| cardoe | So I wrote a little CLI inspired by clif and TBN. I’ll contribute it for the inspection rules. | 00:42 |
|---|---|---|
| TheJulia | ++ | 01:02 |
| TheJulia | https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/971233 should allow us to remove non-voting on the advanced ops tests, so we can unwind the issues alegacy and the otehr reporter highlighted regarding dhcpless paths | 01:05 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 03:45 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: ci: fix and log errors on advanced tests disqualification https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/971233 | 04:12 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 04:33 |
| opendevreview | Jacob Anders proposed openstack/ironic master: Add hardware health monitoring via management interface https://review.opendev.org/c/openstack/ironic/+/966946 | 04:42 |
| opendevreview | Jacob Anders proposed openstack/python-ironicclient master: Add support for node health status field https://review.opendev.org/c/openstack/python-ironicclient/+/967055 | 04:59 |
| rpittau | good morning ironic! o/ | 07:45 |
| rpittau | TheJulia: the bifrost jobs running out of ram at grub is a known issue, I think I mentioned it when I started working on the migration of the bifrost jobs from tinyipa to DIB based | 07:47 |
| rpittau | the migration to debian based ramdisks should mitigate that https://review.opendev.org/c/openstack/bifrost/+/969886 | 07:47 |
| rpittau | cc dtantsur ^ | 07:47 |
| rpittau | uh jsut realized there's a comment there I need to address :D | 07:48 |
| opendevreview | Riccardo Pittau proposed openstack/bifrost master: Switch IPA ramdisk to debian DIB based https://review.opendev.org/c/openstack/bifrost/+/969886 | 07:52 |
| opendevreview | cid proposed openstack/ironic master: Omit `CapacityBytes` for Redfish RAID MAX size_gb https://review.opendev.org/c/openstack/ironic/+/971319 | 12:57 |
| TheJulia | rpittau: then lets turn off bookworm on the centos10 fix for now, then re-enable in your change | 14:11 |
| TheJulia | since we can get centos10 to pass now minus bookworm | 14:12 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: Centos: set masked: false for firewalld https://review.opendev.org/c/openstack/bifrost/+/971212 | 14:22 |
| TheJulia | rpittau: ^ | 14:22 |
| rpittau | TheJulia: ok, sounds good :) | 14:22 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: ci: undo non-voting for centos10 upgrade https://review.opendev.org/c/openstack/bifrost/+/971258 | 14:33 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: add an OCI artifact registry https://review.opendev.org/c/openstack/bifrost/+/961388 | 14:34 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: bifrost-registry-install: install the ORAS client https://review.opendev.org/c/openstack/bifrost/+/968355 | 14:34 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: Upload a disk image to OCI https://review.opendev.org/c/openstack/bifrost/+/968416 | 14:34 |
| opendevreview | Julia Kreger proposed openstack/bifrost master: Upload to and use artifact from OCI registry https://review.opendev.org/c/openstack/bifrost/+/968417 | 14:34 |
| dtantsur | cardoe: on the Metal3 meeting right now, a few people are basically asking for the Redfish quirks mechanism. I'll point them at you. | 14:53 |
| cardoe | heh. okay. I actually just wanted to get back to that. | 14:53 |
| dtantsur | Now you have extra motivation :) (and I as wel) | 14:53 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: WIP: re-enable simple-init by default https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971330 | 15:02 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Revert "ci: stabilize ironic-standalone-redfish" https://review.opendev.org/c/openstack/ironic/+/956515 | 15:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Revert "ci: stabilize ironic-standalone-redfish" https://review.opendev.org/c/openstack/ironic/+/956515 | 15:07 |
| TheJulia | cardoe: to answer your question, we should likely merge it based upon requests back in august, but since nobody has really reviewed it... *shrugs* | 15:08 |
| cardoe | TheJulia: uhhh what did I ask? | 15:11 |
| cardoe | dtantsur: honestly I'd love if sushy remained standalone and we could do rapid releases of it without an upper-contraints so we can push out quirks fixes. | 15:11 |
| TheJulia | cardoe: if we should drop or merge that job config revert | 15:14 |
| cardoe | oh | 15:14 |
| * TheJulia tosses a wet cat at zuul and a node_failure | 15:14 | |
| cardoe | dtantsur: https://review.opendev.org/c/openstack/ironic/+/971142 speaking of redfish... | 15:15 |
| cardoe | dtantsur: btw where are your metal3 meetings? I can toss those on my calendar and come sometimes | 15:15 |
| TheJulia | Well, so far so good for the stack of bifrost changes | 15:19 |
