: Fix linter errors https://review.opendev.org/701278 | 20:56 |
| clarkb | corvus: the tests on that puppet-zuul changing are failing due to a linter issue fixed with ^ but also beacuse of that zuul web js tarball encoding problem | 20:56 |
| clarkb | trying to remember where we ended up on that problem and iirc the idea was to not set the encoding on .gz files when uploading them to the fileserver? that way they are served as normal files? | 20:57 |
| clarkb | (maybe it was for .tar.gz specifically?) | 20:57 |
| corvus | clarkb: oof, i have completely forgotten that | 20:57 |
| corvus | clarkb: stop setting content-encoding header on already compressed log uploads; stop compressing in fetch-subunit-output and stage-output | 20:58 |
| corvus | clarkb: that's the note i took on that ^ | 20:58 |
| clarkb | corvus: ok I think the order we want is to stop compressing first then stop setting content-encoding header on already compressed uploads so that we don't have a time where the wrong behavior happens for end users | 20:59 |
| clarkb | I'll take a look at the stop compressing bit here once i get the infra meeting agenda mailed out | 20:59 |
| clarkb | (maybe there is already a change for that?) | 20:59 |
| corvus | clarkb: ok. that'll be at the expense of maybe uploading uncompressed data? | 20:59 |
| corvus | i don't recall writing any related changes | 20:59 |
| clarkb | corvus: yes that may result in uploading uncompressed data, but I believe both swift and regular server uploads do compression themselves? | 21:00 |
| clarkb | swift does for sure, not sure about regular file server upload role | 21:00 |
| fungi | it was a matter of apache deciding to do deflate on gzipped files, right? | 21:00 |
| clarkb | fungi: its swift doing it | 21:00 |
| corvus | clarkb: ah good point | 21:01 |
| fungi | er, right, or was a double-header being added by the cdn | 21:01 |
| clarkb | fungi: so when we wget the zuul js tarball we get it uncompressed | 21:01 |
| clarkb | double header isn't an issue in this case (we fixed that problem) | 21:01 |
| fungi | ahh, okay, i'm confusing the two issues | 21:01 |
| clarkb | fungi: I think this other issue happened when you were boarding a boat | 21:01 |
| fungi | quite likely, i did that a loy | 21:02 |
| fungi | er, a lot | 21:02 |
| clarkb | fungi: basically we upload foo.tar.gz and tell swift it is gzip compressed. Then when you download it if you don't set accept-encoding gzip it will helpfully decompress it for you. This hurts us because our upload to the permanent tarballs location has an intermediate step through swift | 21:02 |
| clarkb | end result is we upload foo.tar.gz uncompressed as a valid tar file but not gzip file. | 21:03 |
| clarkb | and this confuses downstream consumers | 21:03 |
| fungi | oh, so we were inadvertently copying uncompressed tarballs to the tarballs site with a .tar.gz extension on them? | 21:03 |
| clarkb | yes | 21:03 |
| fungi | neat | 21:03 |
| fungi | (or i suppose we still are?) | 21:04 |
| clarkb | yup still are | 21:04 |
| *** eharney has quit IRC | 21:04 | |
| fungi | so all the release artifacts on the tarballs site since... some time in recent-ish history.... no longer match the associated openpgp signatures? | 21:05 |
| clarkb | plan is to stop setting the encoding field on upload for gzip files then stop gzipping files explicitly and rely on implicit at upload unless we actualyl want a tarball or similar | 21:05 |
| *** zbr has quit IRC | 21:05 | |
| clarkb | fungi: I think we decided it is only a problem if the intermediate file goes through rax (or ovh one of the two) | 21:05 |
| clarkb | so some of the tarballs | 21:05 |
| fungi | and only tarballs.opendev.org, not tarballs.openstack.org? | 21:06 |
| *** zbr has joined #openstack-infra | 21:07 | |
| fungi | we don't seem to be signing anything published to tarballs.opendev.org anyway, at least | 21:07 |
| fungi | also i note there's a stray, stale py2 wheel of zuul in the mix | 21:08 |
| fungi | at https://tarballs.opendev.org/zuul/zuul/ i mean | 21:08 |
| openstackgerrit | James E. Blair proposed zuul/zuul-helm master: Added nodepool to charts https://review.opendev.org/700462 | 21:09 |
| fungi | if this is also happening for tarballs.openstack.org i may need to work out an audit and try to fetch backups from pypi (those are presumably unaffected at least) | 21:09 |
| clarkb | fungi: it depends on whether or not we stage the files in swift first | 21:10 |
| clarkb | fungi: also we may properly sign the uncompressed data | 21:11 |
| clarkb | in which case the file extension would simply be misleading but the signatures would be valid (as are the tar files) | 21:11 |
| openstackgerrit | Merged opendev/system-config master: lists: Add Airship VMP mailing lists https://review.opendev.org/689271 | 21:12 |
| *** zbr has quit IRC | 21:14 | |
