Monday, 2025-11-10

rpittaugood morning ironic! o/07:42
opendevreviewRiccardo Pittau proposed openstack/ironic-python-agent master: Build and publish updated CS10 and debian images  https://review.opendev.org/c/openstack/ironic-python-agent/+/96651310:22
opendevreviewRiccardo Pittau proposed openstack/ironic-python-agent master: Build and publish updated CS10 and debian images  https://review.opendev.org/c/openstack/ironic-python-agent/+/96651310:38
opendevreviewRiccardo Pittau proposed openstack/ironic bugfix/31.0: Handle HTTP 400 and 409 race condition in Redfish power operations  https://review.opendev.org/c/openstack/ironic/+/96651610:54
opendevreviewRiccardo Pittau proposed openstack/ironic bugfix/30.0: Handle HTTP 400 and 409 race condition in Redfish power operations  https://review.opendev.org/c/openstack/ironic/+/96651710:57
opendevreviewnidhi proposed openstack/ironic master: Add PCIe function fields to redfish inspection  https://review.opendev.org/c/openstack/ironic/+/96317911:25
* cid Updates https://wiki.openstack.org/wiki/Meetings/Ironic with updates you might want to look at before the meeting :) 13:12
*** dking is now known as Guest3084613:12
*** Guest30846 is now known as dking13:14
rpittaudidn't we rmeove it ? https://bugs.launchpad.net/ironic/+bug/212946913:34
dkingGood morning, folks. Are there Metal3 guys available? Particularly, could somebody confirm what the latest version tag should be for the metal3 ironic image?13:58
rpittaudking: latest version tag is... latest :)14:09
rpittaudking: https://quay.io/repository/metal3-io/ironic?tab=tags14:10
dkingrpittau: Well, I do like to version tag, and somewhere back in the day, we had we had tags like "camp3-v1..."14:19
cardoehello Ironic. As folks enjoy they're Monday before they get too swamped with the week's things... have a look over at https://review.opendev.org/q/hashtag:%22ironic-week-prio%22+status:open even if you're not a core member, you can help your own changes land by reviewing other changes. Your reviews are helpful for us all.14:19
rpittaudking: you're talking about 2 years ago :D14:20
dkingrpittau: Yeah, that was a while ago. I'm just trying to sanity check. I'm asking because we're having some trouble with BMO re-populating Ironic. I'm pretty sure they're talking because Ironic logs have complaints about the UUIDs missing (because nothing is there), and I'm trying to work backwards. The last time this happened, it was a version mismatch. 14:24
rpittaudking: I see, can this maybe help? https://book.metal3.io/version_support14:26
dkingrpittau: That's very helpful. Thank you!. We're pretty close. I'll add that our local documentation.14:33
opendevreviewcid proposed openstack/ironic master: Filter null NIC firmware versions from cache  https://review.opendev.org/c/openstack/ironic/+/96656714:33
dkingrpittau: Would you have thoughts on where to start troubleshooting BMO not repopulating Ironic? 14:33
dtantsurdking: I assume you've already checked BMO logs?14:36
dtantsurif you're using BMO from the main branch, we've recently dropped support for really old ironics and ironic-inspector14:37
dkingdtantsur: Yeah, I'm looking there. It's complaining a lot about "could not get node for BIOS settings: host not registered"14:38
dkingdtantsur: ...but that could perhaps just be a symptom? There's a few other errors, too, and it looks like it had been in crashloobackoff.14:40
dking"failed to wait for hostfirmwarecomponents caches to sync kind source: *v1alpha1.HostFirmwareComponents: timed out waiting for cache to be synced for Kind *v1alpha1.HostFirmwareComponents"14:42
opendevreviewMerged openstack/ironic-specs master: Add 2026.1 workitems to index  https://review.opendev.org/c/openstack/ironic-specs/+/96632214:44
cidrpittau, re... PostgreSQL. Yup. It's been removed. I couldn't mark as "Won't fix" without a concrete reason. I remember the CI jobs for Postgres were removed back in 2024.14:54
TheJuliaGood morning15:00
jandershey TheJulia o/15:01
dtantsurdking: none of these immediately ring any bells..15:01
dtantsurWho's our chair this time around?15:02
dtantsur#startmeeting ironic15:04
opendevmeetMeeting started Mon Nov 10 15:04:31 2025 UTC and is due to finish in 60 minutes.  The chair is dtantsur. Information about MeetBot at http://wiki.debian.org/MeetBot.15:04
