15:00:24 <TheJulia> #startmeeting ironic
15:00:25 <openstack> Meeting started Mon Mar 29 15:00:24 2021 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:28 <openstack> The meeting name has been set to 'ironic'
15:00:31 <TheJulia> Good morning everyone!
15:00:35 <rpittau> o/
15:00:38 <kaifeng> o/
15:00:38 <erbarr> o/
15:00:48 <ajya> o/
15:00:55 <rpioso> o/
15:00:59 <TheJulia> Our agenda can be found on the Wiki this week
15:01:01 <TheJulia> #link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting
15:01:04 <dtantsur> o/
15:01:06 <bdodd> o/
15:02:06 <TheJulia> #topic Announcements / Reminders
15:02:10 <rloo> o/
15:02:12 <stendulker> o/
15:02:26 <TheJulia> #info This week is R-2, in other words ironic most likely must release for the cycle this week.
15:02:42 <TheJulia> #info PTG will be April 19th through 23rd
15:02:51 <TheJulia> #link https://april2021-ptg.eventbrite.com/
15:03:01 <TheJulia> #link https://etherpad.opendev.org/p/ironic-xena-ptg
15:03:40 <TheJulia> One last item, I will likely be taking the week before the PTG off.
15:03:47 <TheJulia> Does anyone have anything to announce or remind us of this week?
15:04:11 <rpioso> Bill Dodd will be retiring.
15:04:32 <rpioso> bdodd :-)
15:04:32 <bdodd> That is true!
15:04:36 <TheJulia> #info bdodd is retiring! \o/  Thank you Bill for all your hard work!
15:05:20 <ajya> thanks bdodd and enjoy!
15:05:24 <bdodd> Thank all of you for your help and support!
15:05:35 <rpioso> Thanks, again, bdodd, for all of your contributions, especially promoting Redfish in ironic :-)
15:05:44 <rloo> Congrats bdodd! wait, now bdodd has more time to contribute to ironic! (just kidding)
15:05:52 <rpioso> LOL
15:06:01 <bdodd> :)
15:06:37 <TheJulia> rloo: I was going to wait a few days before trying the "but now you have time to contribute more!
15:06:39 <TheJulia> approach :)
15:06:56 <rloo> ha ha, good luck with that approach TheJulia!
15:07:13 <TheJulia> Anyway, moving on, we had no action items from last week.
15:07:21 <rloo> or err.. are you bored now bdodd? We've got some fun stuff...
15:07:22 <TheJulia> Do we want to review subteam statuses since it is basically release week?
15:08:28 <bdodd> rloo Not bored yet. But if that changes I know where to come. :)
15:08:38 <TheJulia> bdodd: Thanks :)
15:08:59 <TheJulia> I'm going to take the lack of responses for subteam updates, meaning we should just move on to priorities for the coming week
15:09:08 <dtantsur> yep
15:09:37 <TheJulia> #topic Deciding on priorities for the coming week
15:10:10 <TheJulia> So we've basically entirely moved over to the hashtag in gerrit from the whiteboard, but are there any items that need attention or need visibility this week?
15:10:21 <TheJulia> #link https://etherpad.opendev.org/p/IronicWhiteBoard
15:10:26 <TheJulia> Line 112
15:10:28 <dtantsur> https://review.opendev.org/c/openstack/ironic/+/744117 is already there
15:10:51 <dtantsur> but would be nice to Jacob to get it in, he's spent significant time on it this cycle
15:11:13 <TheJulia> Yeah
15:11:42 <TheJulia> Ideally I'd like to get the repository in ready to go condition by wednesday
15:11:53 <rloo> of those, which ones are highly desirable for the release?
15:12:18 <TheJulia> I'd like to see the anaconda stuff get in
15:12:36 <dtantsur> ajya: FYI https://review.opendev.org/c/openstack/ironic/+/759428 needs rebasing
15:12:42 <ajya> can this be revisited, it got rebased and removed dependency from another patch - https://review.opendev.org/c/openstack/ironic/+/755821
15:12:44 <TheJulia> The tempest plugin items are not required
15:13:00 <ajya> dtantsur: yup, will rebase today
15:13:10 <TheJulia> ajya: adding hashtag
15:13:23 <ajya> TheJulia: thanks
15:13:29 <rloo> would it be ok to remove the hashtag for the ones that are not required? (if it helps wrt prioritizing for the release?)
15:13:52 <dtantsur> ajya: approved
15:14:01 <openstackgerrit> Merged openstack/ironic master: Validate the kickstart template and file before use  https://review.opendev.org/c/openstack/ironic/+/777448
15:14:09 <rpioso> :-)
15:14:17 <TheJulia> rloo: sure
15:14:42 <openstackgerrit> Merged openstack/ironic-python-agent stable/train: Do not try to read raw disk data as unicode  https://review.opendev.org/c/openstack/ironic-python-agent/+/782881
15:14:48 <rloo> and the goal is to review/merge by EOD Wed?
