Monday, 2026-03-16

blake_good evening :) anyone aware that tarballs.opendev.org is down?02:38
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Suppress misleading ERROR logs from OVN port lookups  https://review.opendev.org/c/openstack/networking-baremetal/+/98054604:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix L2VNI trunk port binding for switch config  https://review.opendev.org/c/openstack/networking-baremetal/+/98054704:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add OVN event-driven reconcile for L2VNI trunks  https://review.opendev.org/c/openstack/networking-baremetal/+/98054804:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add targeted single-VLAN reconcile for OVN events  https://review.opendev.org/c/openstack/networking-baremetal/+/98054904:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add VNI to L2VNI trunk subport binding profiles  https://review.opendev.org/c/openstack/networking-baremetal/+/98057804:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Suppress misleading ERROR logs from OVN port lookups  https://review.opendev.org/c/openstack/networking-baremetal/+/98061204:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix incorrect is_partial_segment() call on PortContext  https://review.opendev.org/c/openstack/networking-baremetal/+/98064004:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix router interface HA binding for baremetal VLAN networks  https://review.opendev.org/c/openstack/networking-baremetal/+/98064804:35
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: L2VNI multi-link LAG support + fix OVN LLDP filter  https://review.opendev.org/c/openstack/networking-baremetal/+/98061204:39
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix incorrect is_partial_segment() call on PortContext  https://review.opendev.org/c/openstack/networking-baremetal/+/98064004:39
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix router interface HA binding for baremetal VLAN networks  https://review.opendev.org/c/openstack/networking-baremetal/+/98064804:39
rpittaugood morning ironic! o/07:46
opendevreviewMerged openstack/networking-baremetal master: Fix hash ring broken after eventlet removal  https://review.opendev.org/c/openstack/networking-baremetal/+/98054308:21
opendevreviewMerged openstack/networking-baremetal master: fix: correct client ovs/ovn idl object creation  https://review.opendev.org/c/openstack/networking-baremetal/+/98032408:21
opendevreviewMerged openstack/networking-baremetal master: fix neutron port reconcilator  https://review.opendev.org/c/openstack/networking-baremetal/+/98043708:21
opendevreviewMerged openstack/networking-baremetal master: Use other_config for bridge-mapping lookup (OVN)  https://review.opendev.org/c/openstack/networking-baremetal/+/98037408:21
*** jroll03 is now known as jroll008:38
opendevreviewMerged openstack/networking-baremetal master: Allow hostname in l2vni_network_nodes.yaml config  https://review.opendev.org/c/openstack/networking-baremetal/+/98037508:57
opendevreviewJacob Anders proposed openstack/sushy master: Revert "Handle missing RelatedProperties in TransferProtocolType detection"  https://review.opendev.org/c/openstack/sushy/+/98069712:13
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Suppress misleading ERROR logs from OVN port lookups  https://review.opendev.org/c/openstack/networking-baremetal/+/98054613:10
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix L2VNI trunk port binding for switch config  https://review.opendev.org/c/openstack/networking-baremetal/+/98054713:13
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add OVN event-driven reconcile for L2VNI trunks  https://review.opendev.org/c/openstack/networking-baremetal/+/98054813:14
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add targeted single-VLAN reconcile for OVN events  https://review.opendev.org/c/openstack/networking-baremetal/+/98054913:23
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add VNI to L2VNI trunk subport binding profiles  https://review.opendev.org/c/openstack/networking-baremetal/+/98057813:24
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: L2VNI multi-link LAG support + fix OVN LLDP filter  https://review.opendev.org/c/openstack/networking-baremetal/+/98061213:25
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix incorrect is_partial_segment() call on PortContext  https://review.opendev.org/c/openstack/networking-baremetal/+/98064013:26
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add targeted single-VLAN reconcile for OVN events  https://review.opendev.org/c/openstack/networking-baremetal/+/98054913:56
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add VNI to L2VNI trunk subport binding profiles  https://review.opendev.org/c/openstack/networking-baremetal/+/98057813:56
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: L2VNI multi-link LAG support + fix OVN LLDP filter  https://review.opendev.org/c/openstack/networking-baremetal/+/98061213:56