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:04
opendevmeetThe meeting name has been set to 'ironic'15:04
janderso/15:04
dtantsurWell, hello everyone, I'm not even sure we have the meeting today, but I'm going to chair it!15:04
TheJuliao/15:04
dtantsurWho is here for some ironic conversation?15:04
TheJulia\o15:05
alegacyo/15:05
* TheJulia is dancing in the corner15:05
dtantsurOur agenda is where it usually is:15:05
dtantsur#link https://wiki.openstack.org/wiki/Meetings/Ironic Meeting agenda15:05
rpittauo/15:05
dtantsurI'm giving the obviously undercaffeinated community members a couple more minutes to join15:06
kubajjo/15:06
mostepha[m]\o15:06
dtantsur#topic Announcements / Reminders 15:07
cido/15:07
dtantsur#link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/FQJSYWZIUX2LJKUGTVB274K4GB3T2QPS/ there is a proposal to expand the core team, current cores please take a look15:07
TheJuliaDoes the agenda need a coffee break?15:09
dtantsur#link https://specs.openstack.org/openstack/ironic-specs/priorities/2026-1-workitems.html here are the agreed workitems for 2026.115:09
dtantsur(please bear with me, I was not prepared to chair this meeting :)15:09
dtantsurStanding 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-dash15:09
dtantsurFinally, 2026.1 Gazpacho Release Schedule https://releases.openstack.org/gazpacho/schedule.html15:10
dtantsurAnything else to announce or remind of?15:10
* dtantsur sprinkles virtual coffee in the air15:10
dtantsurI wonder how many people are still confused by the daylight changes :)15:11
dtantsuranyway15:11
dtantsur#topic Working Group Updates 15:11
dtantsurTheJulia: should we close the eventlet WG or reuse it for further work like delayed tasks?15:12
TheJuliaI just removed the line item15:12
TheJuliaI'll try to rev the delayed task spec this week, fwiw15:12
dtantsurGood, please ping me if I miss the notification (I most certainly will)15:13
dtantsuralegacy: anything on standalone networking, except for the large patch chain to review?15:13
dtantsur#link https://review.opendev.org/q/topic:%22feature/standalone-networking%22 standalone networking patches15:13
alegacyNo, no further updates.  I managed to split the large patch into smaller chunks.  I hope that's a better configuration.15:13
dtantsurThanks, appreciated!15:14
dtantsurMeanwhile, I don't have anything new on the asyncio front (dragged into downstream priorities). cid, do you have anything?15:14
cidNot at all.15:14
dtantsurokie cool15:15
dtantsur#topic Bug Deputy Updates15:15
dtantsurcid: thats you again :)15:15
cidYup :)15:15
cidThere were a total of 12 bugs, 5 of which were RFEs15:15
TheJuliaClearly bug filing occurs with caffeine15:16
TheJulia:)15:16
dtantsurheh15:16
dtantsurShould we review the RFEs, do you think?15:16
cid:) 15:16
cidI will just drop the full list here15:16
cidhttps://bugs.launchpad.net/networking-generic-switch/+bug/2130384: Treat a DPU like a switch15:16
cidhttps://bugs.launchpad.net/ironic/+bug/2130646: Enable the agent deploy interface to auto-switch to bootc15:16
cidhttps://bugs.launchpad.net/ironic/+bug/2130667: Enable Redfish interactions with network device functions15:16
cidhttps://bugs.launchpad.net/ironic/+bug/2129469: Add PostgreSQL database backend support15:16
cidhttps://bugs.launchpad.net/ironic/+bug/2129690: Make all nodes of a conductor-group write grub config15:16
dtantsurTheJulia: DPU-as-a-switch, is it spec worthy?15:16
TheJuliaThe first one, I filed, the broad idea is the ability for NGS to begin to treat a DPU like a switch15:16
TheJuliaI don't think so, I think its more just a driver15:17
rpittauI think we can jsut close the postgresql one as cid suggested before15:17
TheJuliaor drivers15:17
dtantsur(I know that we don't usually do specs for n-g-s)15:17
TheJuliaIssue is, it can only be developed by someone with their hands on a card15:17
dtantsurYeah, let's prepare a polite response for PostgreSQL. Something along the lines of "Alas, OpenStack moved on from it, and we cannot maintain another database on our own"15:17
TheJuliawhich is similar, really, to physical switcehs15:17
opendevreviewRiccardo Pittau proposed openstack/ironic-python-agent master: Test advertised ip reachability before assigning it  https://review.opendev.org/c/openstack/ironic-python-agent/+/96367015:17
dtantsurTheJulia: I assume you don't have one handy?15:18
