adam_metal3 | Hello Ironic | 06:12 |
---|---|---|
dtantsur | JayF: I think we already have the logic in IPA to heartbeat after a command is finished? | 07:00 |
rpittau | good morning ironic! o/ | 08:14 |
opendevreview | Verification of a change to openstack/bifrost master failed: Set [inspector]require_managed_boot explicitly https://review.opendev.org/c/openstack/bifrost/+/877469 | 11:25 |
dtantsur | huh, something broke in bifrost? | 11:30 |
rpittau | I hope not | 11:31 |
rpittau | cs9 seems unhappy | 11:31 |
rpittau | like all cs9 jobs, coincidence? :/ | 11:34 |
opendevreview | Merged openstack/python-ironicclient master: Accept configdrive as a JSON file https://review.opendev.org/c/openstack/python-ironicclient/+/876199 | 11:39 |
dtantsur | rpittau: yeah, a worrisome one | 11:48 |
iurygregory | good morning Ironic | 12:07 |
opendevreview | Verification of a change to openstack/ironic master failed: Add error logging on lookup failures in the API https://review.opendev.org/c/openstack/ironic/+/877786 | 12:12 |
kaloyank | hello ironic o/ | 12:18 |
opendevreview | Verification of a change to openstack/ironic master failed: Add error logging on lookup failures in the API https://review.opendev.org/c/openstack/ironic/+/877786 | 12:47 |
iurygregory | grenade seems to be failing when creating nova resources https://zuul.opendev.org/t/openstack/build/21790f36aef540d9a7af420356baeafb/log/controller/logs/grenade.sh_log.txt#1903 looking at the nova api it can't find our flavor O.o https://zuul.opendev.org/t/openstack/build/21790f36aef540d9a7af420356baeafb/log/controller/logs/screen-n-api.txt#4224 | 12:59 |
dtantsur | iurygregory: if that's the same issue as the one I was debugging last week, it has a lot of red herrings, but the actual problem is the floating IP not pingable. | 13:20 |
iurygregory | ouch .-. | 13:20 |
dtantsur | that's all I've got really... no clue why, no clue why now | 13:20 |
iurygregory | did you notice if it was specific to some cloud provider? | 13:21 |
dtantsur | haven't checked that | 13:24 |
iurygregory | ack, i will check the failed jobs on the history | 13:34 |
rpittau | dtantsur: I'm quite puzzled by the failure in bifrost CI, there's no console logs and the only interesting thing I'm seeing in the logs is "lient-side error: Can not delete node "4e41df61-84b1-5856-bfb6-6b5f2cd3dd11" while it is in provision state "clean wait"" | 14:21 |
rpittau | it doesn't look like a problem with selinux, for once | 14:26 |
TheJulia | dtantsur: when you get some time, some more context on your point of view is desirable on https://review.opendev.org/c/openstack/ironic/+/879210 | 14:26 |
TheJulia | we're in you seem to want a feature where the bug I'm trying to fix seems distinctly different territory, which is why | 14:27 |
dtantsur | TheJulia: mmm, the thing you're trying will just not work, I'm trying to make it work :) | 14:27 |
TheJulia | I need to understand the case you think we desire | 14:27 |
dtantsur | if you don't account for ports in hash ring, your hash ring will ignore any conductors on non-standard ports | 14:27 |
TheJulia | Again, I think we need to understand the use case, I know that, but why would we have multiple distinct conductor processes sharing the same hostname | 14:28 |
TheJulia | that is not the problem steve is trying to get solved | 14:28 |
TheJulia | hence my confusion | 14:28 |
dtantsur | Oh, well, I was thinking several conductors per machine | 14:28 |
dtantsur | if you don't care about that, apologies. but we need that eventually, so ideally your patch should not make it impossible/hard in the future. | 14:29 |
TheJulia | Do we want to go ahead and force operators to take the hash ring structure change in upgrade? | 14:29 |
TheJulia | I was wanting to avoid that | 14:29 |
TheJulia | *entirely* if at all possible | 14:29 |
dtantsur | we can do a hack. we can skip adding the port if it matches the default | 14:29 |
dtantsur | like you don't normally write google.com:443 in your browser | 14:30 |
TheJulia | true, only for the conductor saving the entry I guess | 14:30 |
TheJulia | which simplifies the overall path... | 14:30 |
dtantsur | it can be processed on any side, but if we keep rolling upgrades in mind, it's better to avoid modifying existing conductor records | 14:31 |
TheJulia | we could entirely hide logic in ironic-lib I guess than for client to grok/support | 14:31 |
TheJulia | we just need to make sure we don't have a raw ipv6 address... | 14:31 |
TheJulia | but likely early on in ironic | 14:31 |
dtantsur | yeaah, IPv6... | 14:31 |
TheJulia | which is why I was like "a port column makes sense!" | 14:31 |
TheJulia | albeit getting that port out and in in the end gets pretty nasty unless it is just the conductor hostname | 14:32 |
dtantsur | true, but then we'll have to change the primary key one day | 14:32 |