| *** michael-beaver has joined #openstack-infra | 21:15 | |
| fungi | we sign before uploading to pypi, so if the files we're signing are uncompressed then so are the files we're uploading to pypi. i would expect that to show up pretty quickly as breakage if it were the case | 21:15 |
| clarkb | unless the pip toolchain handles a tar file just as well as a tarball (file magic numbers?) | 21:18 |
| * mordred waves to the nice people - sorry, I had less wifi today than I thought I was going to | 21:22 | |
| *** zbr has joined #openstack-infra | 21:22 | |
| *** zbr has quit IRC | 21:26 | |
| fungi | clarkb: possibly. performing an audit of tarballs.openstack.org and what's uploaded to pypi.org might not be a bad idea regardless | 21:26 |
| *** zbr has joined #openstack-infra | 21:30 | |
| *** zbr_ has joined #openstack-infra | 21:37 | |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Stop compressing files during intermediate steps https://review.opendev.org/701282 | 21:38 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Test swift upload logs without encoding type for gzip files https://review.opendev.org/701283 | 21:38 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Test swift upload logs without encoding type for gzip files https://review.opendev.org/701284 | 21:38 |
| clarkb | fungi: corvus ^ I think that stack does what we were talking about | 21:38 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Swift upload logs without encoding type for gzip files https://review.opendev.org/701284 | 21:38 |
| *** zbr has quit IRC | 21:39 | |
| fungi | we've got slightly over 30k files on tarballs.o.o whose names end in "gz" so i'm running `file` against all of them looking to see if any aren't "gzip compressed data" | 21:44 |
| fungi | er, that is to say, on tarballs.openstack.org | 21:45 |
| fungi | (we have only a handful on tarballs.opendev.org by comparison) | 21:46 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Stop compressing files during intermediate steps https://review.opendev.org/701282 | 21:48 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Test swift upload logs without encoding type for gzip files https://review.opendev.org/701283 | 21:48 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Swift upload logs without encoding type for gzip files https://review.opendev.org/701284 | 21:48 |
| fungi | good news everyone! | 21:49 |
| fungi | /srv/static/tarballs/mistral/images/mistral-docker.tar.gz: POSIX tar archive | 21:50 |
| fungi | /srv/static/tarballs/openstack-doc-tools/openstack-doc-tools-1.0.0.tar.gz: Minix filesystem, V3, 32099 zones | 21:50 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Test swift upload logs without encoding type for gzip files https://review.opendev.org/701283 | 21:50 |
| openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Swift upload logs without encoding type for gzip files https://review.opendev.org/701284 | 21:50 |
| fungi | openstack-doc-tools-1.0.0.tar.gz is from 2016-08-22 so presumably unrelated | 21:50 |
| fungi | /srv/static/tarballs/mistral/images/mistral-docker.tar.gz is from last week though | 21:50 |
| fungi | neither has associated signatures (the mistral job presumably doesn't create them, and the openstack-doc-tools sdist is too old) | 21:52 |
| fungi | now checking our ~15k whl files to see if any aren't "Zip archive data" | 21:55 |
| openstackgerrit | Clark Boylan proposed openstack/project-config master: Disable compression of files in stage-output role https://review.opendev.org/701287 | 21:55 |
| clarkb | corvus: ^ thats the site var update | 21:55 |
| *** slaweq has quit IRC | 21:56 | |
| fungi | i suspect this means our python release jobs aren't affected (do we know roughly when the behavior started?) | 21:56 |
| clarkb | fungi: I think we want /srv/static/tarballs/mistral/images/mistral-docker.tar.gz: to be gzip compressed data | 21:57 |
| clarkb | posix tar archive implies it is uncompressed I think | 21:57 |
| fungi | yes, the "good news" was that was the only one i found | 21:57 |
| clarkb | oh gotcha | 21:57 |
| fungi | out of >30k files | 21:57 |
| clarkb | we should maybe ask the release team to not release things for a short bit too | 21:58 |
| clarkb | (while we get this fixed() | 21:58 |
| corvus | fixed() -> True | 21:58 |
| fungi | will do | 21:58 |
| fungi | no whl files were harmed | 22:00 |
| fungi | i'm taking a look on tarballs.opendev.org next to see what the extent is there | 22:00 |
| paladox | File uploads now on the roadmap https://gerrit-review.googlesource.com/c/homepage/+/249235 :) | 22:01 |
| *** zbr_ has quit IRC | 22:02 | |
| *** zbr has joined #openstack-infra | 22:02 | |
| fungi | on tarballs.opendev.org we have these reported as "POSIX tar archive": zuul/zuul/zuul-master.tar.gz zuul/zuul/zuul-js-content-master.tar.gz zuul/nodepool/nodepool-master.tar.gz opendev/gear/gear-master.tar.gz | 22:03 |