TheJuliadtantsur: w/r/t postgres, concur15:18
TheJuliadtantsur: I don't, I might be able to grab one out of a lab personally15:18
TheJuliaits more a broad idea than anything else15:18
TheJuliasince each card will be different in the end15:18
* cid Goes to update the bug15:18
TheJuliathe second one, regarding auto-switching the device, I think stevebaker is taking a look at.15:19
dtantsurYeah, I see. I don't mind rfe-approved on this, assuming we don't need to change any data model15:19
TheJulias/device/driver/15:19
dtantsur(this was about the DPU one)15:19
TheJuliaI don't think so, I think the existing model works for that15:20
dtantsurI'm fine with the bootc one, but I know that JayF had reservations, so I'd not approve it until I hear from him15:20
* TheJulia wonders if we should have sifted through these one at at ime15:20
cardoegah time change :/15:20
TheJulias/ime/time/15:20
TheJuliaDaylight savings time is indeed, evil.15:20
TheJuliaPossibly the root of all evil.15:20
rpittauhopefully it will disappear from EU next year15:21
TheJulia(We voted it away here in california, but the US seems hellbent on ignoring our resolution)15:21
dtantsurheh15:22
cardoeWe did as well in Alabama (to my shock) but we gotta wait just like CA.15:22
TheJuliaNext up in the catecory of roots of all evil, cats seeking to cuddle with the cables behind the laptop15:22
* TheJulia can't spell today15:22
TheJuliaSo, back to RFEs15:22
dtantsurCats tend to be.. energized ;)15:23
cardoeSo I wanted to get to the changing of the deploy interface….15:23
cardoeIt feels like we need a way to change that dynamically-ish. From the nova-ironic side it could be an image property.15:23
TheJuliaThe third, network device functions, I see that as first step spend some time in sushy modeling the options and largely in sushy but it proposes a general idea of some clean steps to toggle the options around.15:23
TheJuliafor going to something like anaconda, I think that makes a lot of sense15:24
TheJuliathe same general idea is present though15:24
dtantsurNetwork device functions is cool, I wonder if you know how much of a mess NetworkInterfaces are...15:24
* dtantsur summons janders15:24
TheJuliaaiui, fairly messy, unfortuantely.15:24
cardoeHe’s in EU time zone no now?15:24
jandersyes15:25
janderswe had a chat with TheJulia in Paris about the NetworkAdapters bug I'm tackling15:25
dtantsurah, NetworkAdapters. Are these different?15:25
TheJuliaI was largely thinking we might be able to model it on a single platform15:25
TheJuliaso..15:25
TheJuliayou basically need to walk from the system -> Network Adapters -> NetworkDeviceFunctions15:25
dtantsuraha, so you may be forced to ensure that something (IPA?) is running on the machine15:26
* janders is looking at BMC mockups to see where things are tied together15:26
janders++15:26
jandersor at least have some of a fallback for pathological cases15:26
TheJuliadtantsur: based upon the model, not really. That is if vendors have tied things cleanly and/or we already have the MAC addresses15:26
dtantsurI thought the resource was not accessible when the OS is not running?15:27
TheJuliathe host might need to be "on"15:27
jandersdtantsur on some servers, yes15:27
TheJuliait really depends on the level of firmware and underlay architecture when it comes to what is done for the BMC to talk to other devices15:28
jandersso we could 1)either have a requirement to have OS running when we do work on NetworkInterfaces or 2) try and see, if failure -> boot IPA15:28
TheJuliaBut the base idea is to take the devices, model them matching the network ports, and toggle network ports out of networking funciton15:28
TheJuliafunction15:28
TheJuliaand towards a storage function15:28
TheJuliaUltimately, it will all depend on what BMCs return and the error handling which is needed there15:29
dtantsurDoes it make sense to have something on the Port API to mark it? so that the steps themselves only enforce the value?15:29
dtantsur(similarly to target_raid_config?)15:29
TheJuliaI did poke at an idrac10 which was off, and it seemed that the adapters were visible15:29
jandersI concur15:29
dtantsuryeah, the problem is with HPE machines15:30
jandersexactly15:30
TheJuliaI was largely thinking we just use the category function and toggle it to "storage" or some other preconfigurable value15:30