| rpittau | cardoe: https://docs.google.com/document/d/1IkEIh-ffWY3DaNX3aFcAxGbttdEY_symo7WAGmzkWhU/edit?tab=t.0#heading=h.w4ou76o3rp1v | 15:31 |
| dtantsur | cardoe: that could be an argument for keeping sushy, although the "without upper-constraints" part is hard to make happen | 15:43 |
| opendevreview | Verification of a change to openstack/ironic-python-agent-builder master failed: ci: Start running the advaned ironic job https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971232 | 15:44 |
| cardoe | dtantsur: well we can do patch releases in the stable branches. | 15:47 |
| dtantsur | True. Which is equally possible for Sushy standalone and Sushy inside Ironic. | 15:47 |
| dtantsur | In fact, Ironic has more branches to release from. | 15:48 |
| TheJulia | So: thoughts on opinions on un-blocking cloud-init in IPA ramdisks? | 16:04 |
| dtantsur | As long as we don't break something else :) | 16:05 |
| TheJulia | Yeah, trying not to, hopefully the advanced job will be running on IPA-b today | 16:05 |
| TheJulia | well, voting on it most likely | 16:05 |
| cardoe | dtantsur: oh https://review.opendev.org/c/openstack/ironic/+/970918 should re-enable those checks. You want that? The issue is that hacking is broken with "pre-commit run --all-files" | 16:12 |
| cardoe | But we've got another issue that codespell ignores our settings right now as well. | 16:13 |
| dtantsur | le sigh | 16:13 |
| dtantsur | I'd definitely enable the missing checks but not at the expense of breaking something else | 16:13 |
| cardoe | The issue is a few things have merged that aren't compliant with E128 so I'd need to fix those up first. | 16:14 |
| cardoe | Which is why it's failed. | 16:14 |
| cardoe | https://review.opendev.org/c/openstack/ironic/+/968154 is where codespell fails. | 16:14 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: fix: inspection rules docs to match the code https://review.opendev.org/c/openstack/ironic/+/971265 | 16:19 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: fix loading of built-in inspection rules https://review.opendev.org/c/openstack/ironic/+/971266 | 16:19 |
| cardoe | ^ that's another set of backports. | 16:19 |
| TheJulia | [ 22.824997] systemd[1]: /usr/lib/systemd/system/ironic-python-agent.service:6: Unknown key name 'ConditionPathExists' in section 'Service', ignoring. | 16:22 |
| TheJulia | ConditionPathExists=!/etc/.rescued :( | 16:23 |
| JayF | Wait, waht? | 16:23 |
| JayF | systemd removed that?! | 16:23 |
| cardoe | Does the oslo.i18n function... _() take args like _("my string %s", arg) ? | 16:23 |
| TheJulia | looks like it | 16:24 |
| cardoe | Cause I noticed the inspection_rules use that style but everywhere else we do _("some format %s") % arg | 16:24 |
| JayF | TheJulia: it goes in [unit] now | 16:24 |
| TheJulia | yup, just found that | 16:24 |
| JayF | man 5 systemd.unit # still lists it | 16:24 |
| JayF | makes me wonder if it ever worked | 16:25 |
| JayF | and they just now tell us up front it won't | 16:25 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: fix loading of built-in inspection rules https://review.opendev.org/c/openstack/ironic/+/971266 | 16:25 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Fixed rescue ConditionPathExists flag https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971353 | 16:26 |
| opendevreview | Merged openstack/tenks master: Modernize and fix CI https://review.opendev.org/c/openstack/tenks/+/968200 | 16:29 |
| opendevreview | Riccardo Pittau proposed openstack/sushy master: Add connect_timeout parameter for faster failure on unreachable BMCs https://review.opendev.org/c/openstack/sushy/+/971355 | 16:34 |
| opendevreview | Riccardo Pittau proposed openstack/sushy master: Add read_timeout and connect_timeout parameters for faster BMC failure https://review.opendev.org/c/openstack/sushy/+/971355 | 16:40 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: WIP: lockout configdrive reads on network boots https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971236 | 16:44 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: WIP: re-enable simple-init by default https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971330 | 16:45 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Fixed rescue ConditionPathExists flag https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971353 | 16:45 |
| TheJulia | okay, rebases complete | 16:45 |
| TheJulia | Lets see! | 16:46 |
| TheJulia | alegacy: I think I'm close on the issue you've denoted, ultimately you may want to build yourself a new IPA image with the patches to see for sure, but I'm looking to see if I can also just entirely torpedo glean from running at all if we don't detect the needful as well. | 16:49 |
| opendevreview | Merged openstack/ironic master: Omit `CapacityBytes` for Redfish RAID MAX size_gb https://review.opendev.org/c/openstack/ironic/+/971319 | 17:25 |
