Thursday, 2022-09-29

*** rlandy is now known as rlandy|out00:07
*** TheMaster is now known as Unit19300:30
*** ysandeep|sick is now known as ysandeep03:48
*** ysandeep is now known as ysandeep|afk04:45
*** ysandeep|afk is now known as ysandeep06:19
*** jpena|off is now known as jpena06:51
*** lbragstad8 is now known as lbragstad06:56
amorinI did a patchset on mistral last day, but it's not checked by zuul:07:31
amorinI did a recheck this morning, the build appeared on zuul page but disappeared few sec after07:31
amorinis there any issue with zuul at the moment?07:31
*** ysandeep is now known as ysandeep|lunch08:05
mnasiadkaclarkb: seems it booted, thanks08:26
frickleramorin: mistral needs to clean up their queue definitions
fricklersimilar to e.g.
amorinfrickler: ack, should I do something?08:32
frickleramorin: yes, you can propose a patch like ^^08:33
amorinperfect, will do08:33
fricklerseems a lot of projects have problems fixing this issue, maybe the TC should start to be more active in helping them08:33
fricklerwe are still at 444 config errors for the openstack tenant08:34
frickler<ad> might be interesting for some people here, too: Large Scale Ops Deep Dive: Schwarz Group, today 14 UTC </ad>09:37
*** ysandeep|lunch is now known as ysandeep09:44
*** rlandy|out is now known as rlandy10:28
*** dviroel|out is now known as dviroel11:17
fungiyep, thanks for the reminder! i always try to watch those11:29
*** ysandeep is now known as ysandeep|afk12:12
fungiplayed around with a bit earlier today and it gives some pretty helpful suggestions. i was skeptical, but for the most part i agreed with most of its findings12:17
*** dviroel is now known as dviroel|doc_appt12:27
*** dasm|off is now known as dasm12:48
*** ysandeep|afk is now known as ysandeep12:49
*** marios is now known as marios|call14:18
*** ysandeep is now known as ysandeep|dinner14:28
*** marios|call is now known as marios14:45
*** dviroel|doc_appt is now known as dviroel|doc14:46
*** dviroel|doc is now known as dviroel14:46
opendevreviewMerged openstack/project-config master: charms: add new ceph subteam
*** marios is now known as marios|out15:37
*** ysandeep|dinner is now known as ysandeep|out15:56
*** jpena is now known as jpena|off16:05
clarkbfungi: some people were complaining that refurb suggest some far less readable code for file IO operations. Apparntly python has some pythonic apis for that that diverge from classic posix open/close/read/write and refurb will happily suggest you change from the classics16:10
clarkbBut its meant to give suggestions not provide a singular opinion that must be followed16:10
fungiright, and in my case i'm going to add it to tox on one of my projects but set it to skip some of those rules16:37
fungia very old codebase i've been incrementally modernizing, and so has a lot of python 2.x and even ancient 1.x patterns (where those things continued working in py3k  anyway)16:38
JayFIf you're not familiar with it, black is also a useful utility. Sorta like gofmt for python.16:45
fungiyeah, i disagree with rather a lot of black's suggestions, but it's definitely gained a lot of popularity16:46
fungithe codebase in question currently does lint with pycodestyle/pyflakes (via flake8+bugbear), bandit, and codespell16:47
fungiand both its yaml inputs and outputs are checked with yamllint16:48
JayFYou can flip some of the opinionated stuff on/off. Honestly I'm not a person who cares about *what* style, so much as there is *consistent* style :D 16:48
fungicoercing pyyaml to write yaml which yamllint likes is no small feat, let me tell you16:49
clarkbthe thing I like about black is it is specifically built to make code review easier via cleaner diffs16:49
fungii end up shadowing some of pyyaml's writer methods to make indentation work16:49
clarkbI can manage for weirdness if it makes code review easier16:49
JayFThe controversial/awkward thing about code formatting tools is how to run them. e.g.: automatically run in CI and commit back differences? Run in check mode in CI and make the dev do a round trip if they didn't run formatting? I even had someone suggest once to run a nightly job to make the main branch conform (NOPE) 16:51
fungilooking back over the code, i substitute my own yaml.emitter.Emitter.expect_block_sequence() method to get the block sequences indented in a yamllint-compatible way, and then override the yaml.SafeDumper class to call it16:52
fungiapparently yamllint wants block sequences overindented16:52
fungiand pyyaml's emitter lacks an option to do that16:53
opendevreviewMerged zuul/zuul-jobs master: Clarify that stage-output doesn't actually write to zuul_output_dir
*** chkumar|rover is now known as raukadah17:15
clarkbthe mailman mailing list is quite active, but no response to my request for guidance on those issues yet17:18
opendevreviewClark Boylan proposed opendev/system-config master: Update our Gerrit images
clarkbinfra-root ^ that change should run gerrit 3.5 + mina 2.8 + new jgit through our CI system. I don't think we should land that until after the openstack release out of an abundance of caution, but if somethign is horribly broken I want to be able to get that feedback upstream asap hence this change17:29
*** dviroel is now known as dviroel|afk19:42
clarkbour CI testing seems happy enough with updated gerrit 3.5. Thats good20:51
*** dviroel|afk is now known as dviroel20:52
clarkbarg I can't push to gerrit right now apparently I no longer have a valid CLA20:58
clarkbbut also it seems to think my account is deleted in some contexts. I really wish I understood what was going on here20:58
*** tosky_ is now known as tosky21:24
*** dasm is now known as dasm|off21:24
clarkbGerrit is asking for community help to test Gerrit 3.7 RCs: we can probably add 3.7 to our job list and run it through the testing we do21:56
*** rlandy is now known as rlandy|out22:10
*** ysandeep|out is now known as ysandeep23:58

Generated by 2.17.3 by Marius Gedminas - find it at!