TheJuliaand then just dissuade from trying to use the interfaces for anything else.15:30
TheJuliawhich ties into TBN15:30
dtantsurah, right, we have many more fields now that I'm not aware of :)15:30
dtantsurokay, I personally like the idea, but I'd prefer a bit more technical details (and maybe an RFE per step)15:31
TheJuliaI'm not sure if I'll be able to start on it this year or not, still trying to determine the priorities until EoY15:31
dtantsurack15:32
TheJuliaIt is seeming like VXLAN networking is more important on my front, realistically.15:32
dtantsurThen let's table it for now and get back once you have cycles?15:32
TheJuliaI just wanted to get something recorded that is the seed of the idea first15:33
TheJuliaonward!15:33
dtantsurSo, the last is PXE HA15:34
dtantsurhttps://bugs.launchpad.net/ironic/+bug/212969015:34
dtantsurI'd *love* to have this for metal3 HA. I think you even had a draft spec some time ago TheJulia?15:34
dtantsurLike before, I doubt anyone has cycles for that15:35
TheJuliaYes, I know exactly how this would need to work15:35
TheJuliaRealistically, it would take a Conductor RPC call to the remote conductors15:36
dtantsurI'm still worried about RPC multiplication, but I guess we can get back to it on the actual spec if anyone revives it..15:36
TheJuliafor them to do the needful in an abridged form with some additional details in terms of networking15:36
TheJuliaWhere it would need to be wired in, its actually not a big deal and if we have deferred tasks then its feasible to just use that as a mechanism as well15:37
TheJuliaThe key difference is somehow the other conductors need to so some amount of work and it will need to be a shared task under the hood.15:37
dtantsuryeah, there are interesting nuances15:38
dtantsurI'd also prefer the secondary conductors to refer to images from the primary one, not download them from the internet15:38
dtantsur(which is not fully HA though, so I dunno if it fixes the issue in question)15:38
TheJuliaoh, yeah15:38
TheJuliathat was something I had baked into the previous idea15:39
TheJuliajust include a pointer to the other node15:39
TheJuliabut the broad fear of conductor triggering 1 to N additional messages over RPC was sort of the hurdle the discussion died at15:39
cardoeSo do we just need like "conductor can upload to an object storage that's not swift"?15:39
TheJuliafor a super short lived "create a pointer", thats likely not a big deal15:39
TheJuliaSo15:40
TheJuliathat goes into conductor failover theory15:40
TheJuliawhen a failure occurs, the conductor will automatically try to download artifacts15:40
dtantsurAnother approach would be for secondary conductors to download images from the primary one15:41
TheJuliathat is a greater multiplication risk, really15:41
dtantsurit is15:41
TheJuliathe original idea was just generate a pointer to load config/artifacts from the first, and should failure occur and takeover triggers then the conductor and presumably new secondaries would get calls to create new records and new pointers15:42
TheJuliaI guess15:42
dtantsuryeah, fair15:42
TheJuliathe challenge is how much of a lag are we okay with... or not.15:42
dtantsurnow someone needs to do the hardest parts: write/update the spec and the code :D15:42
opendevreviewVerification of a change to openstack/ironic master failed: Fix storing inventory and plugin data in Swift  https://review.opendev.org/c/openstack/ironic/+/96625915:42
TheJuliayeah15:44
dtantsurCan we agree that the last one definitely needs-spec?15:44
TheJuliayeah, definitely15:44
TheJuliaIts totally do-able, but requires agreement on which time/risk gap we're focused on as well15:45
* dtantsur nods15:45
dtantsurcid: anything else on bugs or rfes?15:45
TheJulia(this quickly turns into one of those "perfection is the enemy of good" things as well.15:45
dtantsuralso true15:45
* dtantsur suspects everyone else has falled asleep15:45
cidThat's all on bug deputy updates15:45
dtantsurThanks!15:46
dtantsur#topic Open discussion15:46
dtantsurThe floor is yours, crew15:46
rpittauI have one thing15:46
dtantsuryou have our attention15:46
TheJuliadtantsur: mandatory coffee break at the beginning of meetings?15:46
dtantsur++15:47
dtantsurif it was not 4pm here..15:47
