| opendevreview | Himanshu Roy proposed openstack/ironic master: WIP: Disallow certain steps for users in different deployments https://review.opendev.org/c/openstack/ironic/+/989017 | 09:00 |
|---|---|---|
| opendevreview | Himanshu Roy proposed openstack/ironic master: WIP: Disallow certain steps for users in different deployments https://review.opendev.org/c/openstack/ironic/+/989017 | 09:05 |
| opendevreview | Jakub Jelinek proposed openstack/ironic-python-agent master: Fix skip block devices for RAID arrays with UEFI https://review.opendev.org/c/openstack/ironic-python-agent/+/989083 | 09:36 |
| opendevreview | Jakub Jelinek proposed openstack/ironic-python-agent master: Fix skip block devices for RAID arrays with UEFI https://review.opendev.org/c/openstack/ironic-python-agent/+/989083 | 09:51 |
| dtantsur | Anyone else got tox broken locally? Oo | 10:24 |
| dtantsur | he conflict is caused by: | 10:24 |
| dtantsur | The user requested oslo.context>=2.22.0 | 10:24 |
| dtantsur | The user requested (constraint) oslo-context===6.3.0,===6.4.0 | 10:24 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: Document the Metal3 ironic-standalone-operator job https://review.opendev.org/c/openstack/ironic/+/989014 | 10:26 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: [WIP] Add a BMO job https://review.opendev.org/c/openstack/ironic/+/988766 | 11:29 |
| dtantsur | cardoe: (same ping as in requirements) we probably need to revert https://review.opendev.org/c/openstack/ironic/+/985905 until/unless we understand how it randomly breaks local environments | 12:10 |
| cardoe | Like I said in requirements. That doesn’t make sense. That causes -c to be built into the command line execution rather than manually specified arg. | 12:14 |
| dtantsur | Closure: the problem disappeared just as mystically as it appeared | 12:36 |
| cardoe | I don’t like it. It’s gotta be something with the cache. | 12:38 |
| cardoe | I still don’t like it. | 12:42 |
| * dtantsur agrees | 12:45 | |
| dtantsur | Meanwhile, fixing type conversion in BIOS settings broke sushy-tools again... | 12:45 |
| cardoe | I’m just working with stephenfin to keep us (OpenStack) updated on Python packaging and ecosystem stuff. Cause we have a lot of stuff that’s deprecated for a long time or straight up the author has said they no longer support or maintain it. | 12:46 |
| opendevreview | Dmitry Tantsur proposed openstack/sushy-tools master: Preserve types for BIOS settings https://review.opendev.org/c/openstack/sushy-tools/+/989124 | 12:56 |
| iurygregory | good morning ironic | 13:00 |
| iurygregory | dtantsur, funny | 13:01 |
| dtantsur | very ironic, I would say :) | 13:01 |
| iurygregory | always :D | 13:01 |
| stephenfin | cardoe: dtantsur: I think there's a bug in tox that ironic is hitting because it's unique (?) in setting the constrain_package_deps option | 13:12 |
| stephenfin | When `[testenv] constraints` is set, we always append `-c{file}` *except* when installing the package under test https://github.com/tox-dev/tox/blob/10c431ccc/src/tox/tox_env/python/pip/pip_install.py#L201-L202 | 13:13 |
| *** ykarel_ is now known as ykarel | 13:13 | |
| stephenfin | However, when this `constrain_package_deps` option is set, tox will append another `-c{file}` option when installing package dependencies https://github.com/tox-dev/tox/blob/10c431ccc/src/tox/tox_env/python/pip/pip_install.py#L292-L295 | 13:14 |
| stephenfin | The latter `-c{file}` is a local cached file. My guess is that the cached version of that file is getting out-of-sync with the remote version. idk how, because tox (4) is usually very good to keep these files in sync but that's the only thing I can think of | 13:16 |
| stephenfin | The bug needs to be fixed in tox, but you can just sidestep the whole thing by dropping the constrain_package_deps option since it's entirely unnecessary | 13:17 |
| * TheJulia tries to wake up | 13:20 | |
| stephenfin | Okay, looking at it more, the local constraints.txt is not a copy of the remote one. Rather, it's a "fake" constraints.txt derived from the dependencies installed from `deps`. It's there so you can specify e.g. `foo<=8.0` in deps and not have that overridden during package installation if your package or one its dependencies specified e.g. `foo>8.0`. | 13:22 |
| stephenfin | We're hitting that because we have `-r requirements.txt` in deps, so our "fake" constraints.txt file includes all deps of ironic. So the other potential fix is to drop that | 13:23 |
| opendevreview | Mithun Krishnan Umesan proposed openstack/ironic master: Add TLS configuration to agent lookup response https://review.opendev.org/c/openstack/ironic/+/987887 | 13:37 |
| opendevreview | Stephen Finucane proposed openstack/ironic master: tox: Drop conflicting constraints opts https://review.opendev.org/c/openstack/ironic/+/989132 | 13:40 |
