| opendevreview | Steve Baker proposed openstack/ironic master: OciImageService detect bootc image https://review.opendev.org/c/openstack/ironic/+/966760 | 02:33 |
|---|---|---|
| opendevreview | Steve Baker proposed openstack/ironic master: Add an autodetect deploy interface https://review.opendev.org/c/openstack/ironic/+/973187 | 02:33 |
| opendevreview | Steve Baker proposed openstack/ironic master: Replace ironic-tempest-bios-ipmi-direct with -autodetect https://review.opendev.org/c/openstack/ironic/+/974260 | 02:33 |
| opendevreview | Himanshu Roy proposed openstack/ironic master: Fast track shouldn't happen when firmware update is in progress https://review.opendev.org/c/openstack/ironic/+/974265 | 03:44 |
| rpittau | good morning ironic! o/ | 07:48 |
| bbezak | Good morning Ironic! | 09:28 |
| bbezak | Small patch to ngs - could you please take a look? https://review.opendev.org/c/openstack/networking-generic-switch/+/973461 | 09:29 |
| janders | good morning Ironic o/ | 09:32 |
| janders | cardoe ++ let's coordinate the rebase of health work, I will work on that later in the day | 09:33 |
| priteau | Good morning Ironic! Small Bifrost CI patch ready to approve: https://review.opendev.org/c/openstack/bifrost/+/973458 | 09:42 |
| abongale | good morning o/ | 09:54 |
| opendevreview | Merged openstack/ironic master: Improve image download performance https://review.opendev.org/c/openstack/ironic/+/964502 | 10:20 |
| opendevreview | Harald Jensås proposed openstack/ironic master: devstack: Fix OVN init for hardware deployments https://review.opendev.org/c/openstack/ironic/+/974303 | 10:33 |
| opendevreview | Merged openstack/ironic master: Support `disable_ramdisk` during servicing https://review.opendev.org/c/openstack/ironic/+/973016 | 10:45 |
| opendevreview | David Nwosu proposed openstack/ironic master: Fix firmware update error handling and test empty MessageId https://review.opendev.org/c/openstack/ironic/+/974307 | 11:10 |
| opendevreview | Abhishek Bongale proposed openstack/python-ironicclient master: Fix field names for selection and json https://review.opendev.org/c/openstack/python-ironicclient/+/973948 | 11:28 |
| opendevreview | Abhishek Bongale proposed openstack/python-ironicclient master: Fix field names for selection and json https://review.opendev.org/c/openstack/python-ironicclient/+/973948 | 12:41 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Refactor firmware update temp field handling https://review.opendev.org/c/openstack/ironic/+/974317 | 13:21 |
| janders | ^ https://review.opendev.org/c/openstack/ironic/+/966344 follow up | 13:21 |
| opendevreview | cid proposed openstack/ironic master: Minimum version bump https://review.opendev.org/c/openstack/ironic/+/974321 | 14:00 |
| TheJulia | good morning | 14:18 |
| cardoe | cid: can you remove the depends on from https://review.opendev.org/c/openstack/openstacksdk/+/973287 since that change doesn't depend on NGS but NGS depends on that change? | 14:43 |
| cardoe | janders: you're good to rebase it... all the conflicts have landed. you'll need to update to API 1.109 | 14:50 |
| cardoe | if someone can put some eyes on https://review.opendev.org/c/openstack/python-ironicclient/+/974097 so we can unblock the gazpacho release of the client | 14:51 |
| rpittau | if we can get another core vote for https://review.opendev.org/c/openstack/python-ironicclient/+/974097 so I can unleash the ironicclient release :) | 14:57 |
| opendevreview | Pierre Riteau proposed openstack/bifrost master: CI: Fix previous release for SLURP upgrade jobs https://review.opendev.org/c/openstack/bifrost/+/973458 | 15:00 |
| TheJulia | done | 15:37 |
| opendevreview | Merged openstack/python-ironicclient master: feat: add 'physical_network' and 'category' to portgroup object https://review.opendev.org/c/openstack/python-ironicclient/+/974097 | 15:43 |
| cardoe | rpittau: ^ good to go | 15:54 |
| rpittau | awesome | 15:54 |
| rpittau | I've updated the release patch | 15:57 |
| cardoe | I’ve given it my +1. Now we wait on the release team. | 16:02 |
| janders | cardoe thank you, on it | 16:10 |
| opendevreview | Takashi Kajinami proposed openstack/ironic master: Drop encoding magic comment https://review.opendev.org/c/openstack/ironic/+/974341 | 16:21 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 16:22 |
| opendevreview | Jacob Anders proposed openstack/python-ironicclient master: Add support for node health status field https://review.opendev.org/c/openstack/python-ironicclient/+/967055 | 16:29 |