15:14:49 <ajya> thanks, dtantsur
15:14:56 <dtantsur> review by EOD Tue
15:15:02 <TheJulia> ++
15:15:05 <dtantsur> to allow for a recheck party :)
15:15:12 <rloo> oh. okay...
15:15:55 <TheJulia> I'm going to remove the dib item, since the required diskimage-builder patch has not gone anywhere
15:16:12 <TheJulia> well, ipab
15:17:20 <openstackgerrit> Yogesh proposed openstack/ironic master: Add idrac HW type IPMI interface support  https://review.opendev.org/c/openstack/ironic/+/771862
15:18:52 <TheJulia> I think the current state of https://review.opendev.org/q/hashtag:ironic-week-prio+status:open looks good to me
15:19:44 <rpittau> yup
15:19:48 <dtantsur> yeah
15:19:54 <TheJulia> Well, I guess we're good to move on to Discussion?
15:20:17 <rpioso> TheJulia: https://review.opendev.org/c/openstack/ironic/+/771862 is a Dell priority.
15:22:46 <TheJulia> Okay, but it has two -2's. You need to engage discussion with those reviewers to get them to reconsider or re-evaluate their stance based on the current state of the patch.
15:23:34 <rpioso> TheJulia: Will do.
15:23:38 <dtantsur> if "idrac-ipmitool-shellinabox" in patch: patch.apply(-2)
15:24:03 <TheJulia> Okay...
15:24:12 <TheJulia> Moving on to Discussion since we have a meeting agenda
15:24:17 <TheJulia> #topic Discussion
15:25:03 <TheJulia> We have one discussion item today, specifically adjustment of standing items like Friday SPUC and Tuesday review jam since EU daylight savings time has struck everyone in the northern hemesphere.
15:25:38 <TheJulia> I *believe* we want to try and see if there is consensus to adjust the times forward by an ?hour? or 2 for each?
15:26:13 <dtantsur> for the morning instance I'd like to keep the same *local* time, even though initially SPUC was announced in UTC timezone
15:26:20 <dtantsur> but I need to check with Jacob if that's okay
15:27:43 <TheJulia> I think that is okay
15:28:14 <TheJulia> Are there any objections to possibly moving the times a little bit?
15:28:22 <TheJulia> arne_wiebalck: anything for the baremetal sig?
15:28:25 <dtantsur> janders: if I forget to ping you tomorrow ^^^
15:29:29 * TheJulia hears no objections
15:29:37 <TheJulia> Moving on then
15:29:40 <TheJulia> #topic Baremetal SIG
15:29:54 <TheJulia> The only item I'm aware of is I volunteered to talk about Secure RBAC in two weeks :)
15:30:05 <TheJulia> Tuesday, April 12th, 2021 @ 2pm UTC
15:30:12 <TheJulia> #link https://etherpad.opendev.org/p/bare-metal-sig
15:30:26 <TheJulia> Looks like arne_wiebalck is AFK today
15:30:43 <TheJulia> Anyhow, if nobody has anything sig related, we can move on to RFE review
15:31:05 <dtantsur> ++
15:31:20 <TheJulia> #topic RFE Review
15:31:50 <TheJulia> We have on item to review today. A contributor has proposed improving publish_image to be friendlier to non-openstack environments. Or at least, non-pure openstack environments.
15:31:55 <dtantsur> oh wow
15:32:01 <TheJulia> #link https://storyboard.openstack.org/#!/story/2008671
15:32:07 <TheJulia> the tl;dr is to enable support for S3
15:32:46 <dtantsur> The idea doesn't sound crazy to me, although the implementation probably will be a bit crazy :)
15:33:02 <TheJulia> Which, to me, actually makes a lot of sense because they are absolutely correct, to use swift, or even the swift model APIs, it is all predicated on using keystone as the authentication source, so it becomes harder to integrate
15:33:07 <dtantsur> I suspect $HOME won't work, everything else sounds doable
15:34:15 <TheJulia> eh, ENV['HOME'] ;)
15:34:26 <TheJulia> But yeah
15:34:33 <TheJulia> I was kind of thinking the same
15:34:49 <TheJulia> Any objections or concerns out there?
15:35:01 <dtantsur> I put a question to the RFE, but it's not really blocking
15:35:33 <rpittau> not really, I'm very curious to see the implementation :)
15:35:54 <dtantsur> yeah. only virtual media seems in scope, I wonder if everything else of interest
15:36:00 <dtantsur> (I've asked on the RFE)
15:36:19 <TheJulia> It would be nice if it introduces a clean shim that can be leveraged elsewhere
15:36:26 <TheJulia> Anyway, I guess that comes down to what does the code look like
15:37:48 <dtantsur> yep, similar to how swift is handled
15:37:59 <TheJulia> Anyway, I left a comment as well
15:38:04 <ajya> agree, could be used elsewhere too :)
15:38:10 <TheJulia> ++
15:38:23 <TheJulia> Anyway, I guess that means we're done with RFE review :)
15:38:40 <TheJulia> #topic Open Discussion
15:39:20 <sdanni> Hi folks! lmcgann and I plan to submit a new change on security interface spec later today. Please take a look if you have time!