| stephenfin | cardoe: dtantsur: That'll fix you | 13:40 |
| opendevreview | Mithun Krishnan Umesan proposed openstack/ironic-python-agent master: Apply TLS configuration from Ironic lookup response https://review.opendev.org/c/openstack/ironic-python-agent/+/988150 | 13:40 |
| dtantsur | stephenfin: thank you very much! | 13:58 |
| dtantsur | Now if someone could solve DRM crashes on my laptop so easily... | 14:00 |
| TheJulia | gpu over heating? | 14:01 |
| dtantsur | kernel regression, I suspect | 14:04 |
| iurygregory | =( | 14:07 |
| TheJulia | I remember it being a very unstable kernel interface ages ago,but it seemed like it got a lot better over the years. | 14:08 |
| cardoe | stephenfin: I don't understand the removal of requirements.txt? | 14:20 |
| opendevreview | Merged openstack/python-ironicclient master: remove pbr as runtime dependency https://review.opendev.org/c/openstack/python-ironicclient/+/989001 | 14:23 |
| stephenfin | cardoe: You don't need it. Per that commit message, tox installs packages in 3 phases. Package dependencies are already handled via phase 2. Specifying them via deps is duplcation | 14:23 |
| cardoe | ohhh cause when we build the package it's baking those dependencies in as part of the package. | 14:24 |
| stephenfin | The only reason we've done that historically was that it was the only way to ensure constraints were applied for package dependnecies without overriding install_command. That's not the case now that tox does constraints properly | 14:24 |
| stephenfin | yup | 14:24 |
| cardoe | neat | 14:24 |
| stephenfin | it'll install tox.ini deps, then pull the deps of the package and install those, and finally it'll install the package itself | 14:24 |
| stephenfin | (this ignores extras and dependency-groups, obviously) | 14:25 |
| stephenfin | You could actually go one step further and move all dependencies into project.dependencies in pyproject.toml and delete requirements.txt. All the release tooling should handle that just fine these days, and pbr is a wrapper of setuptools which means it inherits its support of pyproject.toml-specified deps | 14:28 |
| stephenfin | I'd like to see all projects do that in unison though, which is where my governance proposal for formalising dependency-groups naming comes from | 14:28 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: [WIP] Add a BMO job https://review.opendev.org/c/openstack/ironic/+/988766 | 14:51 |
| dtantsur | TheJulia: do we have a CI job on node.network_data? I'm working on one for Metal3, and it doesn't seem to do anything on the Glean side.. | 14:55 |
| TheJulia | The field itself, no, we have a job in baremetal_stndalone_manager which does config drive injection on rendered data. I thought the prupose was more config injection for your cases and not generally for glean | 14:57 |
| TheJulia | because afaik, it doesn't go to configdrive if also submitted | 14:57 |
| dtantsur | TheJulia: it doesn't, but the Glean situation also doesn't work for me. I see network_data injected, I see our scripts mounting /mnt/config, I see no signs of Glean. | 14:58 |
| opendevreview | Verification of a change to openstack/ironic master failed: Graceful browser shutdown for graphical console https://review.opendev.org/c/openstack/ironic/+/987441 | 15:43 |
| TheJulia | today's goal, put "squirrel" into a commit message ;) | 15:51 |
| TheJulia | no signs of glean reading that? or picking it up. There is a super weird case there which I just don't remember the fine details of | 15:52 |
| opendevreview | Merged openstack/ironic master: tox: Drop conflicting constraints opts https://review.opendev.org/c/openstack/ironic/+/989132 | 16:13 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: [WIP] Add a BMO job https://review.opendev.org/c/openstack/ironic/+/988766 | 16:19 |
| dtantsur | TheJulia: I don't see the word "glean" anywhere in the serial console output | 16:19 |
| TheJulia | got an example job link? | 16:20 |
| dtantsur | TheJulia: grab https://github.com/metal3-io/baremetal-operator/actions/runs/26046449462/artifacts/7064236643, I can help you navigate that | 16:21 |
| dtantsur | ironic logs are in logs/baremetal-operator-system/ironic-service/ | 16:21 |
| dtantsur | serial console in pp-network-data/ | 16:22 |
| TheJulia | so, the build has our handling fixes, but looks like its just missing glean/simple-init/etc entirely | 16:28 |
| TheJulia | The model I wanted take originally was the ironic check/triggers, but folks pushed back on that so we had to keep the default launch path. That should independently be generating some logging regardless of the presence of the config or not. | 16:30 |
| TheJulia | is this the stock ramdisk or other? | 16:31 |
