| opendevreview | Mahnoor Asghar proposed openstack/ironic master: Add unit test to ensure full node object is not logged https://review.opendev.org/c/openstack/ironic/+/926487 | 10:39 |
|---|---|---|
| iurygregory | good morning Ironic | 10:50 |
| cardoe | JayF: https://review.opendev.org/c/openstack/python-ironicclient/+/924895 is good btw | 13:30 |
| cardoe | dang rebase was for real as we had touched the same lines of code. :/ | 13:30 |
| JayF | cid: re: cardoe's change, we persist disable_ramdisk for runbooks inside the runbook, yeah? So I wonder if cardoe's change needs to ensure steps/disable_ramdisk and runbook are mutually exclusive | 14:52 |
| JayF | wdyt? | 14:52 |
| * JayF put that comment on the change, will remove it if wrong | 14:54 | |
| cid | JayF, cardoe: Exactly, I think it should be invalid to pass in a runbook and disable_ramdisk at the same time. | 14:54 |
| cardoe | As in I should figure out how to hide that option if you're passing --runbook? | 14:55 |
| JayF | I think cid has it setup now wehre if you did | 14:55 |
| cardoe | Whelp. Back to the editor | 14:55 |
| JayF | clean --runbook blah --clean-steps {} it'll error | 14:55 |
| JayF | saying you can only provide one or the other | 14:55 |
| cardoe | okay I can do that. | 14:56 |
| cardoe | As far as the Python 3.12 support stuff... I'm getting nada from the requirements upstream folks. | 14:56 |
| JayF | there's like one person | 14:57 |
| JayF | and they are doing release activities as well | 14:57 |
| JayF | I know that doesn't make it less annoying, but the only way to fix it long term is to get that team more resources | 14:59 |
| * JayF is working on it as he gets more of his ironic work handed over | 14:59 | |
| JayF | Meeting in a minute here though :) | 14:59 |
| JayF | #startmeeting Ironic | 15:00 |
| opendevmeet | Meeting started Mon Aug 19 15:00:31 2024 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 | Hello everyone and welcome to the Ironic weekly meeting | 15:00 |
| JayF | As always we hold meetings under the OpenInfra Code of Conduct at https://openinfra.dev/legal/code-of-conduct | 15:01 |
| JayF | #topic Roll Call | 15:01 |
| JayF | o/ | 15:01 |
| cid | o/ | 15:01 |
| JayF | As is somewhat obvious, I'm running the meeting this morning as Riccardo is out. Going to give a minute or three to see if we can manage quorum. | 15:01 |
| masghar | o/ | 15:01 |
| JayF | I don't think there are any decision points looking at today's agenda, so I'm going to move on and get our usual things on the record :) | 15:02 |
| JayF | #topic Announcements/Reminders | 15:02 |
| JayF | #info Standing reminder to review patches tagged ironic-week-prio and to hashtag any patches ready for review with ironic-week-prio: https://tinyurl.com/ironic-weekly-prio-dash | 15:02 |
| JayF | This is especially important as the milestone 3 is coming up at the end of August. This is the feature freeze observed by most Ironic projects (but not generally by Ironic). This means we should especially prioritize reviews that touch other OpenStack projects. | 15:03 |
| JayF | #info 2024.2 Dalmatian Release Schedule https://releases.openstack.org/dalmatian/schedule.html | 15:03 |
| JayF | PTL nomination period is currently open. Riccardo is currently the only nominee for Ironic PTL; thanks for his volunteering for another term. | 15:04 |
| JayF | #info the next OpenInfra PTG which will take place October 21-25, 2024 virtually! Registration is now open! https://ptg.openinfra.dev/ | 15:04 |
| JayF | The ironic team has been registered, and people with topics or who want to start gathering context/researching for discussion topics should look at the etherpad: | 15:05 |
| JayF | #topic https://etherpad.opendev.org/p/ironic-ptg-october-2024 | 15:05 |
| JayF | I know this may be the first PTG for some of our new contributors, so just a piece of advice: much of the useful PTG discussion happens beforehand, in the etherpad, while folks have time to do research. | 15:05 |
| JayF | That concludes announcements/reminders from the agenda | 15:06 |
| JayF | #topic Review Ironic CI Status | 15:06 |
| JayF | Seems OK to me? I've not seen too much trouble, I had a patch pushed yesterday pass CI. | 15:06 |
| cid | Generally okay. Yeah! | 15:06 |
| JayF | Good stuff, thanks for the report! | 15:07 |
| JayF | No discussion topics for this week, skipping | 15:07 |
| JayF | #topic Bug Deputy Updates | 15:07 |
| JayF | I believe cid was bug deputy, anything interesting to report? | 15:07 |
| cid | Nothing much. It was a silent week. 2 New bugs | 15:07 |
| JayF | Sounds good. | 15:08 |
| JayF | Does someone want to volunteer to take it this week? | 15:08 |
| JayF | I anticipate a busy week or I'd take my turn :) | 15:08 |
| cid | I don't mind still! Unless there's anybody else | 15:08 |
| masghar | I will be out for 2 days this week, but I can take next week | 15:09 |
| JayF | #info cid and masghar will group up on bug deputization this week :) | 15:09 |