15:39:31 <TheJulia> Any one have anything they would like to discuss?
15:39:33 <dtantsur> at a risk of getting shouted at: what would folks say about explicit support for CoreOS?
15:39:44 <dtantsur> in a sense that our network_data procedure does not work for it
15:39:45 <TheJulia> sdanni: greetings! I can try to take a look
15:40:01 <TheJulia> dtantsur: like an explict api level field?
15:40:10 <TheJulia> on the provision call?
15:40:15 <dtantsur> More like a different way to inject network_data
15:40:36 <dtantsur> this cpio madness https://github.com/coreos/coreos-installer/blob/master/docs/iso-embed-ignition.md
15:40:39 <TheJulia> I guess it would really depend on the what and how.
15:40:48 <dtantsur> yeah, I haven't really thought it through
15:40:51 <sdanni> TheJulia: thanks!
15:41:00 <dtantsur> more like checking how much non Red Hat people are opposed to the very idea
15:41:27 <TheJulia> dtantsur: wow
15:41:32 <rpittau> interesting
15:42:47 <dtantsur> unrelated: we're receiving a pressure with releases: https://review.opendev.org/c/openstack/releases/+/783665
15:43:13 <dtantsur> I think bifrost and inspector should be good to go
15:43:21 <rpittau> that reminds me
15:43:41 <rpittau> there are 2 quick bifrost patches
15:43:48 <TheJulia> I'm really starting to actually feel a great dislike of the release team's behavior
15:43:57 <rpittau> https://review.opendev.org/c/openstack/bifrost/+/782775 https://review.opendev.org/c/openstack/bifrost/+/782700
15:45:08 <TheJulia> dtantsur: I guess such explicit support for ignition would be okay for basic config injection
15:45:26 <openstackgerrit> Merged openstack/ironic master: Move configuration mold utilities  https://review.opendev.org/c/openstack/ironic/+/783630
15:46:35 <TheJulia> what/how that would look like, I don't really know off-hand
15:48:34 <dtantsur> me neither. I'm not sure it's even doable without hardcoding too much coreos knowledge
15:48:47 <dtantsur> maybe it's time for a downstream deploy interface
15:49:19 <TheJulia> or a coreos vmedia specific deploy interface
15:49:57 <dtantsur> yeah, I guess my question boils down to whether something like CoreOSDeploy is acceptable by the community
15:50:11 <dtantsur> (and note that I'll accept no as an answer)
15:51:11 <TheJulia> dtantsur: can that discussion wait until the PTG?
15:51:19 <rloo> it seems very specific to coreos. would it be useful for folks not at redhat?
15:51:22 <TheJulia> the "is it acceptable"
15:51:39 <dtantsur> rloo: there seem to be some people using Fedora CoreOS. I've seen one of them in the wild :)
15:51:52 <rloo> or email. does ptg have an operators track?
15:51:56 <dtantsur> but yes, it's very specific, and yes, most of people interested are related to Red Hat
15:52:23 <rloo> seems a bit similar to the anaconda driver?
15:52:29 <dtantsur> re PTG: the schedule is already packed with things that are arguably a higher priority for *upstream* (even if not for me)
15:52:44 <dtantsur> well.. yes. Anaconda is a precedent of adding support for a specific technology.
15:52:51 <rloo> "similar" -- wrt who might be interested in it
15:53:12 <TheJulia> You know, it could be this is a thing people wnat to use and support in a couple years generally
15:53:29 <dtantsur> CoreOS is very far from anaconda. It's a whole world of its own.
15:53:36 <TheJulia> I'm at least mentally thinking devil's advocate with that though
15:53:42 <rloo> maybe send out email then. although dunno what kind/if any response.
15:54:05 <dtantsur> it could be a part of the Forum if/when it happens
15:54:19 <rloo> similar wrt useful for red-hat related folks.
15:54:20 <dtantsur> I don't think we have enough operators reading the ML :(
15:54:41 <TheJulia> I do like the send out an email, and from there we could ping the operator's group on twitter and see if they can echo that out
15:55:13 <dtantsur> I can do that
15:55:35 <TheJulia> That seems like the "easy" feedback path as well
15:56:45 <TheJulia> It occcurs to me if maybe a coreos ignition deploy interface could possibly also pass the additional parameters needed for early hardware initialization , but I suspect just networking is the bare minimum needed at the super low level.
15:58:18 <TheJulia> Anyway, I guess that is it for the week
15:58:55 <dtantsur> yep. I've got some food for thought, thank you!
15:59:49 <TheJulia> \o/
15:59:55 <TheJulia> Thanks everyone! Have a wonderful week!
16:00:01 <TheJulia> #endmeeting