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix incorrect is_partial_segment() call on PortContext  https://review.opendev.org/c/openstack/networking-baremetal/+/98064013:56
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix router interface HA binding for baremetal VLAN networks  https://review.opendev.org/c/openstack/networking-baremetal/+/98064813:56
cardoerpittau: morning!13:59
rpittauhey cardoe :)13:59
TheJuliagood morning13:59
* TheJulia corrects her calendar due to daylight savings time14:00
dtantsuro/14:01
dtantsurcardoe: hi, do I recall it right that you wanted to move sushy in-tree very soon? (it's relevant for our async redfish work)14:01
cardoeI was thinking right after 2026.1 branching.14:02
cardoeWe pitched it out there and discussed it last PTG. I weighed in why I was -1. But folks gave some guarantees around my concerns. So I think we should proceed since the overwhelming majority was in favor.14:03
cardoeI just think leaving ambiguity in the change is going to be more problematic and lead to possible duplication of work.14:04
cardoeSo I figure giving us a full release cycle to hammer out the rough edges would be best.14:04
cardoeWhat do you think about that approach?14:04
TheJuliaCould someone take a quick look at https://review.opendev.org/c/openstack/ironic/+/978602/7, its small14:05
cardoeTheJulia: the commit message is wrong if we're okay with that?14:12
cardoeIt says it does something different than the change does14:12
TheJuliawe can revise it directly if we want14:13
cardoeI gave it a +214:13
cardoeBut didn't +W if ya wanna fix or whatever14:13
TheJulialets get JayF to ack the issue and make the call then14:14
TheJuliasince it might be something forgotten in the ide14:14
cardoehjensas: https://review.opendev.org/c/openstack/ironic/+/633729 is that something that should be revisited or abandoned? I know we landed some events API stuff already so not sure if there's functionality you've got in your change that we should get on top of what landed or what.14:15
cardoehjensas: same ask here https://review.opendev.org/c/openstack/ironic/+/63784014:16
cardoehjensas: and here https://review.opendev.org/c/openstack/ironic/+/63784114:16
TheJuliaQuick question: Since we're relatively AI tool friendly here, would folks be interested in joining a governing board session during the PTG to discuss impact/perceptions ?14:17
dtantsurcardoe: yeah, having a full cycle will definitely help14:17
cardoeTheJulia: yep. I'd join14:17
dtantsurTheJulia: i'm only relatively AI-freindly ;)14:17
TheJuliaWell, I was more saying friendly from the embrace/recognize/possibly utilize14:18
cardoeTheJulia: I think JayF is right in the release note for https://review.opendev.org/c/openstack/ironic/+/959880 but are we cool with the output to users changing? Ultimately we'll be able to get people detailed messages once the API is covered.14:18
TheJuliastandpoint, opinions either way ;)14:18
JayFI think we have to be cool with that output changing if we want to improve this at all, I just think we need to billboard it with a release though14:18
JayF** Note14:19
JayFAnd we've also already had that effect on endpoints already migrated and just didn't realize it in review14:19
TheJuliacardoe: I think thats for posting *to* ironic, afaik its not covered in the client and I'd never expect a human to touch events14:19
JayFTheJulia: I would suggest that whoever is doing that research sit in on our internal team use of AI session as well14:19
TheJuliaJayF: It was already noted we would have one on the board ML14:20
cardoeSince I tend to have work calls overlapping the ironic meeting I'm gonna toss this out here with the hopes you folks can discuss during the call and I'm okay with the answer being discuss during PTG as well... my team is using the inspection rules tester I wrote plus looking at the TBN simulator. We utilize GitHub Actions so we were going to look at putting together some reusable actions for using the two testers. Do we 14:23
cardoewant that contributed to the Ironic repo or another OpenStack/OpenDev repo or just something my org owns that we'll make public?14:23
TheJuliadefine the scope where it needs a repo?14:26
cardoeIt's just a script and a YAML description14:28
cardoeIt's gotta live in a repo that gets pushed to GitHub, which the Ironic repo does.14:28
cardoeNot sure how familiar with GitHub Actions you are but:       - uses: actions/checkout@v6 for example checks out the code in the current repo using https://github.com/actions/checkout from the "v6" tag.14:29
cardoeWe're gonna write "rackerlabs/ironic/inspection-rules-tester" and "rackerlabs/ironic/tbn-sim" for example.14:30
JayFcardoe: Id just make sure it's open source in a public repo somewhere. And then you start writing the talk for the summit.14:32
hjensas@cardoe: oh, the neutron event handling, so I guess the main driver is https://opendev.org/openstack/ironic/src/branch/master/ironic/dhcp/neutron.py#L150-L15814:32