rpittauwhen moving from tinyipa to DIB for integration tests in bifrost I just realized that the old GRUB can't really work well with big vmedia devices when in UEFI15:47
TheJuliadtantsur: agenda updated.15:47
rpittaumeaning that the vmedia jobs will be broken for bookworm and jammy15:47
rpittauand not feeling very well for noble15:47
TheJuliaWell, that is UEFI in general on some hardware as well.15:47
dtantsurwill it work anywhere at all?15:48
TheJuliaIs it the 512MB barrier?15:48
rpittauthey run ok on centos10 as it has a more recent version of grub15:48
rpittauit';s smaller15:48
TheJuliaSRSLY?15:48
* TheJulia waits for a YA'RLY15:48
dtantsurhmm, don't we have vmedia coverage in the devstack CI? the standalone job?15:48
rpittauanyway jsut wanted to say that I'm going to test with debian DIB15:48
rpittaubut I would also like to propose alpine as an alternative15:48
TheJuliadtantsur: centos based afaik15:48
dtantsurTheJulia: then you had to ask O'RLY?15:48
JayFAlpine will introduce new testing service to us as it uses MUSL15:49
dtantsurTheJulia: ehhmmm? we have devstack jobs on centos??15:49
JayF**surface15:49
TheJuliawell, okay, ubuntu but they boot centos15:49
JayFI don't think it'll matter but it's worth noting 15:49
rpittautrue15:49
rpittauthe alternative is to drop bookworm and ubuntu jammy entirely15:49
rpittauif debian based can't work15:50
dtantsurrpittau: hold on, does it mean that only DIB-building jobs are affected?15:50
rpittauyep15:50
rpittauI wrote that at the beginning :D15:50
dtantsuryou.. did not?15:50
dtantsurDIB building jobs don't even use vmedia, so I'm still confused15:50
rpittauI wrote "when moving from tinyipa to DIB for integration tests in bifrost"15:51
dtantsurright, but not all integration tests are affected, only dibipa ones15:51
dtantsurwhich don't use vmedia, so... wut?15:51
rpittauI probably can't explain myself15:52
rpittauI'm converting ALL the jobs to DIBIPA15:52
rpittauhttps://review.opendev.org/c/openstack/bifrost/+/96440415:52
dtantsurwait, why?15:52
rpittaubecause we don't support tinyipa anymore15:52
dtantsurwe have IPA images published, we don't need to build them in literally every job15:52
rpittauwe don't15:52
TheJuliawhy not?15:52
dtantsur(dibipa jobs build IPA using DIB, hence their name)15:52
rpittauin the integration jobs we don't build the images15:53
rpittauwe used the published ones15:53
TheJuliaSo why don't published images work?15:53
dtantsurAh, so you mean if we switch our official images to Debian, then.. oh15:53
rpittaubecause they're too big for uefi15:53
rpittauso15:53
rpittauif we switch to dib with cs9/10 they're super big >500 MB15:54
rpittauwith debian we're around 300MB I think15:54
dtantsurCS9/10 is already know to work15:54
rpittauwith recent version of grub, yes15:54
rpittauwith grub on bookworm or jammy, nope15:54
dtantsurdoes it mean that the ironic's standalone jobs still use tinyipa?15:55
opendevreviewMerged openstack/networking-generic-switch master: Drop remaining logic for linuxbridge-agent  https://review.opendev.org/c/openstack/networking-generic-switch/+/96514915:55
TheJuliaAre the ipab results still 500+ MB?15:55
rpittauI'm ok making bookworm and jammy non-voting for the time being just to switch to DIB15:55
rpittaubut I'm not sure debian based still make it work15:55
TheJuliaI thought I posted patches to remove like 90+ MB15:56
* TheJulia senses we're in the weeds and doing ourself a disservice from asynchronous discussion15:56
jandersI'd like to briefly touch on iRMC deprecation/removal before we wrap up15:56
dtantsurrpittau: could you check it? I think we might have pre-built images in the ipa-b location15:56
TheJuliarpittau: maybe some deep synchronous discussion after the meeting?15:56
dtantsurif debian IPA images is a workaround, we're fine15:56
TheJuliadtantsur: we do afaik15:57
dtantsurwe don't need a call, we need more information15:57
TheJuliait might be the IPA location is not updated yet if we've not merged anything in that repo yet15:57
rpittaudtantsur, TheJulia, I checked the images, cs9 is 490MB, cds10 is 590MB15:57
rpittauhttps://tarballs.opendev.org/openstack/ironic-python-agent-builder/dib/files/15:57
dtantsurrpittau: let's create a variation of your patch that uses https://tarballs.opendev.org/openstack/ironic-python-agent-builder/dib/files/ipa-debian-master.initramfs15:57
dtantsurand see if it passes15:57