| JayF | Now you have to cooperate, it's in the agenda, can't hashtag-undo that :P | 15:09 |
| masghar | Sounds good :) | 15:09 |
| JayF | There are no RFEs up for review, skipping that agenda item. | 15:09 |
| JayF | #topic Open Discussion | 15:10 |
| cid | Is there an equivalent tag to ironic-week-prio for other projects in OpenStack? | 15:10 |
| JayF | each project has their own way of doing it, more or less | 15:10 |
| JayF | many have a "Review-Priority" +1/-1 but usually ask that to be restricted to cores | 15:10 |
| JayF | e.g. for Nova, I know they track feature status in etherpads and meetings | 15:10 |
| JayF | and you usally have a champion | 15:10 |
| JayF | so if I need reviews I go edit that etherpad, show up in the meeting, or ask the core champion | 15:11 |
| cid | That's indeed different from Ironic's | 15:11 |
| JayF | Yeah, and the reality is, there are not many projects with so many cores you need a "system" beyond "just ask $person" | 15:11 |
| cid | Well, I have the runbook support patch up in openstack-sdk. I will just ask! | 15:12 |
| JayF | Yeah, #openstack-sdks is a good place for that | 15:12 |
| JayF | but they'll often wait for ironic cores | 15:12 |
| JayF | but they'll often wait for ironic cores/ | 15:13 |
| cid | To review it first? | 15:13 |
| JayF | You may want to #ironic-week-prio it so we can look too | 15:13 |
| cid | Ok, great! | 15:13 |
| * cid adds tag | 15:13 | |
| JayF | On a different note: adamcarthur5 and GR-OSS in general are working on enhancing gophercloud/openstack-exporter for prometheus support. If anyone has a specific use case/question/interest/etc feel free to reach out to me or him. | 15:13 |
| JayF | ^^^ (for ironic) | 15:14 |
| JayF | Any other items for open discussion? | 15:14 |
| JayF | Alright, I'm going to close up the meeting. Thanks for coming o/ | 15:16 |
| JayF | #endmeeting | 15:16 |
| opendevmeet | Meeting ended Mon Aug 19 15:16:10 2024 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:16 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/ironic/2024/ironic.2024-08-19-15.00.html | 15:16 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/ironic/2024/ironic.2024-08-19-15.00.txt | 15:16 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/ironic/2024/ironic.2024-08-19-15.00.log.html | 15:16 |
| kubajj | o/ | 15:18 |
| JayF | \o | 15:22 |
| JayF | cid: kubajj was going to help us with stress-ng gpu stuff | 15:22 |
| JayF | cid: CID, meet Kuba, Kuba is slowly-becoming-veteran of Ironic, I think he's now worked in some capacity at the three ironic pillars (RH, GR, and now CERN) :D | 15:23 |
| cid | Oh, Hey kubajj! | 15:23 |
| kubajj | JayF: msavchen picked it up instead, but the same team | 15:24 |
| JayF | kubajj: meet CID, he's working for GR-OSS on various Ironic things including trying to get support for upstream GPU burn-in as mentioned in my email | 15:24 |
| kubajj | cid: hi o/ | 15:24 |
| JayF | You're multiplying! | 15:24 |
| JayF | Actually I probably met msavchen in Geneva, I bet, but don't have the IRC<>Face mapping setup in my brain lol | 15:24 |
| kubajj | JayF: it's Maryna, she's our Ironic expert (alongside Jose, but I was hired to replace her as her program finishes in September | 15:25 |
| JayF | oh, of course | 15:25 |
| JayF | I think I know her more via her patches than via conversations tbh | 15:26 |
| kubajj | JayF: sad I missed you at CERN by just a month or so :/ | 15:26 |
| msavchen | hello hello, yes, I was there on Openinnfra event | 15:27 |
| JayF | Hey, it won't be the last chance, I hope. | 15:27 |
| msavchen | Hope so:) | 15:28 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: [doc] Clarify Step return values https://review.opendev.org/c/openstack/ironic-python-agent/+/926546 | 15:34 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: [doc] Clarify Step return values https://review.opendev.org/c/openstack/ironic-python-agent/+/926546 | 15:35 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: [doc] Clarify Step return values https://review.opendev.org/c/openstack/ironic-python-agent/+/926546 | 15:35 |
| cardoe | So I'm looking at argparse. You can't have a mutually_exclusive_group that has an argument or another group | 15:37 |
| cardoe | Changed in version 3.11: Calling add_argument_group() or add_mutually_exclusive_group() on a mutually exclusive group is deprecated. These features were never supported and do not always work correctly. The functions exist on the API by accident through inheritance and will be removed in the future. | 15:38 |
| cardoe | Cause that's how I see this working. Mutually Exclusive Group of ("runbook" OR "clean-steps && "disable-ramdisk") | 15:38 |
| JayF | I wonder if we can check late in the code | 15:42 |
| JayF | if both are set -> manually kaboom | 15:42 |
| JayF | or alternatively, just find out that it returns an API error that doesn't require a decoder ring and I'll revise my comment | 15:43 |