cardoeYeah it'd be open source for sure.14:33
hjensascardoe: we get the events now, but don't actually use them. It might be worth picking up.14:33
cardoehjensas: yep. Just was looking at that when reviewing your networking-baremetal changes.14:33
cardoeJust trying to understand where all that sits.14:34
cardoeWas just thinking for the upcoming PTG we could talk about picking it up14:35
hjensascardoe: but the events stuff I have in proposed patches are using OVN events - the goal there is to quickly set up the trunk and ha router alignment. So that network create <- deploy node can work if there is no human that will take 10 minutes for coffebreak between net create and deploying a node.14:35
hjensass/proposer patches/proposed networking-baremtal patches/14:38
cardoeYep. It's totally different event stuff but it just made me think of it while reviewing.14:45
TheJuliaI would love to move forward on Deferred Tasks and have the event from neutron be that deferred task14:46
cardoeYeah that'd be a good usage.14:49
cardoeMother Nature is mad. It was 78F/25C yesterday. Today it's 38F/3C with a feels like 22F/-5C. It's also snowing hard to be nearly a whiteout.14:54
JayFso on friday, wake up at 5:30am, it's raining as usual14:54
TheJuliamother nature likely has decided we all got to go14:55
JayFmy wife leaves for work around 6:30am, calls me 5 minutes later "it's pouring snow" and sure enough, at home it switched to just POURING snow14:55
JayFand we got like 4-5 inches just midday outta nowhere when they were calling for flurries at best14:55
JayFnever even got below freezing either14:55
TheJuliaJayF: sounds reminiscent when NC got hit when we were in high school14:55
JayFI didn't live in Raleigh in HS :) but I remember an incident in the late 2000s14:56
TheJuliaI think your area got 2-3 feet, my area got 5-6 inches in like 1-2 hours14:56
TheJuliaearly '0114:56
JayFwhere it was calling for a snowstorm all weekend, it started like 16 hours early and faster than they called for14:56
JayFand there was this photo which has been fully memified of i-40 looking like a zombie apocalypse14:56
TheJuliaoh, that one14:56
JayFhttps://www.ncrabbithole.com/p/raleigh-nc-greatest-snow-picture-winter-weather this one 14:57
JayFalthough that's glenwood ave, it's still hilarious14:57
TheJuliaYeah, the meme photo was 14. I remember it quite well.14:58
TheJuliaHad to jerry rig my heat because the power was out for 8... days.14:58
JayFI had a power outage in rural NC around Spring Hope during a winter storm once that lasted like 10 days14:59
JayFwe were one of the only houses with a fireplace so literally half the neighborhood was sleeping in our den14:59
TheJuliaHas anyone scrubbed the release notes?14:59
TheJuliaand updated release mappings?15:00
cardoeI think no to both.15:00
TheJulia#startmeeting ironic15:00
opendevmeetMeeting started Mon Mar 16 15:00:43 2026 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.15:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:00
opendevmeetThe meeting name has been set to 'ironic'15:00
TheJuliaGood morning everyone and Welcome to the weekly #ironic meeting15:00
TheJuliaI'm TheJulia, and I'll be your host for this roast of the meeting agenda.15:01
cardoeJayF: what say you on https://review.opendev.org/c/openstack/ironic/+/978602 do we need to update the commit msg or no15:01
TheJuliaOur first item on the agenda is to ensure everyone has Coffee. Everyone have coffee?15:01
TheJuliao/15:01
JayFcardoe: meh? I can do it if folks insist, but the work is done by that commit in the one behaind it, just let it land15:01
JayFo/ 15:01
* JayF substitutes a can of diet pepsi15:02
opendevreviewVerification of a change to openstack/ironic master failed: Enable ramdisk autodetect by default  https://review.opendev.org/c/openstack/ironic/+/97860215:02
opendevreviewVerification of a change to openstack/ironic master failed: Correct disk-format in ramdisk boot docs  https://review.opendev.org/c/openstack/ironic/+/98033315:02
rpittauo/15:02
TheJuliaFair enough!15:02
TheJulia#topic Announcements / Reminders15:02
TheJuliaOur first reminder of the week is for cores to review ironic-week-prio hashtaged items.15:02
TheJulia#link https://tinyurl.com/ironic-weekly-prio-dash15:03
TheJuliaOur next item if the 2026.1 release cycle. This week is R-215:03
TheJulia#link https://releases.openstack.org/gazpacho/schedule.html15:03
TheJuliaIn other word: Time to get our end of cycle releases branched15:03
kubajjo/15:03
rpittauwe need to wrap up things, the branch creation will happen this week15:03
TheJuliaOutside of release note cleanup and the release mappings update, what is a MUST ?15:04
kubajjJayF: ok, valid15:04
rpittauthere are still some patches in ironic-week-prio15:05
TheJuliaWe can't hold forever15:05
TheJuliaSome are bug fixes and can be backported15:05
rpittauyep, besides thos4e15:06
TheJuliaI guess what I'm asking for is for us to be specific.15:06