| opendevreview | Doug Goldstein proposed openstack/ironic stable/2025.2: Fix introspection failure handling https://review.opendev.org/c/openstack/ironic/+/974069 | 16:31 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 16:33 |
| cid | cardoe, ack’d. I will quickly do that now. | 16:37 |
| janders | cardoe do https://review.opendev.org/c/openstack/ironic/+/966946 and review.opendev.org/c/openstack/python-ironicclient/+/967055 look good after rebase? I thought that code sections from your changes should be blue not green. I wonder if Claude stuffed up the rebase this time. | 16:37 |
| janders | figured I will ask before I re-attempt these in case it doesn't matter | 16:37 |
| janders | I figured let's sort this out before I start addressing the new comments on the API change | 16:39 |
| janders | upon detailed inspection I think it's OK after all | 16:45 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 16:50 |
| opendevreview | Jay Faulkner proposed openstack/ironic stable/2025.2: Improve image download performance https://review.opendev.org/c/openstack/ironic/+/974345 | 16:52 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 16:53 |
| opendevreview | Rico Lin proposed openstack/ironic master: Set columns in bios_settings table to bigint https://review.opendev.org/c/openstack/ironic/+/968348 | 16:58 |
| opendevreview | Doug Goldstein proposed openstack/ironic bugfix/33.0: fix redfish inspect system product name https://review.opendev.org/c/openstack/ironic/+/971488 | 17:02 |
| janders | cardoe looking at history, when I started working on this I was aiming at microversion 104, now we're at 109. I have a sneaking suspicion it's like 10th rebase :) | 17:07 |
| janders | I need to get more efficient at this although it was really good learning this change | 17:07 |
| cardoe | It happens. It was a big change with a bunch of revisions. | 17:08 |
| cardoe | There's another patch out there to clean up the number of calls against redfish to make it faster that conflicts with your change too. ;) | 17:08 |
| cardoe | But I was gonna ask that person to rebase on top of yours. | 17:09 |
| cardoe | I need to be working on the network stuff with Julia. :/ But I'm squirreling on making sure we land the needful for the release. | 17:09 |
| opendevreview | Merged openstack/networking-generic-switch master: Add Arista bond trunk support https://review.opendev.org/c/openstack/networking-generic-switch/+/973461 | 17:21 |
| TheJulia | cardoe: do the needful for the release, I'm actually making good progress today, at least understanding why my lab setup is unhappy and what is needed to get it working | 17:41 |
| * TheJulia lets current test execute | 17:41 | |
| TheJulia | success! | 17:53 |
| rm_work[m] | Which service is it that takes a really long time to rebuild state when it crashes and comes back up? is it the conductor service? | 18:15 |
| rm_work[m] | and it's because it has to rebuild a bunch of state or something about each host, right? | 18:15 |
| rm_work[m] | I'm trying to remember details, I remember it taking like 24h to come back up at Yahoo after a hard-down | 18:15 |
| JayF | nova-compute services with too many nodes under control act that way, and even worse in the hella-old openstack yahoo ran | 18:17 |
| JayF | you should use sharding to ensure no more than ~300 nodes are handled by each nova-compute | 18:18 |
| rm_work[m] | lol yeah... ok so it was actually the nova-compute side? | 18:18 |
| rm_work[m] | i mean in our case we have so few nodes it only took 10 minutes | 18:18 |
| JayF | yep. most pain is there as we cross the Nova<>Ironic API boundary | 18:18 |
| JayF | it shouldn't be taking 10 minutes anywhere | 18:18 |
| JayF | if you have <300 nodes | 18:18 |
| JayF | are you running old releases and/or really bad DB hardware? | 18:18 |
| rm_work[m] | hmmm | 18:18 |
| rm_work[m] | I don't... think so? I mean our DB is technically just a k8s pod :D | 18:19 |
| rm_work[m] | so maybe that counts as bad lol | 18:19 |
| JayF | so yes | 18:19 |
| JayF | got it | 18:19 |
| JayF | lol | 18:19 |
| rm_work[m] | XD | 18:19 |
| JayF | the faster you make /v1/nodes go, the faster those computes will go, generally | 18:19 |
| JayF | but your pattern doesn't match with the general ones so I wonder if you have a misconfig or something else going on | 18:19 |
| opendevreview | Merged openstack/ironic stable/2025.2: fix redfish inspect system product name https://review.opendev.org/c/openstack/ironic/+/971487 | 18:21 |
| rm_work[m] | hmmm | 18:22 |
| rm_work[m] | can you think of what config options might be related? We did see one that looked like it was limiting the queries it was doing to 3 per second, wondering if that can safely be increased | 18:24 |