dtantsurif yes, we're golden, just need to switch our official images15:58
dtantsurif not, we may need to actually seriously talk tomorrow15:58
rpittauyeah that's what I was going to do15:58
rpittaujust wanted to check if we're open to the alpine alternative for CI15:58
rpittaualright15:58
TheJuliahttps://review.opendev.org/c/openstack/ironic-python-agent-builder/+/965984 will help, a lot.15:58
dtantsurwe have some crazy options like using a different grub binary... but let's wait for the results15:58
TheJuliajanders: might as well :)15:58
jandersI made initial contact with Fujitsu. Most of the discussion is downstream related ( TheJulia I will loop you into this for OSP part ). The upstream-relevant part is FJ are open to deprecation/removal of iRMC from master if it remains in stable branches.15:59
jandersquestion 1) is this doable15:59
* dtantsur issues a 1 minute warning15:59
jandersquestion 2) how would they be proposing critical fixes if needed? straight to stable/whatever branch?15:59
dtantsurit's definitely doable, and we definitely won't remove code from stable branches15:59
* dtantsur doubts they'll propose any fixes now16:00
TheJulia1) yes16:00
dtantsurI've already let janders know my opinion, so please others weigh in16:00
TheJulia2) straight to the branch16:00
jandersdtantsur I agree, unless there's some realisation about CVEs16:00
dtantsurlike in ancient pysnmp? ;)16:00
jandershaha they may need to get creative16:01
jandersit's damage control at this stage anyway16:01
dtantsuranyway. does anyone object tho the approach that janders is proposing?16:01
TheJulianot at all, it is what it is16:01
janderssounds like we have way forward so I will keep pursuing that - there will be some downstream discussion followed by upstream comms once things are clear16:01
jandersthank you for your insights16:02
TheJuliaThanks!16:02
dtantsurWe need to wrap up, thank you folks!16:02
janderso/16:02
dtantsur#endmeeting16:02
opendevmeetMeeting ended Mon Nov 10 16:02:20 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:02
opendevmeetMinutes:        https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-11-10-15.04.html16:02
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-11-10-15.04.txt16:02
opendevmeetLog:            https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-11-10-15.04.log.html16:02
cid++16:02
dtantsurJayF: do we have any requests to support alpine? I thought they were more of base OS for containers...16:02
JayFdtantsur: It was mentioned as an option, I just wanted to point at musl v glibc. To be clear, any Ironic-impacting errors there wouldn't be our bugs, but I'd rather steer clear of other folks' bugs too :D 16:03
rpittaubtw I think we should grab the images from IPA not ipa-builder, hence we need https://review.opendev.org/c/openstack/ironic-python-agent/+/96651316:04
dtantsurrpittau: sure, I meant to do it as an experiment, not final solution16:04
dtantsurrpittau: I dunno about publishing CS10 images without ever testing them16:05
dtantsurwhy?16:05
rpittaummm that's true, and we're going to witch to debian anyway16:07
rpittauok, I'll remove that16:07
opendevreviewRiccardo Pittau proposed openstack/ironic-python-agent master: Build and publish updated debian images  https://review.opendev.org/c/openstack/ironic-python-agent/+/96651316:09
opendevreviewRiccardo Pittau proposed openstack/bifrost master: [WIP] Remove tinyipa support and switch to debian IPA  https://review.opendev.org/c/openstack/bifrost/+/96440416:12
rpittautesting debian ipa on bookworm ^16:13
opendevreviewMerged openstack/bifrost master: [CI] upgrade from 2025.2  https://review.opendev.org/c/openstack/bifrost/+/96628316:13
JayFI'm going to execute the approver/reviewer changes from the mailing list. This is your last few minutes of opportunity to object.16:15
opendevreviewVerification of a change to openstack/ironic master failed: Fix storing inventory and plugin data in Swift  https://review.opendev.org/c/openstack/ironic/+/96625916:26
opendevreviewMerged openstack/ironic-python-agent stable/2025.2: Fix for matching hints with lists of strings  https://review.opendev.org/c/openstack/ironic-python-agent/+/96603116:38
opendevreviewnidhi proposed openstack/sushy master: Add complete LLDP fields to Port.Ethernet.LLDPReceive per DMTF Redfish v1.12.0  https://review.opendev.org/c/openstack/sushy/+/96661616:44