JayFhttps://review.opendev.org/c/openstack/ironic/+/978602 is approved and I'd /like/ it in, but it's not a must15:06
rpittauI will work on release notes cleanup and release mappings update to have all ready by Wednesday15:06
opendevreviewJay Faulkner proposed openstack/ironic master: Enable ramdisk autodetect by default  https://review.opendev.org/c/openstack/ironic/+/97860215:06
TheJuliaI'd ask for us to try and be ready to release tomorrow instead of wednesday15:07
JayF(it needed a rebase to merge, I think)15:07
TheJuliathat way we can take a bump in the road at the last minute15:07
rpittauTheJulia: I can probably then do one of the 2 things, not both :)15:07
TheJuliaThen lets split the work15:08
TheJuliaWho can take on one of the tasks?15:08
rpittauI've kind of started the release mappings update already, anyone up for the release notes cleanup ?15:08
JayFrelease mappings cleanup is just a ^c^v, swap in a version, update tests?15:08
JayFCan we be specific about what cleanup is?15:09
JayFfor release notes?15:09
JayF- add a preamble (whatever it's called)15:09
TheJuliaScrub/cleanup any inconsistencies, and a preamble15:09
JayF- read the rendered version and fix text/grammar if needed15:09
TheJulia++15:09
JayFI didn't wanna take on something text-based today, I'm worried I don't have the juice for it15:10
JayFbut I guess it's me15:10
TheJuliaI can give it a shot15:10
TheJuliaMy brain is likely on the wordy side today15:10
JayFmaybe split it? I'll write the preamble if you wanna do the edits?15:10
TheJuliaAs I have some long emails to aslo reply to as well15:10
TheJuliaSure15:10
JayFit's mainly the proofreading I'm afraid of, I've been sleeping badly and proofreading requires focus I'm not sure I have today tbh15:11
TheJuliafair enough15:11
TheJuliasounds like we have a plan!15:11
TheJuliaDoes anyone have anything else to announce or remind us of?15:11
TheJuliaI'm taking silence as nope, and then onward!15:12
TheJulia#topic Working Group Updates15:13
opendevreviewMerged openstack/python-ironicclient master: Remove dead code: define_command, define_commands_from_module, HelpFormatter  https://review.opendev.org/c/openstack/python-ironicclient/+/98057915:13
TheJuliaAny updates regarding async io aside from the spec?15:13
dtantsurNot really. I'll check your comments when I have a free minute.15:13
TheJuliaOkay, Does it need to become a PTG topic?15:13
dtantsurOtherwise, it slips into the next cycle for me, and I'll also keep cardoe's work to move sushy in-tree in mind15:13
TheJuliaI guess that is a bigger question which doesn't need an immediate answer15:13
dtantsurWell.. maybe. At least as a status check. Maybe together with the sushy future.15:14
TheJuliabut we should likely just sort of have a discussion around those topics15:14
TheJuliamakes a lot of sense, could you make sure something is on the etherpad?15:14
* cid o/15:14
TheJuliaThe next working group is the vxlan working group. By in large, but fixing and rough edge grinding is the order of the week.15:15
TheJuliaAnyone have anything else to add?15:15
JayFI do15:16
JayFI would suggest, as I have, that folks communicate to stakeholders that we may need more bugfix time next cycle15:16
JayFjust as an expectation set, given the quantity of new stuff we added last cycle and how much of that stuff touches hardware15:16
JayF(especially since I suspect "ironic can deliver features" expectations are at an alltime high \o/)15:17
opendevreviewMerged openstack/ironic master: Update TBN docs for group_and_attach_ports  https://review.opendev.org/c/openstack/ironic/+/98037615:17
TheJuliaYeah, I think AI tooling has helped, and hurt us here.15:17
TheJuliaAlso, some of the areas we're touching really requires a dramatically deeper level of testing because its not "out of the box" in the slightest15:18
JayFyep15:18
JayFwhich means some of the "bugfixes" might include improving docs15:18
JayFor support on the list as folks implement this stuff15:18
TheJuliaWhat might work just fine for someone, also may have a sharp edge for another15:18
TheJuliayup15:19
JayFI just wanna make sure we're communicating up to our management chains that we may have more work to do :D 15:19
cardoeyeah this cycle is gonna be a lot of "digest the changes" for me.15:19
cardoeLike another PTG item I just thought of might be to add a "merge" mode for inspection.15:20
TheJuliaYeah, I need to get with hjensas and ensure we're creating items downstream for our recent bugfixes just so we're tracking the time/effort it is taking15:20
TheJuliacardoe: That is a good way to put it15:20
TheJulia"digest the changes"15:20
hjensasDo we want to drop the exsting periodic HA Alignment before releae? I found it is actually a no-op.15:20
TheJuliaAnyway, Discussion Topics!15:20
JayFdigest the changes, finish type checking on ironic repos, plan improvements for nova-compute story15:21
JayFnobody is allowed to talk me into more tasks at ptg ;) 15:21
TheJuliahjensas: At this point, it can be a backported bugfix. We knew there was that risk upfront15:21
