| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: [DNM] dump flows after port plug https://review.opendev.org/c/openstack/networking-generic-switch/+/955739 | 02:52 |
|---|---|---|
| rpittau | good morning ironic! o/ | 07:15 |
| FreemanBoss[m] | Good morning | 07:18 |
| queensly[m] | Good morning o/ | 07:24 |
| opendevreview | Merged openstack/ironic master: [docs] Update manual clean and verify steps https://review.opendev.org/c/openstack/ironic/+/955730 | 07:43 |
| opendevreview | Verification of a change to openstack/ironic master failed: api: Add schema for allocations API (versioning) https://review.opendev.org/c/openstack/ironic/+/945217 | 07:45 |
| adam-metal3 | Hello Ironic o/! It might be a dumb question, I was trying to reduce the logs Ironic was generating, I know I can set the "debug" flag and that works just fine but when I tried to change the "default_log_levels" under the [DEFAULT] section in the ironic config nothing happened I have set the error level of all the libs to ERROR but I have been seeing still a lot of messages, Is there some easier way | 08:02 |
| adam-metal3 | to only allow ERROR level logs to be displayed? | 08:02 |
| adam-metal3 | or if I have to also add the internal libs of Ironic to this list of "default_log_levels" does someone have a config that I could copy? | 08:05 |
| opendevreview | Merged openstack/ironic stable/2025.1: Fix inspection rules validation https://review.opendev.org/c/openstack/ironic/+/956112 | 08:36 |
| opendevreview | John Garbutt proposed openstack/ironic master: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 12:49 |
| opendevreview | John Garbutt proposed openstack/ironic master: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 12:50 |
| opendevreview | John Garbutt proposed openstack/ironic master: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 12:52 |
| TheJulia | adam-metal3: sounds like your wanting to suppress the periodics from logging. That setting is a bit weird, but I would suspect it to work. Could you share what you tried setting? | 13:08 |
| adam-metal3 | TheJulia: https://pastebin.com/1hanYiqR ;;; passoword: sRnHRRJ2eS | 13:20 |
| adam-metal3 | I think my confusion is mainly related to how to get rid of these info messages that are not coming from external libs but from internal modules/libs | 13:21 |
| TheJulia | so to answer your question, you should be able to do exactly that | 13:22 |
| opendevreview | Merged openstack/ironic master: doc: Update contributor docs to include multiple doc sources and preview options https://review.opendev.org/c/openstack/ironic/+/948541 | 13:29 |
| adam-metal3 | TheJulia: Okay Thanks! I will do so | 13:30 |
| TheJulia | Your definitely highlighting an area which could be more clear, or have an easier overall knob. At the same time we're likely logging a bit too aggressively but don't have a good way to say "ignore that" short of the general knob | 13:31 |
| opendevreview | Merged openstack/ironic master: Restore architecture.rst with redirect notice to overview https://review.opendev.org/c/openstack/ironic/+/946615 | 13:40 |
| opendevreview | Morten Stephansen proposed openstack/ironic-python-agent master: Fix for motherboards where efibootmgr returns UTF-8. https://review.opendev.org/c/openstack/ironic-python-agent/+/956068 | 14:26 |
| TheJulia | A wild JayF re-appears | 14:32 |
| JayF | what year is it | 14:33 |
| TheJulia | 2199 | 14:33 |
| TheJulia | or 2025... not entirely sure | 14:33 |
| rpittau | hey JayF :) | 14:38 |
| cardoe | Are we sure that's actually JayF and not an LLM chat bot agent? | 14:39 |
| JayF | I assure you, if it were possible I woulda already done it. | 14:39 |
| cardoe | TheJulia: did you see my question about firing off another task? | 14:40 |
| rpittau | JAIF | 14:40 |
| TheJulia | cardoe: no, when? | 14:40 |
| TheJulia | cardoe: uhh, yes. its obviously a code change... uhhhhhhhhh | 14:42 |
| TheJulia | cardoe: take a look at notify_conductor_resume_operation | 14:45 |
| TheJulia | in ironic/conductor/utils.py | 14:45 |
| cardoe | thank you. | 14:45 |
| cardoe | I've just stalled out digging. | 14:46 |
| TheJulia | it leverages the message bus to have the call route through | 14:46 |
| kubajj | JayF: o/ | 14:47 |
| cardoe | yes this is exactly what I was looking for. | 14:48 |
| cardoe | So another question... https://opendev.org/openstack/ironic/src/commit/e415f2a4410f2aa0fedb8b6d385722d756bfef4b/ironic/drivers/base.py#L1346 | 14:52 |
| TheJulia | Those should be able to be nuked, in theory | 14:53 |