| cardoe | cid: so I've updated that inspection rules stuff. I kinda just wanna merge in those docs fixes and just backport them cause that'll make the docs match the code. The last patch just fixes it so the built-in rules can load but the follow on I'm doing is adding examples and a CLI loader / validator tool. That's why I changed it to take the path in. | 17:55 |
| opendevreview | Merged openstack/bifrost master: Centos: set masked: false for firewalld https://review.opendev.org/c/openstack/bifrost/+/971212 | 17:56 |
| cardoe | TheJulia: https://opendev.org/openstack/ironic/src/commit/5f911ad688e682bb52c2f69b0630a497b67beacf/ironic/drivers/modules/fake.py#L112 is why the tests pass on that pxe boot issue. | 18:01 |
| cardoe | So the only thing he can do is change it to walk through each boot interface and mock those out instead of using the fkae. | 18:02 |
| TheJulia | yeah, lets undo the capability label change which removed pxe_boot when kaifeng went in and further split pxe/ipxe stuffs | 18:05 |
| cardoe | I think we shouldn't be calling it "ipxe_boot" | 18:05 |
| TheJulia | why? | 18:05 |
| TheJulia | I ask becasue that was the signaling mechanism and the prior change created a break | 18:06 |
| TheJulia | so we need to "fix" it in prior branches too | 18:06 |
| cardoe | Yes. That's the fix that's backportable. | 18:06 |
| cardoe | Generically speaking though any of the capabilities that's a superset of another capability needs to be signaled in such a way that the mistake cannot happen. | 18:07 |
| TheJulia | (fwiw, the whole purpose of the flags is so we can further hint to the interfaces about what they can/do without doing name inspection which also allows for derivitiave interfaces to be handled appropriately without splitting out an ipxe flavored neutron interface | 18:07 |
| cardoe | e.g. you identified that when you added http_boot as a separate parameter on the BootInterface instead of as a capability | 18:07 |
| TheJulia | Yes | 18:07 |
| cardoe | So you didn't add http_boot to the capabilities list. | 18:08 |
| cardoe | So Milan can add pxe_boot back. But the test is gonna be hard since he'll have to load the interfaces instead of the fake. | 18:08 |
| TheJulia | okay | 18:10 |
| cardoe | https://opendev.org/openstack/ironic/src/commit/c11e12a45b3c762b9b701e704bc1a8e9623c4ae8/ironic/drivers/modules/pxe.py#L50 that's what I'm referring to btw. | 18:10 |
| TheJulia | lets not worry about a test then, and just fix the flag | 18:10 |
| TheJulia | since the tests using fake do the needful | 18:11 |
| cardoe | We're talking in a follow up how to adjust the interface so that we make it hard to use incorrectly. | 18:11 |
| TheJulia | at what point is there really a gain then? | 18:13 |
| cid | cardoe, ack. I have already added my vote, we just need another core to land it. | 18:14 |
| cid | Or should I single core land the doc updates? | 18:14 |
| TheJulia | (FWIW, I'm super low on spoons for this year, and it might be coming off in my replies, sorry) | 18:14 |
| cid | TheJulia, it's been a good and eventful year. Maybe time for a well deserved rest. | 18:16 |
| cid | Also, mind landing this patch? https://review.opendev.org/c/openstack/ironic/+/971265 | 18:16 |
| TheJulia | I'm trying to get the vmedia stuff and bifrost stuff sorted first ;) | 18:16 |
| JayF | TheJulia: you need any more approvals on that chain? | 18:17 |
| JayF | cid: looking | 18:17 |
| cid | TheJulia, ack, JayF, tks | 18:17 |
| JayF | TheJulia: I am in full holiday motivation mode, and there isn't enough caffeine in the universe to kick me outta it | 18:17 |
| cid | https://review.opendev.org/c/openstack/ironic/+/971264/1 | 18:18 |
| TheJulia | JayF: eh, not really, a look on the bifrost changes on https://review.opendev.org/c/openstack/bifrost/+/961388 and beyond it would be good since that make dmitry smile and help us get the OCI image service tested in CI for ironic | 18:19 |
| JayF | these are tested and high confidence? | 18:19 |
| JayF | I don't usually review bifrost stuff so making sure it's OK I'm going to be a human rubber stamp with a sanity check | 18:20 |
| TheJulia | fairly high confidence, still unbricking the rest of bifrost's CI | 18:20 |
| TheJulia | oh! | 18:20 |
| TheJulia | I already approved https://review.opendev.org/c/openstack/bifrost/+/969886 | 18:20 |
| TheJulia | so bookworm jobs will be happy moving forward | 18:21 |
| JayF | that registry patch is exactly the sorta thing I'm not really good to review in bifrost | 18:22 |
| TheJulia | fair enough | 18:23 |
| TheJulia | :) | 18:23 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Lockout configdrive reads on network boots https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971236 | 18:31 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: WIP: re-enable simple-init by default https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971330 | 18:35 |
