Wednesday, 2022-07-06

*** matt__ is now known as matfechner03:24
rpittaugood morning ironic! o/07:45
jandershey rpittau o/09:29
rpittauhey janders :)09:29
opendevreviewRiccardo Pittau proposed openstack/ironic-python-agent-builder master: Remove CentOS Stream 8 leftovers  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/84882609:30
opendevreviewRiccardo Pittau proposed openstack/bifrost master: [WIP] Move bifrost integration tests to Ubuntu Jammy  https://review.opendev.org/c/openstack/bifrost/+/84882809:50
dtantsurTheJulia: I think we did nuke it in the end10:29
jandersdtantsur iurygregory can I please ask you for a second review on https://review.opendev.org/c/openstack/ironic/+/846859 (verify steps fix on older firmwares/iDRACs)? Thank you!10:55
dtantsurputting on the queue11:03
jandersTY!11:05
iurygregorygood morning Ironic o/11:40
jandershey iurygregory o/11:41
jandersthank you for the +211:41
iurygregoryjanders, yw =)11:42
opendevreviewRiccardo Pittau proposed openstack/bifrost master: [WIP] Move bifrost integration tests to Ubuntu Jammy  https://review.opendev.org/c/openstack/bifrost/+/84882812:28
kubajjHi everyone, I have a question about IPA logs. Where is everything stored? I opened one of the ironic-bm-logs/deploy_logs and I can't figure out if the journal is everything.12:36
dtantsurkubajj: the journal is probably the main thing for DIB-built ramdisks12:40
kubajjdtantsur: thanks12:43
TheJuliagood morning13:01
TheJuliadtantsur: hmm, based upon some unit test stuff yesterday I guess it was only for non-netboot scenarios, but I guess that is totally okay since netboot is going to go poof13:01
dtantsurTheJulia: for netboot you need kernel/ramdisk to do a normal reboot, hence I did not touch it13:04
TheJuliamakes sense now that I'm beginning to sip coffee13:04
opendevreviewDmitry Tantsur proposed openstack/bifrost master: Deprecate os_ironic_node_info  https://review.opendev.org/c/openstack/bifrost/+/84876513:04
dtantsurthis work is by request of the collection team ^^^13:05
dtantsurwe need to clean up our usage of ironic_url (hence also the desire to nuke redeploy-dynamic)13:05
TheJuliaAhh, makes sense13:05
* TheJulia needs https://i.pinimg.com/originals/e8/2f/be/e82fbe370ddc8bc63820f8abc569419a.jpg13:05
dtantsur+++13:06
rpittaulol13:10
opendevreviewMerged openstack/ironic master: Prevent clear_job_queue and reset_idrac failures on older iDRACs  https://review.opendev.org/c/openstack/ironic/+/84685913:15
opendevreviewDmitry Tantsur proposed openstack/ironic master: Trivial: log which state the node is in  https://review.opendev.org/c/openstack/ironic/+/84887414:52
opendevreviewDmitry Tantsur proposed openstack/ironic master: Move logging out of skip_automated_cleaning  https://review.opendev.org/c/openstack/ironic/+/84887515:02
dtantsurJayF: ^^^15:02
TheJuliaI <3 that trivial change15:03
dtantsurIt has a pretty interesting backstory. A person is hitting this error because inspection takes for them exactly 30 minutes, i.e. the inspect wait timeout. So inspector tries to update ports at the same moment as Ironic moves the node to "inspect failed".15:04
TheJuliaugh, all sorts of fun races around there15:05
dtantsuryup. I'm still puzzled why it takes so long.15:05
dtantsursee you tomorrow folks o/15:12
rpittauI wonder if it's pbr not ready for Python 3.10 or viceversa15:17
arne_wiebalckjanders: I *think* sending multiple etags is fine ... did I add this? If I did, I claim that I checked the standard to make sure this is ok :)15:23
TheJuliarpittau: your the second person I've seen remark as such in the last few weeks15:29
rpittauheh having issues with bifrost on jammy, I guess it's early15:30
rpittauoook my head hurts enough, see you tomorrow o/15:45
arne_wiebalckbye everyone o/16:15
EmilienMwe have an interesting (and valid) request on Gophercloud, if anyone wants to have a look too: https://github.com/gophercloud/gophercloud/issues/243116:59
JayFThat's not a library we support, is it?17:03
TheJuliabut it is the support to talk to our api17:03
JayFIs the implication it's our API bug? Or that we should be invested just because we should be invested.17:04
TheJuliaso it has always been "next". I wonder where "nodes_links"17:04
TheJuliacame from17:04
JayFI'm onboard with the latter, just unclear17:04
TheJuliaso, Stephen/Joe wrote that section of code17:05
TheJuliai suspect it was just not an inherent need to understand the pagination and most of their references were much smaller lists of nodes17:06
JayFWe should, at a minimum, comment on the issue and endorse the alternate code linked if it's accurate17:07
JayFI suspect if we just say "this alternate implementation LGTM signed, Ironic dev" it might be sufficient :)17:08
TheJuliaYeah, I'm looking at it now17:08
JayFI am willing to make that comment, the alternate code looks sane to me but I have very little context in gophercloud17:08
JayF(but do write golang for my day job)17:08
TheJuliacommented17:18
TheJuliaYeah, I went back and checked the apiwg stuffs as well17:18
TheJuliaI'm just thinking it would have been prevented if there was more transparency from API docs to details like pagination17:18
opendevreviewJulia Kreger proposed openstack/ironic master: Do not require stage2 for anaconda with standalone  https://review.opendev.org/c/openstack/ironic/+/83576919:54
opendevreviewJulia Kreger proposed openstack/ironic master: Make anaconda non-image deploys sane  https://review.opendev.org/c/openstack/ironic/+/83470919:57
opendevreviewJulia Kreger proposed openstack/ironic master: Do not require stage2 for anaconda with standalone  https://review.opendev.org/c/openstack/ironic/+/83576919:58

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!