| opendevreview | Julia Kreger proposed openstack/ironic-tempest-plugin master: Fix failure cases around test_ports https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/934387 | 01:11 |
|---|---|---|
| dtantsur | JayF: I will take a look | 12:02 |
| dtantsur | (re configdrive) | 12:02 |
| ykarel | Hi can someone check https://bugs.launchpad.net/virtualbmc/+bug/2086617 | 13:41 |
| ykarel | not specific to virtualbmc though, but other branchless project sushy-tools, ironic-tempest-plugin | 13:42 |
| TheJulia | The problem is on the branches, we need to pin the version to use. It has happened in the past :( | 14:05 |
| TheJulia | good morning | 14:15 |
| TheJulia | In reality, it is not a virutalbmc bug, but an issue for each and every impacted project | 14:16 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: Agent deploy: account for disable_power_off https://review.opendev.org/c/openstack/ironic/+/934637 | 14:33 |
| ykarel | TheJulia, yes right, i have pushed the patch to do so but one of the job is failing and i don't know why so | 15:12 |
| ykarel | if someone from ironic team can check it would be good to clear this | 15:13 |
| TheJulia | so https://b3858d48ad9ee4e7fde1-83ee41d1d15b6e7c94a25488c54501c9.ssl.cf5.rackcdn.com/934036/3/check/ironic-tempest-uefi-redfish-vmedia/afeaaf5/testr_results.html likely is a result of a defect in our overall CI job model where we attempt to prepare a partition image, which sometimes doesn't work out so well. Specifically the image preparation falls over and the image extraction never actually completes ... the side | 15:40 |
| TheJulia | effects of which are visible in https://b3858d48ad9ee4e7fde1-83ee41d1d15b6e7c94a25488c54501c9.ssl.cf5.rackcdn.com/934036/3/check/ironic-tempest-uefi-redfish-vmedia/afeaaf5/controller/logs/ironic-bm-logs/node-0_console_2024-11-04-19%3A10%3A01_log.txt | 15:40 |
| TheJulia | Because of this on newer branches, we have dialed back the partition image testing | 15:41 |
| TheJulia | Related question, is https://review.opendev.org/#/q/I61f7a6d1af3a66d0e8cc610eda5829c7703d144f only applied to the two branches which are impacted by/using focal or are there others? | 15:42 |
| ykarel | TheJulia, should apply to even older unmaintained too xena/yoga | 15:47 |
| ykarel | in neutron periodic queue we have one ironic job running in zed+ so only pushed for those branches | 15:47 |
| ykarel | ok seems you meant https://review.opendev.org/c/openstack/ironic/+/913270 with "we have dialed back the partition image testing" | 15:56 |
| cardoe | TheJulia: can I poke ya to look at https://bugs.launchpad.net/ironic/+bug/2086823 whenever you have some $TIME? | 18:00 |
| JayF | cardoe: I find myself wondering if "pointing to an empty policy file" rather than "not specifying a policy file" could be some weird edge | 18:10 |
| JayF | cardoe: as CID asked, can you plop the node object in there too / confirm the node.owner setting on it | 18:10 |
| cardoe | bleh didn't see CID's comment sorry. | 18:12 |
| cardoe | is there a ``` in launchpad to say this is a block of code? | 18:12 |
| * TheJulia looks at her hat, and notes jira items are still attached ot it at the moment | 18:20 | |
| TheJulia | cardoe: ping me after 1pm pacific and you can have the rest fo my brain spoons for the day at that time | 18:21 |
| cardoe | It's an ugly comment but I've attached it.. https://bugs.launchpad.net/ironic/+bug/2086823 cid, you can see the owner is the correct project_id | 18:21 |
| JayF | cardoe: fwiw -f yaml is really good for this kinda thing in the future (I think it's -f yaml) | 18:24 |
| JayF | honestly one of the things I <3 the most about our client is so many output options | 18:24 |
| cardoe | yeah that works | 18:24 |
| opendevreview | Adam McArthur proposed openstack/ironic-tempest-plugin master: WIP: Microversion handling test https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/933970 | 18:34 |
| cardoe | JayF: launchpad eats the whitespace | 18:34 |
| cardoe | Cause why not | 18:34 |
| JayF | whitespace is delicious | 18:35 |
| cardoe | so the yaml output sucks | 18:35 |
| JayF | oh. hahahathud | 18:36 |
| * JayF gotta go again & | 18:38 | |
| opendevreview | Adam McArthur proposed openstack/ironic-tempest-plugin master: WIP: Microversion handling test https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/933970 | 18:56 |
| TheJulia | JayF: https://review.opendev.org/c/openstack/ironic/+/931055 has a quick question | 19:19 |
| opendevreview | Julia Kreger proposed openstack/networking-baremetal master: prevent break on communications failure https://review.opendev.org/c/openstack/networking-baremetal/+/933149 | 19:23 |
| TheJulia | cardoe: just a minor commit message fix on ^ | 19:23 |
| cardoe | +2'd | 19:25 |