TheJulia#topic Discussion Topics15:21
TheJuliaFirst and only item, PTG!15:21
TheJulia#link https://etherpad.opendev.org/p/ironic-ptg-2026.215:21
TheJuliaWere building up quite a few topics, I'd encourage everyone to take a little time this week and add any items and thoughts to existing items15:22
TheJuliaHave we reserved a "room" and times yet?15:23
JayFIsn't one of the DPL roles assigned to that?15:24
TheJuliaIt appears that we have not.15:24
TheJuliaYes, although they are not in channel at present.15:24
JayFDoes someone want to volunteer to help schedule PTG session times and register them?15:26
TheJuliaWell, we likely need to start to get that sorted if we want a good and consistent schedule. Some teams have already started reserving rooms.15:26
JayFIf it's somethign you've not done before, I'm willing to show you what it takes15:26
opendevreviewVerification of a change to openstack/ironic master failed: Fix bug in TBN NoMatch.__str__  https://review.opendev.org/c/openstack/ironic/+/98047715:26
JayFit's fun to whitewash a fence15:26
* TheJulia updates the 'twain reference counter for the day.15:27
cidSounds like something I can likely be able to do with a guide. 15:27
JayF++ thank you for the volunteer15:28
JayFwe can talk it out in here after meetings15:28
TheJuliaNext, you'll have us sitting around sipping mint juleps and negotiating while on a river boat ;)15:28
cid+++15:28
* TheJulia updates the ST:Lower Decks reference counter for the day.15:28
JayFI was expecting the ST/Twain crossover to be more TNG-coded15:28
TheJuliaWell, Lower Decks, they have to hit the entire spectrum of things.15:29
TheJuliaOkay, onward?15:29
TheJulia#topic Bug Deputy Updates15:30
TheJuliacid, you were our deputy, what chaos do you have to report! :)15:30
cidSaw a lot of new bugs... :D15:30
mumesan[m]lol15:30
cidMostly against the networking baremetal repo15:30
TheJuliaYeah, we knew this was going to happen on some level15:31
cidYup. Out of the 10 total, 7 against NB15:31
TheJuliaI think it would be helpful to try and have a high bandwidth call regarding that stuff today. hjensas would you have a time which could work?15:31
TheJuliaI think we're going to just have to work through them, although I think patches are up for a number of them already15:32
hjensasSure, we can do that - I am free most of the evening.15:32
TheJuliahjensas: ack, okay15:32
hjensasMost new bugs there are fixed in my proposed NB changes.15:32
cidI saw that. Most are just awaiting reviews15:33
TheJuliaCool, who would like to volunteer to be the bug deputy for this week?15:33
MahnoorI could give it a go15:34
JayF\o/15:34
TheJulia\o/15:34
TheJulia#info Mahnoor has volunteered to be the bug deputy for the week!15:34
TheJuliaSince have no RFE's to review, onward to Open Discussion15:34
TheJulia#topic Open Discussion15:34
cidHey, Mahnoor, been a minute :-) tks15:34
MahnoorIt has, hello hello :)15:35
cid\o/15:35
TheJuliaSo! What crazy topics of discussion do we have for today?15:35
TheJuliahjensas: would you be good to chat in say, 2 hours?15:36
TheJuliaOr 2.5 hours?15:36
JayFWe should mention the summit Call for Papers is open15:37
JayFif you wanna go hang out in Shanghai and talk about Ironic please submit a talk :)15:37
hjensasTheJulia: yes 2 hours or 2.5 hours. 15:38
TheJuliaOkay, I'll send you a meeting invite and I'll past the meet link in irc here so folks can join in15:39
TheJuliaJayF: Great call out!15:39
mumesan[m]Can you send me an invite as well please 😅15:40
TheJuliasure!15:40
TheJuliaso in just under 2 hours, we'll discuss on this following gmeet link; https://meet.google.com/zqs-irew-agb?authuser=0&hs=12215:43
TheJuliaInvite sent, if anyone else would like an invite, please let me know15:43
TheJuliaThanks15:43
TheJuliaAnything else to discuss today before we move on to figuring out who we're going to make run the next meeting?15:43
* TheJulia takes silence as a no15:46
TheJulia#topic Who will run our next meeting?15:46
TheJuliaVolunteers?15:46
JayFOK, my turn again I guess :)15:46
TheJuliaThank you JayF for volunteering to run next week's meeting.15:47
* JayF has a reminder set so he'll be here to run it15:47
TheJulia\o/15:47
TheJuliaIs there anything else before we rap this week's meeting?15:48
cidDoesn't look like it.15:49
TheJuliaThank folks, have a good week!15:49
JayF\o15:50
TheJulia#endmeeting15:50
opendevmeetMeeting ended Mon Mar 16 15:50:42 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:50
opendevmeetMinutes:        https://meetings.opendev.org/meetings/ironic/2026/ironic.2026-03-16-15.00.html15:50
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/ironic/2026/ironic.2026-03-16-15.00.txt15:50
opendevmeetLog:            https://meetings.opendev.org/meetings/ironic/2026/ironic.2026-03-16-15.00.log.html15:50
JayFcid: so basically there's a few steps to doing the room bookings15:52
JayFcid: first of all, I always would check the times we booked last cycle and use that as a base template15:52