| cardoe | We've got ESSENTIAL_PROPERTIES of memory_gb, local_gb, and cpu_arch. But only redfish, irmc, and ilo enforce that inspection gets those set. | 14:53 |
| TheJulia | the traits are more important | 14:53 |
| cardoe | okay so my change to make redfish work like IPA inspector becomes much easier if I can remove the ESSENTIAL_PROPERTIES test from redfish first. | 14:54 |
| TheJulia | I suspect so, but that also becomes moderately drastic. I think it makes sense to remove though. I just don't know what else really expects it (still...). | 14:55 |
| cardoe | well redfish runs the test to ensure they're set in the inspect_hardware() call. But with my change they're being populated by the inspection hooks which are happening later. | 14:57 |
| cardoe | It's all still set and correct. | 14:57 |
| cardoe | But I have to rewrite all those tests to check that it's set in a later call. | 14:58 |
| cardoe | I originally went to just copy the IPA inspector's tests for ESSENTIAL_PROPERTIES, but it doesn't have any at all. | 14:58 |
| TheJulia | #startmeeting ironic | 15:00 |
| opendevmeet | Meeting started Mon Aug 4 15:00:26 2025 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
| opendevmeet | The meeting name has been set to 'ironic' | 15:00 |
| TheJulia | What soul is running this meeting today? | 15:00 |
| TheJulia | ... I guess it is me? | 15:01 |
| TheJulia | do we have quorum? | 15:01 |
| kubajj | o/ | 15:02 |
| opendevreview | John Garbutt proposed openstack/ironic master: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 15:03 |
| cardoe | You didn't hit the JayFbot trigger word to make him respond. | 15:03 |
| cardoe | JayF: welcome back btw. I'm just messing around with ya. | 15:03 |
| TheJulia | #topic Announcements / Reminders | 15:03 |
| JayF | \o | 15:04 |
| TheJulia | Reminder, please review items on the priority review dashboard. | 15:04 |
| TheJulia | #link https://tinyurl.com/ironic-weekly-prio-dash | 15:04 |
| TheJulia | According to the schedule, we're in week R-8. We're in the last of the cycle so we need to begin to wrap up items for this cycle. | 15:04 |
| TheJulia | #link https://releases.openstack.org/flamingo/schedule.html | 15:05 |
| TheJulia | Anyone have anything else to remind us of this week? | 15:05 |
| cid | o/, welcome back, JayF! | 15:05 |
| TheJulia | Yes, Welcome back Jay! | 15:07 |
| TheJulia | Onward! | 15:07 |
| TheJulia | #topic Working Group Updates | 15:07 |
| TheJulia | First up, standalone networking. alegacy do you want to provide a brief update? | 15:07 |
| TheJulia | I guess he is busy. He gave an awesome quick demo last week. He expects to post code this week. | 15:09 |
| TheJulia | Next up is eventlet working group. Work has moderately stalled on needing changes to futurist. That has been workflowed so hopefully we can get a release cut soon. Overall I also wrote a blog post for the ironic website as well to kind of highlight details. If anyone is interested, please let me know. | 15:10 |
| TheJulia | Looks like we don't have any discussion topics, so onward to Bug Deputy Updates! | 15:10 |
| TheJulia | #topic Bug Deputy Updates | 15:10 |
| JayF | I'll review that post if you can link it in | 15:10 |
| JayF | would be good catchup reading anyway perhaps | 15:10 |
| cid | Quiet week, 3 new books and a carryover RFE from the past week | 15:11 |
| TheJulia | JayF: https://github.com/OpenStackweb/ironic-website/pull/62 | 15:11 |
| TheJulia | books?! :) | 15:11 |
| cid | https://bugs.launchpad.net/ironic/+bug/2118746 Records cannot be deleted when ironic-conductor service is down (could be a bug?) | 15:11 |
| JayF | That's right on the border of RFE/bug | 15:12 |
| TheJulia | Yeah, that is by design IMHO | 15:12 |
| JayF | we *never* clean them up, right? | 15:12 |
| TheJulia | .... I think we do, on startup of another conductor | 15:13 |
| TheJulia | I'd have to double check, I typically only run environments where they are expected to be up | 15:13 |
| JayF | If that's the actual behavior; IMO not a bug not a feature | 15:13 |
| JayF | same | 15:13 |
| TheJulia | The API surface though, for this is afaik intended to be read only | 15:14 |
| TheJulia | I suspect if we don't, we would just want to look at auto-purge or something as an automatic workaround of sorts | 15:14 |
| JayF | yeah | 15:15 |
| JayF | similar to node history cleanups | 15:15 |
| JayF | if dead conductor older than X, cleanup entry | 15:15 |
| TheJulia | yeah | 15:15 |
| TheJulia | I can take the action item to look after the meeting | 15:15 |
