Monday, 2021-09-20

*** ramishra_ is now known as ramishra05:38
*** ysandeep|away is now known as ysandeep05:45
dpawlikhey fzzf1: please join #softarefactory-project:matrix on matrix or #sofwarefactory-project on libera server06:04
dpawlikfzzf1: maybe better join on matrix, we moved from IRC server 06:06
dpawlikfzzf1: tl;dr you want to configure zuul to use not a Openstack instance but your local host (https://docs.softwarefactory-project.io/sf-config-3.6/user/nodepool_user.html?highlight=cloud - this one is default) . I guess what you are looking for is a static driver https://zuul-ci.org/docs/nodepool/static.html 06:17
dpawlikfzzf1:if your job can be executed in containers, you can try hypervisor-k1s. More info we can share on matrix  :)06:19
dpawlikbtw, clarkb, fungi: hey. I need to push one more change for PS https://review.opendev.org/c/opendev/puppet-log_processor/+/809424 . Please take a look in free time ;)06:24
*** ykarel_ is now known as ykarel|afk06:54
*** jpena|off is now known as jpena07:27
*** rpittau|afk is now known as rpittau07:55
*** ykarel|afk is now known as ykarel07:58
*** ykarel_ is now known as ykarel08:09
*** fressi_ is now known as fressi08:40
*** ysandeep is now known as ysandeep|lunch08:53
*** ykarel_ is now known as ykarel|lunch10:10
*** ysandeep|lunch is now known as ysandeep10:35
*** rlandy is now known as rlandy|rover10:56
*** ykarel|lunch is now known as ykarel11:12
*** jpena is now known as jpena|lunch11:29
*** jcapitao is now known as jcapitao_lunch11:30
*** ysandeep is now known as ysandeep|afk11:37
*** jpena|lunch is now known as jpena12:22
*** jcapitao_lunch is now known as jcapitao12:27
fungidpawlik: clarkb: i've raised a question and a -1 on that change for now, pending confirmation it won't cause a problem12:33
fungisee my review comment for details12:34
dpawlikfungi: sure, will reply soon ;) Thanks for review12:35
fungiyw, other than the python dependencies being present for the new interpreter, i'm fine with that change12:35
opendevreviewMerged openstack/project-config master: Fix missing charm-interface-ironic-api  https://review.opendev.org/c/openstack/project-config/+/80819012:47
*** ysandeep|afk is now known as ysandeep12:47
opendevreviewYuriy Shyyan proposed openstack/project-config master: Scaling down proposed instance count.  https://review.opendev.org/c/openstack/project-config/+/81000913:03
dpawlikGood point fungi++. On clean environment it require one more patch13:18
opendevreviewMerged openstack/project-config master: Scaling down proposed instance count.  https://review.opendev.org/c/openstack/project-config/+/81000913:33
dpawlikfungi, may I ask you to check https://review.opendev.org/c/opendev/base-jobs/+/809765 ? Do you think that it can be merge in the future? 14:04
fungilookin14:19
*** ykarel is now known as ykarel|away14:39
dpawlikthanks for feedback fungi++ !14:42
fungiyw14:46
clarkbdpawlik: fungi  I left a comment on 809765 I've -1'd it because I can probably be convinced to chagne my mind but feel pretty strongly about that. Consider it more like a soft -215:25
fungii do feel like if it's a short-term stopgap while a polling-based replacement is being developed, then it's probably not terible in the interim, but in that case i'd want to have some sort of a deadline on the horizon for being able to remove it15:26
dpawlikfungi, clarkb: of course it's something that it would work temporary. I have another idea: a CI job that will pull new logs from previous job and push it to RDO endpoint. It does not need to modify base-jobs...15:37
clarkbdpawlik: you don't even need that. You can just modify the jobs that need to push the records to do it from their post-run playbook I think15:41
clarkbbut then it is controlled by the interested party and not present in an untestable base job15:42
fungibut also if it's just periodically polling the builds or buildsets endpoint in the zuul api, making that a ci job is probably overkill. it could be as simple as a cronjob on the server coordinating the ingestion work, since all the necessary context is provided by the zuul api anyway15:49
*** rpittau is now known as rpittau|afk16:00
dpawlikclarkb, fungi: I get it. Thanks for that idea16:04
*** ysandeep is now known as ysandeep|dinner16:25
fungifor those not following along in #opendev, the openstack integrated gate queue is currently not starting any builds, we're looking into it and should hopefully have it going again shortly16:52
*** ysandeep|dinner is now known as ysandeep17:05
*** jpena is now known as jpena|off17:19
fungiand things are moving again17:22
*** ysandeep is now known as ysandeep|away18:04
clarkbfungi: heads up that nova is apparently rejecting changes from our bots because we don't appropriately cherry pick them or annotate them as stable only even though they are clearly stable only if actually reviewed22:29
clarkbI think that nova should fix that in the bot if they are going to insist on silly rules for this stuff22:29
clarkbhttps://zuul.opendev.org/t/openstack/build/5ce992bbc3c243bf8b118c6ec097a0cd is an example22:29
fungiwell, those changes are provided as a convenience. they're free to ignore, abandon or revise them too22:31
clarkbya I'm mostly just grumping that we keep piling on silly rules to stable branches and no one ever seems to understand that sometimes changes originate on a stable branch and that is ok. We don't need special tags or authorization to handle those. Reviewers are capable of reviewing a change and determining if it is appropriate22:32

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!