JayFcid: then you can reserve that time using the ptgbot in #openinfra-events (ptg.opendev.org has a howto for using the bot)15:52
JayFcid: then send an email to the list proposing those times, and asking if it works for time zones -- we can add more times that are friendly to other TZs (including yours if needed!) if folks ask15:53
zigoGuys, there's a missing __init__.py in sushy/oem/dell/resources/system/15:53
JayFcid: other thing to consider: cross-project sessions (we appear to want one with neutron/nova)15:53
zigoHow come it's only something that I see only in the Ironic namespace ! :)15:54
* zigo writes a patch...15:54
JayFI wasn't aware a __init__.py was a hard requirement in all cases?15:54
TheJuliaI think for some packagers, it is :)15:55
JayFI'm mainly curious if this caused runtime issues or just a thing you noticed15:55
cidJayF, acknowledged. tks!15:55
zigoIf there's no __init__,py, then you cannot import sushy.oem.dell.resources.system15:55
TheJuliaoh brain, where were you15:55
JayFzigo: I'm mainly trying to answer the question "if this is breaky, how did it get in the codebase in the first place"15:55
zigoWell, failure to discover tests...15:55
zigohttps://paste.opendev.org/show/b3QP9JYhVOJg7sC7gMU4/15:55
JayFcardoe: ^15:55
zigoAnyways, please give me the pleasure to make this zero char patch ! :)15:56
zigoFuny stuff, my last contrib to the sushy repo was: https://review.opendev.org/c/openstack/sushy/+/83151515:56
cardoeWhich version of Python?15:56
zigoExact same thing ...15:56
zigocardoe: Any version. Currently 3.14, but that's IMO, not relevant.15:57
cardoehrm. I don't know how I'm not hitting it.15:57
JayFthat's basically the question I'm trying to figure out cardoe :)15:58
JayFbecause I figured that was running downstream at your place15:58
JayFwe should fix it anyway, but I think finding out how zigo can find these may help us in the future to detect them15:58
cardoeIt is.15:58
* JayF hopes the answer isn't "I read the output very carefully"15:59
JayFlol15:59
cardoeI even looked and we don't have sushy_oem_dell installed15:59
opendevreviewThomas Goirand proposed openstack/sushy master: Add missing __init__.py  https://review.opendev.org/c/openstack/sushy/+/98078115:59
opendevreviewRiccardo Pittau proposed openstack/ironic master: Update release mappings for new ironic releases 34.0 and 35.0  https://review.opendev.org/c/openstack/ironic/+/98078216:00
cardoehttps://www.irccloud.com/pastebin/hFn5D9jK/16:02
cardoeEven works locally16:02
JayFthat imports dell16:02
cardoeThat's what the code zigo linked is doing.16:03
JayFnot sushy.oem.dell.resources.system16:03
cardoeInside of that its importing sushy.oem.dell.resources.system16:03
JayFah16:03
zigocardoe: Are you saying "it works on my laptop" ? :)16:03
* zigo runs and hides ...16:04
JayFI think he's saying "it works on Rackspace Cloud control plane"16:04
JayFwhich is a bit of a stronger statement16:04
JayFand sounds more like "it doesn't work on /your/ laptop" ;) 16:04
cardoeI mean I pasted it from my laptop16:04
zigolol16:04
JayFI legitimately don't care about adding the __init__.py, I'll +2 it16:04
cardoeBut I confirmed that the container we're using doesn't have sushy_oem_dell installed which would be the fallback import16:04
JayFbut finding out what's different is super duper important and useful16:04
cardoeYeah I'll 100% +2/+W adding the __init__.py I was just trying to understand16:04
JayFcardoe: lets try your test w/3.1416:05
cardoeI'll have to install it.16:05
JayFI'm trying to get there too16:06
JayF we can race lol16:06
rpittauseems to work?16:06
rpittauPython 3.14.3 (main, Feb  4 2026, 00:00:00) [GCC 15.2.1 20260123 (Red Hat 15.2.1-7)] on linux16:06
rpittauType "help", "copyright", "credits" or "license" for more information.16:06
rpittau>>> from sushy.oem.dell.resources import system16:06
rpittau>>> 16:06
zigoIn 2022, when I wrote https://review.opendev.org/c/openstack/sushy/+/831515 I didn't need Python 3.14 (that didn't exist) to trigger an issue... :P16:06
JayFcan you give more detail about your testing environment?16:07
zigoHistorically, setuptools just skiped installing .py files recursively if __init__.py is missing. Not sure these days though.16:07
JayFis this from an installed deb?16:07
zigoYeah, sbuild.16:07
JayFusing setuptools instead of pbr?16:07
zigohttps://wiki.debian.org/zigo/mysbuild16:07
zigoNo no, using pbr of course.16:07
zigoGosh, hold on the +2w16:07
JayFthe of course is not assumable :) there are downstreams that used to use !pbr for openstacky stuff16:07
zigoIt's missing other __init__.py it seems.16:07
JayFjust W-1 your own patch if you want to make it not-land16:08
opendevreviewThomas Goirand proposed openstack/sushy master: Add missing __init__.py  https://review.opendev.org/c/openstack/sushy/+/98078116:09
JayFI'm asking claude, seeing if maybe it finds something we missed16:09