| cid | So, it's related to this one ? https://bugs.launchpad.net/ironic/+bug/2069771 | 15:16 |
| TheJulia | Yup | 15:17 |
| JayF | they aren't related, it's the same bug IMO | 15:17 |
| cid | :D | 15:17 |
| cid | Okay, 'cus I was working on a patch. | 15:17 |
| TheJulia | cid: what path were you taking? | 15:18 |
| cid | I got some ideas from Kaifeng's response in the later bug link | 15:20 |
| TheJulia | Yeah, I was thinking the same | 15:20 |
| cid | Great. I'll gerrit what I have, I will take you too, if you don't mind :) | 15:21 |
| cid | *tag/ping | 15:21 |
| TheJulia | sounds good! | 15:21 |
| TheJulia | Anything else? | 15:21 |
| cid | Yeah, that's all with bug deputy updates | 15:22 |
| TheJulia | if not, who shall be the bug deputy this next week? | 15:22 |
| cid | Mua ? | 15:22 |
| TheJulia | ? | 15:23 |
| cid | I meant me, :D | 15:23 |
| TheJulia | if you so desire :) | 15:23 |
| cid | ++ | 15:23 |
| TheJulia | Since we have no RFEs, on to Open Discussion! | 15:23 |
| TheJulia | #topic Open Discussion | 15:23 |
| TheJulia | Does anyone have anything for us to discuss? | 15:23 |
| TheJulia | JayF: as an fyi, the website build is very broken right now, I talked to Wes about it last week and it slipped off a radar. I've been promised that the dependency issues should be resolved this week. | 15:24 |
| TheJulia | Well folks, thanks everyone! Lets have a great week! | 15:26 |
| TheJulia | #endmeeting | 15:26 |
| opendevmeet | Meeting ended Mon Aug 4 15:26:49 2025 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:26 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-08-04-15.00.html | 15:26 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-08-04-15.00.txt | 15:26 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/ironic/2025/ironic.2025-08-04-15.00.log.html | 15:26 |
| opendevreview | Merged openstack/ironic master: api: Add schema for allocations API (versioning) https://review.opendev.org/c/openstack/ironic/+/945217 | 16:21 |
| TheJulia | cid: yeah, thinking about it, if we've not seen it in ?5 days? then maybe okay to drop the record? | 16:32 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Replace GreenThreadPoolExecutor in conductor https://review.opendev.org/c/openstack/ironic/+/952939 | 16:34 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Set the backend to threading. https://review.opendev.org/c/openstack/ironic/+/953683 | 16:34 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Clean-up misc eventlet references https://review.opendev.org/c/openstack/ironic/+/955632 | 16:34 |
| opendevreview | cid proposed openstack/ironic master: Add periodic cleanup of dead conductor entries https://review.opendev.org/c/openstack/ironic/+/956500 | 16:35 |
| opendevreview | cid proposed openstack/ironic master: Add periodic cleanup of dead conductor entries https://review.opendev.org/c/openstack/ironic/+/956500 | 16:50 |
| cid | TheJulia, Alright, see the current shape ^^ | 16:52 |
| opendevreview | John Garbutt proposed openstack/ironic master: Fix inspection IB port client-id https://review.opendev.org/c/openstack/ironic/+/955806 | 17:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: API optional indirection API use https://review.opendev.org/c/openstack/ironic/+/956504 | 17:23 |
| TheJulia | cid: one aspect which seems weird is it is introducing both a periodic and the ability to do so via the api surface. While not an awful idea, it will still require the conductor depending on case, and I'm also sort of wondering if that routine should actually guard it out since deleting an active conductor is sort of a bad idea and can cause chaos | 17:33 |
| JayF | https://github.com/openstack/ironic/commit/03fd6f992454fe2d609efcebc81bf0b627ea2abd FWIW that configuration was added specifically because we regressed that configuration once. If we can find a way to do it while keeping CI happy, it'd be ideal. | 17:44 |
| cid | TheJulia, do you advice to get rid of that aspect entirely or adding a guard will be good enough | 17:50 |
| JayF | I haven't reviewed your specific code yet, but I would generally think we're better off without an API endpoint to trigger that clean up | 18:16 |
| opendevreview | cid proposed openstack/ironic master: Add periodic cleanup of dead conductor entries https://review.opendev.org/c/openstack/ironic/+/956500 | 18:18 |
| cid | I thought so too, updated | 18:18 |
| JayF | dtantsur: your change from local RPC to automated-json-rpc appears to have broken tox -elocal-ironic-dev entirely? | 18:34 |
| JayF | I don't think I have anything weird/broken locally, but it errors unable to create the socket. | 18:35 |