opendevreviewnidhi proposed openstack/sushy master: Add complete LLDP fields to Port.Ethernet.LLDPReceive per DMTF Redfish v1.12.0  https://review.opendev.org/c/openstack/sushy/+/96661616:46
opendevreviewnidhi proposed openstack/sushy master: Add complete LLDP fields to Port.Ethernet.LLDPReceive per DMTF Redfish v1.12.0  https://review.opendev.org/c/openstack/sushy/+/96661616:47
dkingdtantsur: When setting IRONIC_IP in BMO, does that default to an HTTP or HTTPS connection?16:55
opendevreviewVerification of a change to openstack/ironic-python-agent-builder master failed: Wait up to 30 seconds for config drive  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/96267316:56
JayFGerrit auth changes have been finalized: cid cardoe should have approvers access now, kubajj has reviewer access now, mgoddard removed from approvers16:58
JayFCongratulations16:58
kubajjJayF: 🫡 thanks16:58
cid\o/. Thanks Ironic for the vote of confidence.16:59
TheJuliawoot!17:00
dtantsurdking: there is no such option in BMO, did you mean ironic-image?17:00
dtantsurthe BMO option is called IRONIC_ENDPOINT and includes the whole URI17:00
cardoeThank you all :)17:01
dkingdtantsur: I suppose that's why I couldn't find it. :) I think it might be sharing some tings in the configmap. I'm starting to grasp at straws. The Ironic endpoint is set correctly. I just found some line in the error logs that looked like it was trying to connect with the wrong handshake.17:03
dtantsurdking: the current deployment scripts in BMO are a mess. They do indeed mix options for BMO, option for Ironic and options for the CI.17:03
dkingThat probably explains that. I don't think that we're doing anything too custom here.17:05
dtantsurThis all is one of the main reasons I started ironic-standalone-operator17:06
dkingThat might be worth looking at. At the moment, though, our setup has been pretty stable for a few years, and we're not quite ready to rock the boat.17:07
dtantsurI can imagine17:07
dking... but after this latest update, something just isn't happy. Fortunately, it's not a production environment, but we'll be doing that soon.17:08
dtantsurdking: which versions do you use?17:09
dkingdtantsur: So, currently, BMO has a pod continually going into CrashLoopBackoff, and also, our first indication of a problem was that Ironic was never repopulated. I see several issues in the logs... and I just noticed that it has errors that "spec.architecture" is an unknown field.17:12
dkingdtantsur: BMO: v0.11.1, Ironic: v31.0.0 That's what we just updated to.17:13
dtantsurspec.architecture, huh? maybe you did not update CRDs properly?17:15
dtantsurspec.architecture was added in BMO 0.4, so something must be really off17:16
dkingdtantsur: I'm ashamed to say that in this non-prod environment, we are using an older version than that. So, that might be why it's not there.17:17
dkingWe were around v0.317:18
dtantsurdking: did you upgrade from 0.3 to 0.11? Could it be that you did not upgrade CRDs or something went wrong there?17:20
dkingdtantsur: I didn't do the upgrade myself, but yes, it was bumped directly from 0.3 to 0.11.1, and it doesn't look like anything further was done.17:24
dkingThese machines are currently running, so I'm hoping to not have to re-provision them if I can help it, but I do have at least one that can be completely removed as it was already ready to clean when we noticed the issue.17:24
opendevreviewClif Houck proposed openstack/ironic master: Trait Based Networking Simulator  https://review.opendev.org/c/openstack/ironic/+/96620217:25
dtantsurif only the code was bumped, it can explain the symptoms17:26
dtantsur(I'm not sure why you mention reprovisioning, BMHs themselves are fine probably)17:26
dtantsurI cannot provide specifics since I don't know how your installations and upgrades work17:27
dkingdtantsur: Okay. So, forgive my ignorance, but what is the best way to update these BMH properly after the image update?17:27
dtantsurYou don't need to update BMH17:28
dtantsurMake sure CRD (not CR!) are up-to-date. I.e. that you've applied the latest version of https://github.com/metal3-io/baremetal-operator/blob/main/config/base/crds/bases/metal3.io_baremetalhosts.yaml17:28
dtantsur(or, well, the 0.11 version)17:28
dtantsurThis is how the fields are defined, they don't come from the BMO code implicitly17:29