| JayF | e.g. if you do clean --runbook X --disable-ramdisk True # if it just runs the runbook that is bad; if it returns a 400: Give me a runbook OR a disable_ramdisk/steps | 15:43 |
| JayF | then that's more OK :) | 15:43 |
| JayF | mainly just trying to avoid a case where someone passes --disable-ramdisk=true and ironic ignores it and/or they get an impenetrable error | 15:44 |
| cid | cardoe: Unless I did not understand your last message, you will not need to call `add_argument_group`, just including `disable_ramdisk` with `add_argument` is what you need. | 15:54 |
| cardoe | Yeah that's fair. I'll play with it. I gotta now stand up one that has runbooks in it. | 15:55 |
| cardoe | cid: Where am I adding "disable_ramdisk" too? Which group? | 15:55 |
| cardoe | Cause you created a mutually_exclusive_group which means in that group only 1 can be specified. "clean_steps" OR "runbook" | 15:56 |
| cardoe | I can't add "disable_ramdisk" to that cause that's an optional for "clean_steps" | 15:56 |
| cid | Oh, I see what you mean now. | 15:56 |
| JayF | I'd totally suggest just going the route of seeing how the API complains about it | 15:56 |
| cid | You will have to manually validate that | 15:56 |
| JayF | and I'll note the local API setup I espouse in my youtube video, and is documented, is probably enough for this problem | 15:56 |
| JayF | so you shouldn't need a full devstack to determine the behavior if our logging is good enough | 15:57 |
| cardoe | So I've got kind and k3s locally that run ironic. I just haven't gotten around to building my own containers. | 15:57 |
| cardoe | I'm consuming loki built containers so they're on stable versions. | 15:57 |
| cid | Right! I think adding the check at the API level is probablly sufficient | 15:57 |
| JayF | runbooks hasn't made it into Ironic releases yet at all, fwiw, even bugfix ones | 15:58 |
| *** dking is now known as Guest886 | 16:09 | |
| JayF | Do we have any working examples of firmware interface being used to update firmwares on ilo? | 16:37 |
| TheJulia | via ilo itself, or redfish? | 16:39 |
| JayF | via the new redfish interface | 16:39 |
| JayF | I mean: via IRONIC'S IMPLEMENTATION of the new redfish interface | 16:41 |
| TheJulia | @iurygregory ^ | 16:44 |
| iurygregory | JayF, I can provide some examples later, I'm in hurry right now, I need to go to my girlfriend's cousin's funeral | 16:51 |
| *** Guest886 is now known as dking | 16:57 | |
| TheJulia | iurygregory: please ensure some of those examples go into docs. And condolences. | 17:09 |
| dking | TheJulia: It might be a bit late in the day for folks, but regarding the fact that there's no current introspection, is there some way that I could see something from the `data` object from a collector in the `node` object passed on clean/deploy steps? Perhaps some internal way to set some data? | 18:54 |
| TheJulia | hmmmmmmmmm | 18:55 |
| TheJulia | dking: I don't think so off hand, but it seems like it could be added to the task for it to be optionally loaded in | 19:02 |
| TheJulia | but I don't remember how the data and all gets precisely handled on the newest code | 19:02 |
| cid | o/ | 19:28 |
| iurygregory | TheJulia, sure, I will push a patch | 19:40 |
| JayF | We'll need a final release of Ironic libraries before the end of the week, FWIW. I assume release managers have been on that | 20:14 |
| JayF | iurygregory: my condolances, and zero rush on the ask -- Mainly we were curious about the idea of "components" given it appears from their end-user-hp-tooling perspective they mostly just run one command and it gets $stuff. I think they might be operating at a slightly higher abstraction layer and I'm trying to piece together what they need ... on a timeline of like "over the next couple months" | 20:15 |
| iurygregory | JayF, the idea is for us to be able to keep track in our DB, the redfish call itself is just a SimpleUpdate Action with a given url for the firmware (but in the redfish spec there is something about Targets where you could specify the resource that would be updated, but atm no vendors have this working from what I've tested - Dell&HPE) | 20:31 |
| JayF | aha so *in practice* it's generally one package | 20:32 |
| JayF | and that's just a label for future use or for operators with interesting hardware | 20:32 |
| iurygregory | correct | 20:33 |
| iurygregory | since we have plans to expand the "components" front | 20:33 |
| iurygregory | allowing NIC for example | 20:33 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Support Automatic Lessee from instance metadata https://review.opendev.org/c/openstack/ironic/+/924887 | 20:41 |
| opendevreview | cid proposed openstack/ironic master: [WIP] Add inspection rules https://review.opendev.org/c/openstack/ironic/+/918303 | 22:35 |
| opendevreview | cid proposed openstack/ironic master: [WIP] Add inspection rules https://review.opendev.org/c/openstack/ironic/+/918303 | 22:49 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!