| JayF | also we have a warning saying to only use ironic/sqlite as single process... and I think we removed the ability to run as a single process? | 18:35 |
| JayF | What's the piece here I'm missing? | 18:35 |
| JayF | referring to https://review.opendev.org/c/openstack/ironic/+/954755 | 18:36 |
| JayF | cid: https://github.com/openstack/ironic/commit/a4e7939b66a938e0001d542bd5173bf779b2f6dc would you mind following this up with a release note? | 18:54 |
| JayF | obviously not urgent :D | 18:54 |
| JayF | that's a great change btw | 18:56 |
| JayF | https://github.com/openstack/ironic/commit/94948bb1945c83b54a178d4e7e5ea54eb1af6f3e is also awesome but needs a post-facto release note, too | 18:59 |
| TheJulia | JayF: well, the VMs don't have cross-host dns resolution, so nova-compute blows up. Maybe an eventlet removal side-effect ? | 19:27 |
| JayF | I mean, it could be a million things | 19:27 |
| JayF | but that job was added during The Horrible Image Time(tm) | 19:27 |
| JayF | because we had a version of the patch that regressed on that config | 19:27 |
| TheJulia | JayF: no, we didn't remove single process mode fwiw, and we expect it to break soon if we don't fix db queries which I've started on | 19:28 |
| JayF | just shifted the default? | 19:28 |
| JayF | so I gotta figure out how to manually set it up to fix ironic-dev-env | 19:28 |
| TheJulia | JayF: the code now routes locally for rpc through a socket instead of direct code invocations. by chance did you read my blog pr yet? | 19:29 |
| JayF | it's in a tab | 19:29 |
| JayF | that change specifically is what appears to have broken our tox api magic | 19:30 |
| TheJulia | maybe we might need to sync up on what to expect and the path we're heading as a result | 19:31 |
| JayF | I just know there are folks (on my team) who depend on that, I was reading commits to ironic (this part is done) and tested it when I saw that one | 19:32 |
| JayF | sync is not a bad idea, but not today | 19:32 |
| JayF | my brain is already on overflow | 19:32 |
| JayF | TheJulia: there is a hole in my mind?! :D lol | 19:33 |
| TheJulia | yeah, I guess it would really help to understand how it is suddenly broken but at the same time hiding tooling in tox is sort of unexpected by many so I can see how it could be problematic, but things are going to break at times and we're going to need to fix them | 19:34 |
| TheJulia | ESPECIALLY around eventlet stuffs. | 19:34 |
| JayF | tox -elocal-ironic-dev is just a nice trick for folks like adamcarthur5 who are mainly doing r/o client work :) | 19:35 |
| JayF | My concern is that we fix it and move forward :D I figured this was wrapped in eventlet, somehow | 19:35 |
| JayF | everything is | 19:35 |
| * JayF hasn't typed the word 'eventlet' in 6 weeks /s | 19:35 | |
| TheJulia | this is honestly the first I'm hearing its broken | 19:36 |
| JayF | it landed July 30, it hasn't been that long | 19:36 |
| JayF | I'm not surprised, we'll figure it out | 19:36 |
| opendevreview | Verification of a change to openstack/ironic master failed: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 19:37 |
| TheJulia | oh, heh, I thought it landed a few weeks ago at this point | 19:38 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: [DNM] dump flows after port plug https://review.opendev.org/c/openstack/networking-generic-switch/+/955739 | 20:07 |
| cardoe | free candy for those who fix bad test data... https://review.opendev.org/c/openstack/ironic/+/955536 | 20:21 |
| cardoe | we didn't branch 2025.2 yet right? | 20:22 |
| opendevreview | cid proposed openstack/ironic master: Follow-up: Clear `last_error` on power match/sync https://review.opendev.org/c/openstack/ironic/+/956511 | 20:22 |
| TheJulia | no, we've not, maybe 6 weeks until we have to | 20:23 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: fix redfish processor inspection https://review.opendev.org/c/openstack/ironic/+/955537 | 20:30 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: allow running inspection hooks on redfish interface https://review.opendev.org/c/openstack/ironic/+/933066 | 20:30 |
| opendevreview | Verification of a change to openstack/ironic master failed: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 20:48 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: API optional indirection API use https://review.opendev.org/c/openstack/ironic/+/956504 | 21:00 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Remove direct mapping from API -> DB https://review.opendev.org/c/openstack/ironic/+/956512 | 21:00 |