JayF> if sushy uses implicit namespace packages (no __init__.py) anywhere in the chain, Debian's dh_python3 can sometimes fail to install the full tree correctly, especially when the package hierarchy is deep.16:10
JayFis what claude says16:10
JayFI'm asking it to prove it :D 16:11
TheJuliaSpeaking of claude, I spotted something it couldn't see, and that makes me feellike there is hope!16:11
cardoeThat could be it. Clearly I screwed up not including those __init__.py files.16:11
rpittauwell it should work without those though :)16:12
rpittauI blame debian :D16:12
JayFTheJulia: have you read https://www.linkedin.com/posts/jayfaulkner_ive-thought-a-lot-about-the-last-few-days-activity-7437570841652920320-y6LN ... I sorta think if we get to the point where claude "knows more" than us, we need to stop clauding and start learning :D 16:12
JayFI mean, it's bad of us to be inconsistent16:12
JayFI believe it's our bug no matter what16:12
zigorpittau: Well, blame python itself, the only thing Debian is doing is "python3 -m install" ...16:12
JayFit just might be BOTH of us have bugs16:12
zigorpittau: I don't agree it "should work". If it doesn't have an __init__.py, then it's not a module, technically speaking.16:13
JayFzigo: you mean python -m pip install (?)16:13
zigoThat's by definition from Python.16:13
zigoJayF: No, I mean what I wrote.16:13
rpittauzigo: I was definitely not serious there16:13
JayFzigo: can you link me to upstream docs on the python install module then? I'm not familiar with it. 16:13
JayFzigo: and can't find it, that's why I assumed type16:13
* TheJulia very carefully reads claude's latest output and begins to wonder if someone gave sonnet 4.5 some drugs16:13
JayF*type16:13
JayF*typo16:14
* JayF notes that omitting __init__.py is OK for downstream modules, the problems only start coming in when they are inconsistently included16:14
JayFzigo: https://github.com/pypa/setuptools/issues/1097 is suggested by claude to be a related bug, with an alternate fix. Might be good to include the nspkg.pth file as well as adding __init__.py16:15
zigoIf you ask google "what's the difference between python package and a python module" it leads to this:16:15
zigohttps://stackoverflow.com/questions/7948494/whats-the-difference-between-a-module-and-package-in-python16:15
zigowhich is 100% accurate.16:15
JayFzigo: after reading the issue it sounds like it matches up16:15
opendevreviewThomas Goirand proposed openstack/sushy master: Add missing __init__.py  https://review.opendev.org/c/openstack/sushy/+/98078116:16
dtantsurTheJulia: why not opus?16:23
TheJuliaEh, I've not really found a need for opus, but it is clear that it is acting more humanlike on parsing lots of text16:24
dtantsurHeh, I just follow the logic "the fancier - the better" :D16:27
JayFopus is like asking your really smart friend who goes way in depth on everything16:27
JayFsometimes I just want a sentence and sonnet acts more like that16:27
JayFopus has ... tied itself in knots trying to do something for me before when I'm fairly sure sonnet would've said no/reprompted. It's clear claude values "uninterrupted agent time" over accuracy, and I think it's making their product worse even while it's getting smarter16:28
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Suppress misleading ERROR logs from OVN port lookups  https://review.opendev.org/c/openstack/networking-baremetal/+/98054616:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix L2VNI trunk port binding for switch config  https://review.opendev.org/c/openstack/networking-baremetal/+/98054716:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add OVN event-driven reconcile for L2VNI trunks  https://review.opendev.org/c/openstack/networking-baremetal/+/98054816:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add targeted single-VLAN reconcile for OVN events  https://review.opendev.org/c/openstack/networking-baremetal/+/98054916:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Add VNI to L2VNI trunk subport binding profiles  https://review.opendev.org/c/openstack/networking-baremetal/+/98057816:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: L2VNI multi-link LAG support + fix OVN LLDP filter  https://review.opendev.org/c/openstack/networking-baremetal/+/98061216:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix incorrect is_partial_segment() call on PortContext  https://review.opendev.org/c/openstack/networking-baremetal/+/98064016:31
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: Fix router interface HA binding for baremetal VLAN networks  https://review.opendev.org/c/openstack/networking-baremetal/+/98064816:31
TheJuliaYea, the couple times I tried opus, similar result, got in a loopy knot and I was getting frustrated.16:35
TheJuliatbh, even sonnet has gotten there at times, but I use a model of firm feedback when it gets like that16:36
JayFthe newest idea I saw, which makes sense and has helped me some16:42
JayFis being fast to "fire" agents / sessions16:42
JayFif it gets too far off the path? ^c, reprompt with more explicit direction16:42