| rm_work[m] | but I think that was going to placement | 18:24 |
| JayF | I'm unsure. You also haven't provided information on numbers of nodes, versions of ironic/nova, etc | 18:24 |
| rm_work[m] | ah, we're on 2024.2, and ~240 nodes | 18:25 |
| JayF | That should /not/ be taking ten minutes I believe. | 18:26 |
| JayF | To the point where I'd potentially consider that a bug? Just not sure where to even start looking | 18:26 |
| rm_work[m] | ok i got the info from you i need (basically, this isn't normal) and we should make sure to verify we're not doing something really dumb before you waste any more brain cycles on this :D | 18:27 |
| rm_work[m] | thanks a bunch <3 | 18:27 |
| rm_work[m] | BTW we did get the new version of dnsmasq and it helped a bunch :) | 18:28 |
| rm_work[m] | but we're still looking at doing our own thing, maybe less deeply rewriting something like Y! did, but something we can better control | 18:28 |
| JayF | write your own dhcp implementation that's library driven and go implement it in neutron and ironic | 18:29 |
| JayF | **please | 18:29 |
| JayF | **please ;) | 18:29 |
| rm_work[m] | I would actually probably really enjoy that, but currently fighting to stay above water with our bonkers openstack install | 18:29 |
| rm_work[m] | once we get things handled and all replaced with sane deployment architecture for us (which hopefully will be within the next month or three) I might have better time to look at these things | 18:30 |
| rm_work[m] | I spent like 6 months with legal here working out an upstream contribution model, so at least that part is finally handled XD | 18:30 |
| rm_work[m] | and we're also pulling in ... most of the redfish related changes we see coming through, because we're trying to switch entirely over to that with our funky hardware, and so far it has been going well, if not smoothly. progress! if you need early feedback on any redfish stuff we may be able to help with real-world testing :) | 18:32 |
| cardoe | definitely would like to hear about redfish stuffs | 18:34 |
| rm_work[m] | ping me if there's specific patches of interest you are hoping for feedback on! just talked to Stan and I guess we don't really do inspection yet, but we are using the redfish driver full time now I think | 18:37 |
| cardoe | Nothing specific just if you're seeing issues | 18:38 |
| cardoe | JayF: I'll trade ya a +2/+W on https://review.opendev.org/c/openstack/ironic/+/974345 for a +2/+W on https://review.opendev.org/c/openstack/ironic/+/974069 or https://review.opendev.org/c/openstack/ironic/+/973532 dealers choice. :-D | 19:01 |
| JayF | I'll review patches but I don't do trading | 19:08 |
| JayF | since the one had a conflict, it technically should have a second review | 19:19 |
| JayF | if nobody gets around to it in the next few hours give me a poke and I'll +A anyway | 19:19 |
| cardoe | We need to re-enable the over-indent check that ruff doesn't actually have implemented. | 19:19 |
| cardoe | The issue is that I've tried to re-implement it but that meant I had to fix up the repo. But folks are committing badness faster than I can get the re-enable change to land. | 19:20 |
| JayF | do you have a fixup patch in place now? | 19:20 |
| JayF | you should just coordinate | 19:20 |
| cardoe | So I fix it all up and push the patch. Then it goes to merge and it fails cause someone has added another over-indent error. | 19:20 |
| JayF | you gotta get a lock() with the team | 19:21 |
| JayF | maybe mention it at the meeting or something | 19:21 |
| cardoe | I can do that on Monday. | 19:21 |
| opendevreview | Merged openstack/ironic stable/2025.2: Improve image download performance https://review.opendev.org/c/openstack/ironic/+/974345 | 20:22 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: [doc] Releasenote for iDRAC LLDP https://review.opendev.org/c/openstack/ironic/+/974366 | 20:24 |
| JayF | cardoe: ^ that's a release note for that idrac lldp change if you want something easy to approve | 20:25 |
| cardoe | Done | 20:27 |
| cardoe | The initial functionality is part of this upcoming release and that note incorrectly mentioned Dell so that’s why I had her leave this off. | 20:28 |
| JayF | if behavior changes, we need a release note. I'm not sure what hair you're trying to split but that change is 100% behavior changing :D | 20:29 |
| cardoe | Originally the patch had the basic functionality and Dell as one patch. But she was asked to split off the Dell | 20:29 |