TheJulia | pkey to what one day? Why do you say that? | 14:32 |
TheJulia | (it seems like out of the blue) | 14:32 |
TheJulia | ... and pkey for what | 14:33 |
dtantsur | sorry, not pkey, the unique constraint on conductor hostname | 14:33 |
dtantsur | if we ever have multiple conductors per hostname | 14:33 |
TheJulia | yeah, which is a plus to just reuse hostname then | 14:33 |
TheJulia | and add the port | 14:33 |
dtantsur | the V6 thing.. we could have a release note saying "if you use a naked IPv6 as a host name, we will change it to add brackets. be prepared" | 14:34 |
TheJulia | I was thinking of just adding some sanitization/guard code in the process of registering | 14:35 |
TheJulia | the conductor on startup | 14:35 |
TheJulia | seems easy enough | 14:35 |
TheJulia | .. one of those "I'll spend longer on the reno things if I don't just add logic to fix things | 14:36 |
TheJulia | " thing | 14:36 |
dtantsur | yep. sorry for making you do more work, but I think running several conductor per machine is very cool | 14:36 |
TheJulia | Nah, I can see the reason/logic there, It just came out of left field compared to the problem | 14:36 |
TheJulia | and honsestly, it is the same amount of work and likely removes a patch | 14:36 |
TheJulia | ... that I didn't like anyhow | 14:36 |
JayF | meeting in two | 14:58 |
JayF | #startmeeting ironic | 15:00 |
opendevmeet | Meeting started Mon Apr 3 15:00:04 2023 UTC and is due to finish in 60 minutes. The chair is JayF. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
opendevmeet | The meeting name has been set to 'ironic' | 15:00 |
JayF | Good morning all, who all is here? | 15:00 |
rpittau | o/ | 15:00 |
JayF | I'm going to try and make this fast since many of us spent the majority of last week in planning | 15:00 |
JayF | #topic Announcements/Reminder | 15:00 |
JayF | Reminder: tag your patches ironic-week-prio; and review priority patches | 15:00 |
TheJulia | o/ | 15:00 |
JayF | I'd say, with PTG just ending, put a special emphasis on getting specs landed on topics discussed at PTG while they are fresh | 15:01 |
TheJulia | ++ | 15:01 |
JayF | #note Ironic meeting cancelled for next week, April 10, due to many contributors being out of office or on holiday. | 15:01 |
matfechner | o/ | 15:02 |
JayF | Congrats on an Antelope release, it's done and announced. It's the 18th release of Ironic as part of an integrated OpenStack release. | 15:02 |
rpittau | \o/ | 15:02 |
JayF | There were no action items from last meeting; skipping that piece of the agenda | 15:02 |
dtantsur | o/ | 15:02 |
JayF | #topic Ironic CI status | 15:03 |
JayF | It sounded like someone was saying Bifrost+CentOS 9 was unhappy this morning? | 15:03 |
rpittau | yeah | 15:03 |
rpittau | I've rechecked but it doesn't look like a one time issue | 15:03 |
iurygregory | o/ | 15:04 |
rpittau | not many info from the logs | 15:04 |
JayF | Ack; so please do not recheck bifrost things unless you fixed a CI issue | 15:04 |
JayF | Everything else doing OK? | 15:04 |
iurygregory | grenade job is failing | 15:04 |
iurygregory | https://zuul.opendev.org/t/openstack/builds?job_name=ironic-grenade&project=openstack%2Fironic&branch=master&skip=0 | 15:04 |
dtantsur | not always, but frequently | 15:05 |
iurygregory | yeah | 15:05 |
JayF | ack; not awesome but we'll get it sorted | 15:05 |
JayF | going to move on | 15:06 |
JayF | #topic ironic-python-agent-builder changes | 15:06 |
JayF | this was added to the agenda by someone else; is there someone who can talk about this? | 15:06 |
rpittau | yep | 15:06 |
rpittau | it was me | 15:06 |
rpittau | I checked the ipa bulder jobs and they look a bit outdated | 15:06 |
rpittau | also we're still suipporting ussuri with one job | 15:06 |
rpittau | I think we should remove the ussuri job and move over from bionic-focal to focal-jammy | 15:07 |
TheJulia | seems reasonable | 15:07 |
JayF | SGTM | 15:07 |
opendevreview | Kaifeng Wang proposed openstack/ironic master: Remove instance image after deployment https://review.opendev.org/c/openstack/ironic/+/878680 | 15:08 |
JayF | rpittau: good to move on? | 15:09 |
rpittau | also on the arm64 topic, we're running the jobs on debian bullseye, it would be worth to evaluate running that using jammy instead | 15:09 |
JayF | I'm onboard for anything that makes that CI look/act more like our existing CI | 15:09 |
rpittau | not sure when bookworm is being released | 15:09 |
JayF | *other existing ci | 15:10 |
rpittau | alright, I'll get to the nodeset migration first, then the arm64 build | 15:10 |
rpittau | thanks, all good :) | 15:10 |
JayF | Thanks for championing that! | 15:11 |
JayF | #topic PyPI Maintainer Audit | 15:11 |
JayF | Ironic is making progress on this. I've emailed our various emeritus-contributors and asked them to clean stuff up, we've got a couple folks who have taken action. | 15:11 |