| *** slaweq has joined #openstack-infra | 22:11 | |
| fungi | paladox: is that the last remaining bit needed for users to be able to create new changes without ever touching git? | 22:12 |
| paladox | well 99% yes, but you carn't change chmod on it. | 22:13 |
| paladox | it makes our tool at wikimedia redudant | 22:13 |
| paladox | https://tools.wmflabs.org/gerrit-patch-uploader/ | 22:13 |
| fungi | having one less thing to maintain will be nice, i'm sure | 22:15 |
| *** slaweq has quit IRC | 22:15 | |
| clarkb | fungi: are you able to review https://review.opendev.org/#/c/701287/ as the first step in sorting this out? then https://review.opendev.org/701283 is next <- corvus your input on that one would be good too | 22:18 |
| fungi | yeah, i was looking through the stack to make sure i understood | 22:18 |
| fungi | oh though i missed 701287 because it uses a different topic | 22:19 |
| *** zbr has quit IRC | 22:25 | |
| *** zbr has joined #openstack-infra | 22:25 | |
| *** zbr has quit IRC | 22:30 | |
| openstackgerrit | Merged openstack/project-config master: Disable compression of files in stage-output role https://review.opendev.org/701287 | 22:32 |
| smcginnis | Could someone explain why this job gets picked up for python-openstackclient patches? https://opendev.org/openstack/openstackclient/src/branch/master/.zuul.yaml#L3 | 22:40 |
| *** rcernin has joined #openstack-infra | 22:40 | |
| smcginnis | I don't see it being referred to anywhere. | 22:40 |
| smcginnis | Oh, it's part of openstackclient-plugin-jobs which actually is referred to in the local .zuul.yaml | 22:41 |
| fungi | glad we could help! ;) | 22:41 |
| smcginnis | :) | 22:44 |
| *** jamesmcarthur has joined #openstack-infra | 22:44 | |
| smcginnis | So that job assumes and pulls in master, even though projects at least back to stable/rocky use that job template. I think it's a valid job, but does the job's playbook need some sort of handling to make sure it pulls in the right branch. | 22:45 |
| smcginnis | It's causing failures right now because it pulls in master, which pulls in a newer lib than the upper-constraint for a stable/rocky job. | 22:45 |
| smcginnis | Or another way, does required-projects have any awareness of the stable branch it's running in, or will things defined there always just grab the tip of head? | 22:46 |
| *** Goneri has quit IRC | 22:46 | |
| smcginnis | My guess is that at least for now, that job needs to be updated to only run if the patch is on master. | 22:47 |
| smcginnis | Or stable branches that are affected need to not use that template I guess. | 22:48 |
| smcginnis | I *think* this is the right thing to do, but if not, please let me know - https://review.opendev.org/#/c/701293/ | 22:53 |
| fungi | sorry, catching up | 22:54 |
| smcginnis | I am curious if there actually is a way to refer to zuul jobs in the same stable branch so it doesn't lose test coverage, but behavior appears to be that it pulls from master. | 22:55 |
| fungi | smcginnis: so... openstackclient is branchless (well, master-branch only) right? and so should be its plugins | 22:55 |
| smcginnis | Oh right! | 22:55 |
| smcginnis | So really no way to test this on stable branches from this approach. | 22:56 |
| fungi | unless i'm mistaken, this is similar to tempest and its plugins | 22:56 |
| smcginnis | Since openstackclient is branchless, the restricts the job to only really be relevant on master. | 22:56 |
| fungi | so we want the stable/stein job to test that current openstackclient and its plugins work with a stable/stein deployment of openstack | 22:56 |
| smcginnis | Python-openstackclient is broken by this since the stable/rocky version is getting the current (master) release of osc-lib. | 22:57 |
| *** jamesmcarthur has quit IRC | 22:57 | |
| smcginnis | So that probably works well for consumers of osc, but for osc and its plugins themselves, this approach can't work. | 22:57 |
| fungi | yeah, the intent was probably to test that openstackclient works against old clouds... but old clouds consuming python-openstackclient with constrained versions are probably going to need slightly different jobs? | 22:58 |
| smcginnis | Good, we're at least coming to the same general conclusions at least, so that does help. ;) | 22:59 |
| fungi | yeah, so openstackclient (frontend) branchless and uses latest python-openstackclient | 22:59 |
| fungi | python-openstackclient itself (backend) branched with openstack releases, consumed by frontends and also for service-to-service interactions | 23:00 |
| fungi | what's an example of one of the plugin repo names? | 23:01 |
| *** tkajinam has joined #openstack-infra | 23:01 | |
| smcginnis | This was brought to my attention by a patch to python-openstackclient's stable/rocky branch trying to run this job and failing. | 23:01 |