| JayF | aha | 20:36 |
| opendevreview | Merged openstack/ironic stable/2025.2: fix: port endpoints did not return vendor and category and fix docs https://review.opendev.org/c/openstack/ironic/+/973532 | 20:36 |
| cardoe | JayF: I'm happy now ironic-week-prio is down to 1 page. :-D | 21:00 |
| JayF | you keep saying that | 21:00 |
| JayF | I just hear you admitting to having your gerrit default page length too low | 21:00 |
| JayF | lol | 21:00 |
| cardoe | lol fair point | 21:01 |
| opendevreview | Steve Baker proposed openstack/ironic master: Speed the slowest tests https://review.opendev.org/c/openstack/ironic/+/974369 | 21:04 |
| TheJulia | brraaaina | 21:16 |
| TheJulia | err, braaains | 21:16 |
| cardoe | packets are like data traveling over synapses | 21:16 |
| TheJulia | sorry, shared carrier collission which timed up perfectly | 21:17 |
| TheJulia | The whole sense aspect failed in that other neuron | 21:17 |
| opendevreview | Merged openstack/bifrost master: CI: Fix previous release for SLURP upgrade jobs https://review.opendev.org/c/openstack/bifrost/+/973458 | 21:19 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: l2vni baremetal mech driver https://review.opendev.org/c/openstack/networking-baremetal/+/973889 | 21:24 |
| TheJulia | cardoe: ^^^ a few changes, but some I think you'll really like | 21:25 |
| opendevreview | Steve Baker proposed openstack/ironic master: Add an autodetect deploy interface https://review.opendev.org/c/openstack/ironic/+/973187 | 21:39 |
| opendevreview | Steve Baker proposed openstack/ironic master: Replace ironic-tempest-bios-ipmi-direct with -autodetect https://review.opendev.org/c/openstack/ironic/+/974260 | 21:39 |
| opendevreview | Merged openstack/ironic master: Add LLDP collect for DRAC Redfish inspection https://review.opendev.org/c/openstack/ironic/+/970630 | 21:49 |
| opendevreview | Julia Kreger proposed openstack/ironic-specs master: VXLAN networking https://review.opendev.org/c/openstack/ironic-specs/+/959401 | 21:58 |
| TheJulia | hmm, claude has selective focus | 22:02 |
| cardoe | Yes. it literally just wrote a new spec file out after I gave it some feedback on changes I wanted it to make. Then claimed it didn't write a new file that I did that. | 22:03 |
| opendevreview | Steve Baker proposed openstack/ironic master: Speed the slowest tests https://review.opendev.org/c/openstack/ironic/+/974369 | 22:04 |
| cardoe | So the default_physical_network will only work if you have 1 leaf switch | 22:14 |
| TheJulia | Only work if you don't care enough to explicitly set it everywhere because you have a network which is not a split architecture | 22:15 |
| TheJulia | i.e. simpler deployments | 22:15 |
| cardoe | hrm actually good point | 22:19 |
| cardoe | I make a few small comments | 22:19 |
| TheJulia | k, fixing tests | 22:19 |
| TheJulia | claude seems to be super high for an AI right now | 22:20 |
| cardoe | That's really spot on. It's Towelie from South Park. | 22:27 |
| TheJulia | Yes, very yes. | 22:41 |
| TheJulia | Like, borderline shrooms | 22:41 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: l2vni baremetal mech driver https://review.opendev.org/c/openstack/networking-baremetal/+/973889 | 23:05 |
| opendevreview | Merged openstack/ironic master: Speed the slowest tests https://review.opendev.org/c/openstack/ironic/+/974369 | 23:06 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: l2vni baremetal mech driver https://review.opendev.org/c/openstack/networking-baremetal/+/973889 | 23:07 |
| TheJulia | well, that seems to work in my preliminary tests | 23:11 |
| JayF | https://review.opendev.org/c/openstack/ironic/+/974321 I think the release note on this is either wrong, or it's a security vuln | 23:18 |
| JayF | we already were checking the auth algo, we just weren't checking it via openstacksdk at download time | 23:18 |
| JayF | whereas I think that release note reads as "yeah we actually enforce sha256 now" which has always been true aiui | 23:18 |
| TheJulia | from the library, fine to block it and revise it or just follow-it up | 23:19 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Minimum version bump https://review.opendev.org/c/openstack/ironic/+/974321 | 23:20 |
| JayF | I just revised it, I think that's better, I saw the W+1 but didn't pay attention to it not being merged | 23:21 |
| TheJulia | cardoe: fwiw, I'll rebuild my devstack tomorrow and try to verify fully passing packets but also getting claude started on the reconcillation stuffs for subports | 23:21 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!