| opendevreview | Merged openstack/ironic stable/2024.2: add qemu-img to necessary dependency list https://review.opendev.org/c/openstack/ironic/+/934336 | 19:35 |
| opendevreview | Merged openstack/ironic master: update doc and drop deprecated plugin https://review.opendev.org/c/openstack/ironic/+/898157 | 19:35 |
| opendevreview | Adam McArthur proposed openstack/ironic-tempest-plugin master: Testing microversions on v1/shards https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/935219 | 19:39 |
| JayF | TheJulia: I thought I had fixed that. That's my mistake. We can either land it as is and I'll follow up or I'll get an updated change posted at some point. | 19:41 |
| * JayF at the vet office | 19:41 | |
| adamcarthur5 | https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/935219: Jay and I want to start testing wrong microversion handling in our tempest tests. This is the first of many, starting with shard. | 19:42 |
| adamcarthur5 | It will better ensure we are not changing any behaviour while we are making changes like https://review.opendev.org/c/openstack/ironic/+/928919 going forward. | 19:42 |
| JayF | 💪💪💪 | 19:43 |
| adamcarthur5 | And hopefully its a pattern we can keep up: write the tests in tempest, then write the validation change. | 19:43 |
| adamcarthur5 | cc @gtema stephenfin - when this gets merged, I will recheck against it, then we can get the validation for shard merged :) | 19:44 |
| opendevreview | Adam McArthur proposed openstack/ironic master: api: Introduce new mechanism for API versioning https://review.opendev.org/c/openstack/ironic/+/928919 | 20:03 |
| opendevreview | Adam McArthur proposed openstack/ironic master: WIP: api: Add schema validation framework https://review.opendev.org/c/openstack/ironic/+/928920 | 20:03 |
| opendevreview | Adam McArthur proposed openstack/ironic master: WIP: api: Add schema for allocations API https://review.opendev.org/c/openstack/ironic/+/928921 | 20:03 |
| adamcarthur5 | Just rebasing further up the relation chain, sorry for the extra messages | 20:04 |
| TheJulia | JayF: ack, no worries | 21:03 |
| TheJulia | done, and change should be showing up in a moment | 21:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: trivial: remove libpq from bindep file https://review.opendev.org/c/openstack/ironic/+/935226 | 21:07 |
| TheJulia | cardoe: so... | 21:07 |
| JayF | wait, what | 21:08 |
| JayF | https://review.opendev.org/c/openstack/ironic/+/931055/7/bindep.txt | 21:08 |
| JayF | I'm showing it's not in there | 21:08 |
| JayF | DEVSTACK/FILES/BINDEP.TXT OhHhhhhhh | 21:08 |
| JayF | I checked /bindep.txt approximately 9000 times :D | 21:08 |
| TheJulia | heh | 21:13 |
| TheJulia | no worries, I got it :) | 21:16 |
| TheJulia | cardoe: i love your config file btw :) | 21:20 |
| TheJulia | redaction text, that is :) | 21:20 |
| cardoe | I had fun with it. :) | 21:27 |
| TheJulia | okay, yeah, this is a bug | 21:27 |
| cardoe | JayF: yeah that libpq I even asked if you meant to leave it in | 21:28 |
| TheJulia | i think it is a quick fix, let me confirm | 21:28 |
| JayF | like I said, I was looking at the wrong file the whole time | 21:28 |
| JayF | I'm glad we got it sorted in any event | 21:28 |
| TheJulia | yeah, this is a easy-ish bug to fix | 21:30 |
| TheJulia | cardoe: so your issue is basically that if one is doing any node policy comparison against a rule which requires subfield matching, the handler needs to populate that data and be aware of it | 21:34 |
| TheJulia | one cannot just call check_policy directly | 21:34 |
| TheJulia | we have a special handler for that in the api utilities | 21:34 |
| TheJulia | but cid didn't know that nor the reviewers really were not concious of it when the runbooks change was reviewed | 21:35 |
| JayF | Yeah, I'm very curious to see the fix | 21:35 |
| JayF | because this is news to me too | 21:35 |
| TheJulia | yeah, looks like with the runbook policy check itself, it was done right | 21:42 |
| jamesdenton | when IPA is pushing introspection data to inspector:5050, i'm getting a read timeout (30 sec default), but it looks like data eventually comes back maybe 10 seconds later. Is that timeout value configurable? | 21:42 |
| TheJulia | JayF: this is one of those things where the tests will take me like 30 minutes | 21:47 |
| JayF | yeah, makes sense, I'll still make a point to look at it once it's up, please tag cid in it too | 21:48 |
| cardoe | jamesdenton: inspector is deprecated. long live agent | 21:49 |
| JayF | I'm still looking , Idon't see anything in config | 21:49 |
| cardoe | jamesdenton: read on which side? | 21:53 |
| JayF | jamesdenton: https://opendev.org/openstack/oslo.service/src/branch/master/oslo_service/_options.py#L86 looks like the only config option we have, and it's set to ... 15 minutes afaict | 21:53 |