opendevreviewJulia Kreger proposed openstack/ironic master: docs: revise release notes for 2026.1 release  https://review.opendev.org/c/openstack/ironic/+/98078616:47
TheJuliaYeah, I have found that *generally* feedback as you go better re-enforces the session, but depending on how you access also changes the pattern entirely16:54
opendevreviewHarald Jensås proposed openstack/networking-generic-switch master: Fix Cisco NX-OS VNI cleanup check command  https://review.opendev.org/c/openstack/networking-generic-switch/+/98059017:03
opendevreviewHarald Jensås proposed openstack/networking-generic-switch master: Configure L2VNI mappings for trunk subports  https://review.opendev.org/c/openstack/networking-generic-switch/+/98058617:04
JayFTheJulia: when I found out that basically most of the chat history sessions work by just feeding the entire history of the chat back into the llm, that started to inform how I treat those kind of failure cases. Just stop it, pull out any of the good that was in there and feed it back in17:06
JayFI almost think of it as a way to manually edit context17:06
TheJuliasort of an interesting idea actually17:26
TheJuliain about 4 minutes, vxlan issues and fun! Lets discuss vxlan issues and solutions! meet.google.com/zqs-irew-agb17:27
opendevreviewMerged openstack/ironic master: Enable ramdisk autodetect by default  https://review.opendev.org/c/openstack/ironic/+/97860217:48
opendevreviewMerged openstack/ironic master: Fix bug in TBN NoMatch.__str__  https://review.opendev.org/c/openstack/ironic/+/98047719:18
cardoecid: can we make some constants of https://review.opendev.org/c/openstack/ironic/+/980153 in a follow up? Also do we need to back port?20:46
TheJuliahttps://review.opendev.org/c/openstack/ironic/+/980786 <-- minor release note cleanups20:49
cidcardoe. Alright let me see if I can get a follow-on up in a moment.20:55
cidre' backportability. We can backport it just one release.20:55
* JayF needs to get clif to point his docs-speed-magic at releasenotes target when he's back21:20
JayFNext release will be my last one before I loop21:21
JayFwhen Hibiscus is released, I'll have contributed to every openstack letter21:22
cidyes21:35
* cid err21:36
opendevreviewcid proposed openstack/ironic master: Replace boot property literals with constants  https://review.opendev.org/c/openstack/ironic/+/98082921:36
opendevreviewMerged openstack/networking-generic-switch master: vxlan: Multicast traffic options for cisco driver  https://review.opendev.org/c/openstack/networking-generic-switch/+/97933221:51
opendevreviewJulia Kreger proposed openstack/ironic master: ci: gently increase ci callback timeouts  https://review.opendev.org/c/openstack/ironic/+/98083222:05
opendevreviewJay Faulkner proposed openstack/ironic master: Add preamble for Gazpacho  https://review.opendev.org/c/openstack/ironic/+/98083322:06
JayFI went 47 rounds with restructured text to bring you that22:06
JayF100% human work too, so don't ask about the AI header :D 22:06
cardoeLook at JayF's callused fingers as a result of that effort!22:07
TheJulialol22:08
TheJuliaJayF: one minor issue on that change which is a quick correction22:08
opendevreviewHarald Jensås proposed openstack/networking-baremetal master: docs: l2vni - Clean up admin documentation  https://review.opendev.org/c/openstack/networking-baremetal/+/98083522:08
opendevreviewVerification of a change to openstack/ironic master failed: Update release mappings for new ironic releases 34.0 and 35.0  https://review.opendev.org/c/openstack/ironic/+/98078222:21
JayFI also apparently spelled remote as remove so I'll fix that as well when I get back to a desk. Walking the dog right now.22:27
JayFI also owe you a review on your release notes work which I will get before my EOD22:27
opendevreviewJay Faulkner proposed openstack/ironic master: Add preamble for Gazpacho  https://review.opendev.org/c/openstack/ironic/+/98083322:39
JayFTheJulia: quick one for you, too https://review.opendev.org/c/openstack/ironic/+/980786/1#message-a53267bc4746a6fc6092bb21425f307c9467be0822:40
JayFTheJulia: if you are already away from pc lmk and I can edit22:41
opendevreviewJulia Kreger proposed openstack/ironic master: docs: revise release notes for 2026.1 release  https://review.opendev.org/c/openstack/ironic/+/98078622:46
TheJuliadone22:46
TheJulialol22:46
TheJuliaMy brain didn't catch remove, it just kind of went "thats awkward!"22:46
TheJuliadoh22:46
opendevreviewJay Faulkner proposed openstack/ironic master: Add preamble for Gazpacho  https://review.opendev.org/c/openstack/ironic/+/98083322:50
JayFfound another one where I accidentally a word22:50
TheJuliaheh22:52
TheJuliaJayF: you have a spelling issue in that23:57
TheJuliatox -epep823:57
opendevreviewJay Faulkner proposed openstack/ironic master: Correct disk-format in ramdisk boot docs  https://review.opendev.org/c/openstack/ironic/+/98033323:59

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!