JayF | Just wanted to update the group. | 15:11 |
JayF | #topic VirtualPDU | 15:11 |
JayF | #note VirtualPDU rename happening Friday, April 7 during a maintenance window. | 15:12 |
rpittau | awesome | 15:12 |
JayF | There should be no need for action from Ironic devs, and AIUI the best thing we can do is just stay away from that repo before/during the migration | 15:12 |
JayF | #topic Open Discussion | 15:12 |
JayF | any items for open discussion? | 15:12 |
JayF | TheJulia: planning on mashing buttons to enable LP this week? | 15:13 |
JayF | if so we should probably note it | 15:13 |
TheJulia | JayF: yes | 15:13 |
JayF | #note TheJulia to execute on a launchpad-default-bug migration. No data to be migrated automatically. | 15:13 |
JayF | Anything else? | 15:13 |
JayF | Thanks everyone o/ | 15:16 |
JayF | #endmeeting | 15:16 |
opendevmeet | Meeting ended Mon Apr 3 15:16:47 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:16 |
opendevmeet | Minutes: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-04-03-15.00.html | 15:16 |
opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-04-03-15.00.txt | 15:16 |
opendevmeet | Log: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-04-03-15.00.log.html | 15:16 |
opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: Remove ussuri CI jobs https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/879357 | 15:18 |
JayF | FYI; I have a 2023.2 workstreams spec in progress, probably about 75% done | 15:22 |
JayF | I'll have it up this morning, likely before noon | 15:22 |
TheJulia | bugs.launchpad.net/ironic lives again | 15:44 |
TheJulia | now to clean it up | 15:44 |
JayF | pretty sure there are some project-config things too? but I don't know exactly what | 15:45 |
TheJulia | there are someplace | 15:45 |
rpittau | https://media.tenor.com/ytDtlG2gxLsAAAAC/its-alive-young-frankenstein.gif | 15:45 |
clarkb | TheJulia: JayF https://opendev.org/openstack/project-config/src/branch/master/gerrit/projects.yaml#L1227-L1231 and so on for the various ironic projects | 15:47 |
clarkb | you would drop use-storyboard: true. I think the groups: - ironic can stay? I'm not positive on that at the moment | 15:48 |
JayF | FYI I have a downstream item which will be the primary focus for me this week; I'll keep up with reviews and post-PTG items but in general need to finish a thing downstream | 15:53 |
JayF | and next week I'm gone :) | 15:53 |
JayF | (vacation) | 15:53 |
rpittau | good night! o/ | 16:02 |
opendevreview | Mohammed Boukhalfa proposed openstack/sushy-tools master: Add fake_ipa inspection, lookup and heartbeater to fake system https://review.opendev.org/c/openstack/sushy-tools/+/875366 | 16:21 |
opendevreview | Verification of a change to openstack/ironic master failed: Add error logging on lookup failures in the API https://review.opendev.org/c/openstack/ironic/+/877786 | 16:23 |
opendevreview | Verification of a change to openstack/ironic master failed: Add error logging on lookup failures in the API https://review.opendev.org/c/openstack/ironic/+/877786 | 16:34 |
TheJulia | https://bugs.launchpad.net/ironic/ is down to 248 open bugs | 17:22 |
TheJulia | I suggest we have a high bandwidth bug scrub and aggressively wontfix/update items | 17:22 |
TheJulia | I did a cursory pass updating stuff and took care of about `50 | 17:22 |
TheJulia | 150 | 17:22 |
JayF | I'd be game to do that sync with a couple of folks | 17:28 |
opendevreview | Julia Kreger proposed openstack/ironic-lib master: Add jsonrpc client port capability https://review.opendev.org/c/openstack/ironic-lib/+/879211 | 18:31 |
opendevreview | Verification of a change to openstack/ironic master failed: Add error logging on lookup failures in the API https://review.opendev.org/c/openstack/ironic/+/877786 | 19:45 |
NobodyCam | Good Afternoon Ironic folks, happy Monday | 20:29 |
TheJulia | o/ NobodyCam | 20:46 |
opendevreview | Jay Faulkner proposed openstack/ironic-specs master: Add 2023.2 Workitems discussed at Ironic PTG https://review.opendev.org/c/openstack/ironic-specs/+/879381 | 20:58 |
JayF | https://review.opendev.org/c/openstack/ironic-specs/+/879381 is up, it represents my best understanding of what things were discussed and planned at PTG. It is littered with TODO questions inside for when I ran out of info. | 20:58 |
JayF | Please review it, answer the questions in todos if you can o/ | 20:58 |
opendevreview | Julia Kreger proposed openstack/ironic master: Utilize the JSON-RPC port https://review.opendev.org/c/openstack/ironic/+/879215 | 21:55 |
iurygregory | JayF, tks for the review in the spec! | 22:03 |
* iurygregory shouldn't write spec during midnight looking at the BIOS Settings spec for inspiration :D | 22:04 | |
TheJulia | heh :) | 22:13 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!