| fungi | yeah, just trying to work out if the openstackclient plug-ins are branched or master-only | 23:02 |
| fungi | whether they follow openstackclient or python-openstackclient basically | 23:02 |
| *** jamesmcarthur has joined #openstack-infra | 23:03 | |
| smcginnis | I haven't looked at the plugins themselves, but python-openstaclient is branched, so I'd assume the plugins are too. | 23:03 |
| fungi | so they're plug-ins to python-openstackclient, not openstackclient? | 23:04 |
| *** zbr has joined #openstack-infra | 23:04 | |
| smcginnis | Yeah, I always need to check that too. | 23:04 |
| smcginnis | python-openstackclient is all the actual code. | 23:04 |
| smcginnis | openstackclient is a metapackage that just pulls in python-openstackclient and the most common plugins | 23:05 |
| fungi | ahh, right, the openstackclient repo is metadata, docs and tests/jobs | 23:05 |
| smcginnis | Mostly a convenience thing so you can just install that and not have to install eleventy different individual packages. | 23:05 |
| fungi | https://opendev.org/openstack/openstackclient/src/branch/master/requirements.txt pulls in the plug-ins | 23:05 |
| fungi | which i guess aren't plug-ins, they're simply the client libraries? | 23:06 |
| smcginnis | Yeah, that appears to be the case. | 23:06 |
| fungi | but i guess python-openstackclient treats them as plug-ins | 23:06 |
| smcginnis | But also why that job being branchless just gets all the latest things which then conflict with what is needed on a stable branched repo. | 23:07 |
| openstackgerrit | Mohammed Naser proposed zuul/zuul-jobs master: Add basic Helm jobs https://review.opendev.org/700222 | 23:07 |
| fungi | going back to the original premise, python-openstackclient (like the other python-*client libraries) is supposed to work with older openstack service releases, presumably there is a desire to test that | 23:08 |
| *** zbr_ has joined #openstack-infra | 23:08 | |
| *** zbr has quit IRC | 23:09 | |
| fungi | so, yes, 701293 seems like the solution if that's the case | 23:09 |
| smcginnis | That's the remaining issue here. This "fixes" things so those branches aren't blocked, but at the expense of actually testing things. | 23:09 |
| *** mattw4 has quit IRC | 23:09 | |
| smcginnis | My guess at this point is that job should actually be defined in python-openstackclient so it follows the stable branching. | 23:10 |
| *** mattw4 has joined #openstack-infra | 23:10 | |
| smcginnis | And it will just have to handle getting everything that openstackclient conveniently pulls in for it. | 23:10 |
| *** Goneri has joined #openstack-infra | 23:10 | |
| *** rlandy|afk is now known as rlandy | 23:11 | |
| fungi | well, i'm unconvinced that it was really testing anything for stable branch changes to the clients | 23:12 |
| fungi | client libraries i mean | 23:12 |
| smcginnis | True, that's a good point. It was running on stable branches, but not necessarily really testing stable. | 23:13 |
| *** zbr_ has quit IRC | 23:14 | |
| fungi | i'm struggling to think what that job *could* test on stable branches of client libraries | 23:14 |
| fungi | i think it's being run for python-openstackclient and possibly other client libraries to make sure they won't break openstackclient, but it will always use the latest versions of them, not stable branch versions | 23:14 |
| smcginnis | I think the general idea is to install python-openstackclient, its plugins and/or the other project clients and make sure that they actually install and run. | 23:14 |
| smcginnis | Or at least I'm guessing that was the intent. Maybe. | 23:15 |
| fungi | yeah, i guess the question is whether there's a desire to try installing and exercising stable branch client libraries and a contemporary version of python-openstackclient | 23:16 |
| fungi | is that something stable branch services rely on for service-to-service communication? | 23:17 |
| *** jamesmcarthur has quit IRC | 23:17 | |
| smcginnis | Probably of limited value related to the effort it would take. | 23:17 |
| smcginnis | I know of services relying on specific service libraries, but not openstackclient managed ones. | 23:17 |
| fungi | that's what i was really wondering, so yeah in that case your proposal there seems sane | 23:21 |
| *** zbr has joined #openstack-infra | 23:21 | |
| *** tosky has quit IRC | 23:23 | |
| *** dchen has joined #openstack-infra | 23:45 | |
| *** rh-jelabarre has quit IRC | 23:51 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: DNM: testing https://review.opendev.org/701237 | 23:52 |
| *** Lucas_Gray has quit IRC | 23:53 | |
| *** mattw4 has quit IRC | 23:56 | |
| *** KeithMnemonic has quit IRC | 23:56 | |
| *** KeithMnemonic has joined #openstack-infra | 23:57 | |
| *** KeithMnemonic has quit IRC | 23:58 | |
| *** KeithMnemonic has joined #openstack-infra | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!