| JayF | I suspect I'm looking at the wrong seam | 21:54 |
| cid | I will love to see , re policy fix | 21:55 |
| jamesdenton | brb | 21:57 |
| TheJulia | still working on the tests :) | 22:02 |
| opendevreview | Merged openstack/ironic master: Remove postgresql testing and documentation https://review.opendev.org/c/openstack/ironic/+/931055 | 22:02 |
| TheJulia | hmm, the extra set_provision_state code is fun | 22:07 |
| jamesdenton | cardoe deprecated, schmeprecated | 22:14 |
| jamesdenton | JayF gonna try ipa-http-request-timeout on the kernel command line | 22:17 |
| jamesdenton | that defaults to 30 seconds | 22:17 |
| TheJulia | almost got it | 22:20 |
| JayF | aha, good stuff jamesdenton | 22:24 |
| JayF | jamesdenton: btw; almost all of those you can bake into a config and drop in your ramdisk | 22:24 |
| JayF | jamesdenton: if you find it's static and want to not have the worlds longest kernel command line :D | 22:25 |
| jamesdenton | if i were good at this I would consider that | 22:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Fix policy checks https://review.opendev.org/c/openstack/ironic/+/935230 | 22:31 |
| TheJulia | There you guys go | 22:31 |
| TheJulia | that will be one ice cream sandwich or tasty whiskey | 22:31 |
| TheJulia | :) | 22:32 |
| TheJulia | I guess my subject should have been slightly better | 22:32 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Fix policy checks added with runbooks https://review.opendev.org/c/openstack/ironic/+/935230 | 22:32 |
| TheJulia | there we go | 22:32 |
| opendevreview | Verification of a change to openstack/ironic master failed: Use patched dnsmasq from PPA https://review.opendev.org/c/openstack/ironic/+/933104 | 22:36 |
| JayF | so we have to call check_owner_policy for things we want for project-scoped users? | 22:36 |
| JayF | at least in cases like that where we check late | 22:36 |
| TheJulia | if you want to compare anything which has a node.*something match in the policy | 22:37 |
| JayF | I guess node is the only real scoped object we have (so far) | 22:37 |
| TheJulia | if you look at the runbooks thing, it got its own owner handler as well, and that was good | 22:38 |
| JayF | oooh, runbooks are too! that's right! | 22:38 |
| TheJulia | yeah | 22:38 |
| TheJulia | and ports through inheratance relationshiop | 22:38 |
| JayF | I was wondering about that | 22:38 |
| TheJulia | its a little weird, admittedly | 22:38 |
| JayF | makes sense why it's not explicit | 22:38 |
| TheJulia | but it is also super easy to miss | 22:38 |
| TheJulia | because if your not aware of the need on the check, you'll never know about it | 22:38 |
| JayF | honestly my takeaway is more like, we should've had these provision state tests | 22:38 |
| TheJulia | Yes, exactly | 22:38 |
| JayF | this is the reason I'm asking adamcarthur5 to do tempest-microversion-testing ahead of us changing them | 22:39 |
| TheJulia | I didn't see added tests for the policies and that would have highlighted it... or at least started to | 22:39 |
| TheJulia | I did shuffle it slightly because the rbac checks were deferred | 22:39 |
| TheJulia | and they really shouldn't be | 22:39 |
| TheJulia | so I fixed that as well which brought validity to the check, so success we know is 409, failure with 403 is insufficient access | 22:40 |
| TheJulia | the lesser access checks likely fail because they don't have rights, but that should be okay. | 22:40 |
| TheJulia | RBAC checks are not easy, that is for sure | 22:40 |
| TheJulia | myself or stevebaker are good resources for that | 22:40 |
| cid | thanks cardoe, for that catch and, TheJulia, for the fix :D | 23:00 |
| JayF | there's always a bright side to someone finding a bug you wrote: someone used the code you wrote :D | 23:08 |
| JayF | and apparently Rackspace is gonna be runbooking before our downstream is :D | 23:08 |
| TheJulia | nope, it was the if not runbook then apply new ACLs path | 23:29 |
| TheJulia | so they were not doing runbooks. | 23:29 |
| cardoe | JayF: yeah this was just attempting to run stuff with steps. | 23:35 |
| cardoe | I wanna be runbooking | 23:36 |
| JayF | ah, I see | 23:39 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Task Manager: Log if the lock takes too long https://review.opendev.org/c/openstack/ironic/+/892412 | 23:42 |
| cardoe | JayF: you okay with me using ruff like sushy? | 23:44 |
| JayF | I am generally going to be onboard with using more modern python toolsets, with two caveats: 1) it's being used somewhere else in openstack already and 2) packages are generally available on most OSes our devs will use | 23:52 |
| JayF | but I am probably the lowest bar to clear for stuff like that among team consensus | 23:53 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!