| TheJulia | alegacy: I sort of see why you didn't see it until a couple days ago too, looks like the images in CI didn't get picked up to actually start using it until this past week really. I'm curious if your end CI job result would have encountered the case you expect or not, otherwise we could consider building a duplicate of the multinode jobs we have, but the downside is we moved those over to be vmedia based instead of network | 18:50 |
| TheJulia | boot based for performance reasons. :\ | 18:50 |
| opendevreview | Steve Baker proposed openstack/ironic-tempest-plugin master: Add graphical console validate to basic ops scenario https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/963643 | 18:51 |
| opendevreview | Merged openstack/ironic-python-agent-builder master: Fixed rescue ConditionPathExists flag https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971353 | 18:56 |
| opendevreview | Merged openstack/ironic-python-agent-builder master: ci: Start running the advaned ironic job https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971232 | 19:11 |
| opendevreview | Merged openstack/ironic master: update inspection rules docs and code to the same order https://review.opendev.org/c/openstack/ironic/+/971264 | 19:47 |
| opendevreview | Merged openstack/ironic master: fix: inspection rules docs to match the code https://review.opendev.org/c/openstack/ironic/+/971265 | 19:48 |
| opendevreview | Doug Goldstein proposed openstack/ironic stable/2025.2: update inspection rules docs and code to the same order https://review.opendev.org/c/openstack/ironic/+/971370 | 20:26 |
| cardoe | JayF: you have a few? | 20:27 |
| JayF | I have enough experience to know better than to say "yes" without knowing the next question :D | 20:27 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Make multiple integration jobs voting https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/971371 | 20:27 |
| JayF | cardoe: seriously what you need? | 20:29 |
| cardoe | Just following up on your comment on the patch for inspection rules | 20:29 |
| JayF | oh, yeah, happy to walk thru that now if you want | 20:30 |
| JayF | meetpad.opendev.org/p/ironic-review-cardoes-thing | 20:30 |
| cardoe | heh beat me to it... I joined https://meetpad.opendev.org/ironic-adhoc | 20:30 |
| cardoe | I'm in yours now | 20:30 |
| * TheJulia blinks | 20:38 | |
| opendevreview | Merged openstack/bifrost master: Switch IPA ramdisk to debian DIB based https://review.opendev.org/c/openstack/bifrost/+/969886 | 21:00 |
| JayF | cardoe: https://docs.openstack.org/ironic/latest/admin/runbooks.html#access-control-for-runbooks | 21:02 |
| opendevreview | Mikkel Vestergård proposed openstack/ironic stable/2025.1: Provide pw_file to ipmitool-socat console https://review.opendev.org/c/openstack/ironic/+/971377 | 21:14 |
| cardoe | TheJulia: tl;dr if you're curious... I'm gonna write a bug for https://review.opendev.org/c/openstack/ironic/+/971266 and we'll backport that to 2025.2 to fix the crasher | 21:14 |
| opendevreview | Julia Kreger proposed openstack/bifrost stable/2025.2: Centos: set masked: false for firewalld https://review.opendev.org/c/openstack/bifrost/+/971378 | 21:15 |
| cardoe | As a follow up we'll turn off mutable on the built_in_rules per JayF and I'll add an oslo type check for validating the input file rather than doing it when the file is first used but that won't be backported. | 21:16 |
| cardoe | Plus my examples and my tools/inspection_rules/tester.py inspired by clif's tools/trait_based_networking/simulator.py will be another commit which won't be backported. | 21:16 |
| TheJulia | cool cool | 21:17 |
| cardoe | Then I'll make a spec for inspection rules API access updates like runbooks so that its not entirely under system scope and implement that but that's likely in scope for 2026.2 | 21:17 |
| cardoe | Trying hard to get our flow to be "see initial box and inspect with redfish... apply some hooks and rules.... switch to agent inspect and apply other hooks and rules" | 21:18 |
| JayF | yeah I suggested to cardoe that he steal the rbac approach for runbooks around having a project + a public field that are mutually exclusive | 21:31 |
| TheJulia | ++ | 21:41 |
| cardoe | Do you want me to update the commit message / release note with a link to https://bugs.launchpad.net/ironic/+bug/2136776 ? | 22:25 |
| cardoe | Cause that'll wipe away your review. | 22:25 |
| JayF | do it and ping me to re-add if I don't | 22:30 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: fix loading of built-in inspection rules https://review.opendev.org/c/openstack/ironic/+/971266 | 22:44 |
| cardoe | JayF: ^ | 22:44 |
| JayF | cardoe: curious if you use openstack-exporter at all (https://github.com/openstack-exporter/openstack-exporter) | 22:45 |
| cardoe | I am | 22:47 |
| * JayF -> DM you | 22:48 | |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!