dkingAh, thanks. I'll see if I can get that done. 8 versions is a big jump, so we should have probably taken more care, but that's why it's in dev.17:32
dtantsurTrue. I don't remember breaking changes but better safe than sorry.17:32
JayFhttps://review.opendev.org/c/openstack/ironic/+/961498 (this is the tip of Trait Based Networking) could use a review 17:41
JayFhttps://review.opendev.org/c/openstack/ironic/+/965090 (r pittau's docs update for releasing) could use a review 17:47
TheJuliaI'm likely not going to be able to circle to reviews until much latter today, fwiw17:53
JayFSimilarly I'll be mostly gone tomorrow so trying to get a review on all the things today :)18:07
TheJuliaheh, I'm over here trying to get groundwork done so I can get reviews done later. joy!18:09
TheJuliarpittau: any thoughts on disabling or non-voting the arm64 job on ipa-b ?18:43
JayFDo we need some kind of Ironic-netowrking sync? I'm thinking about https://review.opendev.org/c/openstack/ironic-specs/+/940861/2#message-5800acbabaf63de96df0e7320f7d988d654cb67b and how with TBN, ironic-networking (standalone) and the VXLan work ongoing + maybe this tags thing, we should make sure we're not making a different decoder ring for each finger :D 18:47
JayFcc clif 18:48
JayFI'm all for figuring out things as we implement them but especially with so many moving parts it seems like an ounce of prevention could be worth a pound of cure here18:48
TheJuliawe do likely need one, just hesitant to commit at the moment18:58
JayFsame but I wouldn't hate us all making sure we're working together18:58
JayFfor instance, in that spec, I could invision it turning into "syntax sugar" to activate TBN18:58
JayFeven if we need a simpler door to scheduling, I'd rather not reimplment the world18:58
JayFespecially since I may desire a phase 3 involving hooking this up to standalone ironic-networking :D 18:59
TheJuliaOn my end it is more a issue of I have a ton of requirements and more so its indecision on which is the. most pressing19:22
opendevreviewMerged openstack/ironic master: Fix storing inventory and plugin data in Swift  https://review.opendev.org/c/openstack/ironic/+/96625919:37
opendevreviewMerged openstack/ironic stable/2025.2: Handle HTTP 400 and 409 race condition in Redfish power operations  https://review.opendev.org/c/openstack/ironic/+/96629020:14
*** milan is now known as Guest3088320:15
cardoeJayF: I think it would be good to have an how will this work end to end.21:15
JayFMy main thing is I just wanna make sure all the network folks are talking to each other21:15
JayFwe have enough parallel work happening that in 6-12 months Ironic networking could be massively improved21:15
JayFbut when that much work is going on at the same time the risk for making it worse (or more confusing... maybe more confusing that necessary) if we don't 21:16
opendevreviewMerged openstack/ironic master: fix: ensure that portgroup physical_network is updated for tests  https://review.opendev.org/c/openstack/ironic/+/96598522:02
cardoeJayF: agreed. happy to have a call at some point.22:27
opendevreviewDoug Goldstein proposed openstack/ironic master: pass along physical_network to neutron from the baremetal port  https://review.opendev.org/c/openstack/ironic/+/96457022:29
cardoeJayF: ^ like that I'd love to have a chat with how multiple VXLAN fabrics would work with scheduling.22:30
cardoeCause we've said use category which could work but we've also said use category for bonding.22:30
opendevreviewMerged openstack/ironic-python-agent master: Test advertised ip reachability before assigning it  https://review.opendev.org/c/openstack/ironic-python-agent/+/96367022:40
JayFyou should be talking to clif about this more than me22:41
JayFand I think your concerns are sorta mixing TBN + non-TBN22:41
cardoeHappy to speak with clif as well.22:41
JayF"what bonds together" + "what VXLAN fabric to attach" from an Ironic standpoint would all be inside the TBN logic22:41
JayFI'm just saying I am speaking in general he has the specifics better nailed down22:41
cardoeAgreed22:41
opendevreviewMerged openstack/ironic master: fix: local_link_connection inspection hook does not fail on missing port  https://review.opendev.org/c/openstack/ironic/+/96637322:47

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!