| TheJulia | so that *should* sort of become a clean switchover for the API-> DB interactions | 21:01 |
| TheJulia | I guess we'll see | 21:01 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: set a maximum systemd journal size https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/954527 | 21:16 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Revert "ci: stabilize ironic-standalone-redfish" https://review.opendev.org/c/openstack/ironic/+/956515 | 21:19 |
| TheJulia | JayF: ^ should do the needful, fwiw regarding the change your panicing about for the job config | 21:19 |
| JayF | I wouldn't say panic, just making a specific note about why that config was set up the way it was. | 21:20 |
| JayF | Here's the part where I would normally say that I'll take a look at seeing if I can make it more stable, but I don't want to make any promises until I get to the bottom of my what have I missed pile. | 21:20 |
| TheJulia | JayF: regarding your tox -elocal-ironic-dev, it works for me so... I guess more insight is needed. Ultimately dmitrys' change is correct for what we're trying to tackle but it might be local state? your saying a socket error. | 21:23 |
| JayF | I'm going to feel 10/10 stupid if adding a -r fixes it | 21:23 |
| TheJulia | dunno, it might! worth a try anyhow? | 21:24 |
| JayF | already running | 21:24 |
| TheJulia | success?! | 21:25 |
| JayF | Similarly shaped failure. OS error no socket could be created | 21:28 |
| JayF | I'll stick a pin in it for now. | 21:29 |
| TheJulia | does it say what socket? | 21:29 |
| TheJulia | Just curious if something else is already running | 21:29 |
| JayF | 2025-08-04 14:24:21.167 416640 ERROR oslo_service.backend.eventlet.service OSError: No socket could be created -- (('::1', 8089, 0, 0): [Errno 99] Cannot assign requested address) | 21:30 |
| JayF | I have IPv6 access, not a disabled v6 situation | 21:30 |
| JayF | and there's no services already on those ports | 21:31 |
| JayF | ...what the hell, I don't have a v6 localhost IP?! | 21:31 |
| TheJulia | doh?! | 21:34 |
| JayF | TheJulia: as I put it in #gentoo once I figured it out: some idiot with root on my desktop disabled ipv6 via sysctl and didn't notice because networkmanager re-enables it on the interfaces that matter | 21:52 |
| JayF | (I'm the idiot) | 21:52 |
| TheJulia | c'est la vie | 21:53 |
| JayF | I wonder how many random broken systems things are fixed now | 21:53 |
| JayF | ...and now it works \o/ | 21:54 |
| JayF | TheJulia: dtantsur: Sorry for the false alarm re: tox -e local-ironic-dev :D | 21:54 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Replace GreenThreadPoolExecutor in conductor https://review.opendev.org/c/openstack/ironic/+/952939 | 22:56 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Set the backend to threading. https://review.opendev.org/c/openstack/ironic/+/953683 | 22:56 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Remove direct mapping from API -> DB https://review.opendev.org/c/openstack/ironic/+/956512 | 22:56 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: API optional indirection API use https://review.opendev.org/c/openstack/ironic/+/956504 | 22:56 |
| opendevreview | Verification of a change to openstack/ironic master failed: Populate switch_info with lldp system name https://review.opendev.org/c/openstack/ironic/+/956471 | 23:14 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: [DNM] dump flows after port plug https://review.opendev.org/c/openstack/networking-generic-switch/+/955739 | 23:45 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: Create service plugin to configure security groups https://review.opendev.org/c/openstack/networking-generic-switch/+/949834 | 23:47 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: Support security groups in netmiko base class https://review.opendev.org/c/openstack/networking-generic-switch/+/949835 | 23:47 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: Add security group support to netmiko_cisco_nxos https://review.opendev.org/c/openstack/networking-generic-switch/+/952314 | 23:47 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: Add security group support to netmiko_sonic https://review.opendev.org/c/openstack/networking-generic-switch/+/955252 | 23:47 |
| opendevreview | Steve Baker proposed openstack/networking-generic-switch master: WIP Add security group support to ovs https://review.opendev.org/c/openstack/networking-generic-switch/+/956519 | 23:47 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!