| dtantsur | TheJulia: I *think* it's a stock ramdisk, but I'm checking with Adam | 16:36 |
| TheJulia | k, because if I look at our vmedia jobs, we have glean in the logs. https://de15138de803b0793abc-bb5986a7d7bea17c037d57ac4735bc87.ssl.cf5.rackcdn.com/openstack/6dca99017c0645bd848c286d39eb571c/controller/logs/ironic-bm-logs/node-0_console_2026-05-19-14%3A09%3A42_log.txt but we also may be building that up | 16:37 |
| * TheJulia checks | 16:37 | |
| dtantsur | https://artifactory.nordix.org/ui/native/openstack-remote-cache/ironic-python-agent/dib has a suspicious modification date. I wonder if it's just reported wrongly or it's actually ancient. | 16:38 |
| TheJulia | and it looks like we pull it from tarballs | 16:38 |
| TheJulia | at least for our upstream job | 16:38 |
| TheJulia | yeah, thats suspect | 16:40 |
| dtantsur | We had to use a cache since tarballs.o.o was a source of transient issues.. | 16:40 |
| dtantsur | I'll wait for Adam's response before distracting you further with this | 16:40 |
| TheJulia | ok | 16:40 |
| dtantsur | On the bright side, if I finish this test and manage to run the BMO job on Ironic, we'll get extra coverage for this and other odd corners | 16:40 |
| TheJulia | ++ | 16:40 |
| dtantsur | Like, I've just filed my 5th bug from just the BIOS settings test alone | 16:41 |
| TheJulia | wheeee ? | 16:43 |
| TheJulia | You could always play "Otep" as the theme music for those bugs | 16:43 |
| dtantsur | I need a good soundtrack for sure | 16:51 |
| TheJulia | I started a station on pandora with "Kidneythieves" and it got me into a pretty good sequence of music. | 16:53 |
| TheJulia | unfortunately, pandora is not available in the EU afaik | 16:54 |
| opendevreview | Merged openstack/ironic unmaintained/2024.1: stable only: cleanup 2024.1 ci jobs https://review.opendev.org/c/openstack/ironic/+/988767 | 16:55 |
| dtantsur | A related request. Could someone who is *NOT* a Metal3 developer check the CI docs I'm adding in https://review.opendev.org/c/openstack/ironic/+/989014 and https://review.opendev.org/c/openstack/ironic/+/988766 and tell me if they're usable? | 17:04 |
| dtantsur | (you can use the most recent job runs as a playground) | 17:07 |
| opendevreview | Verification of a change to openstack/ironic master failed: Graceful browser shutdown for graphical console https://review.opendev.org/c/openstack/ironic/+/987441 | 17:31 |
| dtantsur | TheJulia: update for you: the test passed with the upstream IPA image. I'll chase Adam then. Thanks for looking! | 17:31 |
| TheJulia | dtantsur: thanks for the update! | 17:34 |
| Jad_Haj | TheJulia: Hi. implmentation moved to boot.py as you suggedted | 17:44 |
| TheJulia | I noticed :) | 17:47 |
| Jad_Haj | just in case :) | 17:48 |
| TheJulia | I haven't gotten a chance to review it, still curious if there was success | 17:49 |
| Jad_Haj | we usually test fixes after they been brought d/s | 17:50 |
| Jad_Haj | TheJulia: Not sure how to test a not merged u/s fix | 17:50 |
| TheJulia | I just said I was curious, I'm not holding off. And we generally attempt to right size our testing efforts. FWIW. | 17:52 |
| cardoe | dtantsur: I'll read the docs tonight | 17:54 |
| Jad_Haj | TheJulia: Ack, thanks | 17:55 |
| JayF | abongale: Sorry we cancelled the team group review, you are the first non-grosser to ever connect up so I haven't been habitually notifying upstream if it gets canned :) | 18:01 |
| JayF | abongale: show up next time and/or ping me / c i d / c l i f here and we'll make sure it happens even if we were gonna cancel | 18:02 |
| opendevreview | Merged openstack/python-ironicclient master: fix incorrect annotation for list_ports method https://review.opendev.org/c/openstack/python-ironicclient/+/989000 | 18:26 |
| opendevreview | Merged openstack/ironic master: Graceful browser shutdown for graphical console https://review.opendev.org/c/openstack/ironic/+/987441 | 18:38 |
| TheJulia | so w/r/t deferred tasks/actions, I've long wanted to fix the neutron behavior handling, but it looks like johnthetubaguy's change to polling, while sort of invasive, just makes it harder to do. That being said it looks like redfish firmware upgrades is much easier to model into the dferred action flow, at least I hope. | 19:01 |
| cardoe | TheJulia: got an example? | 19:20 |
| TheJulia | for which one? | 19:21 |
| cardoe | john's polling change? | 19:21 |
| TheJulia | oh, sure | 19:21 |
| TheJulia | one moment | 19:22 |
| cardoe | JayF: I can rebase Karan's changes on top of https://review.opendev.org/c/openstack/python-ironicclient/+/989016 but I'll have to drop the one bad one we identified cause it legit double imports. It was on my baddie list but got included on my good list. | 19:22 |
| cardoe | TheJulia can also go crazy with the machete. | 19:22 |
| JayF | I am not contexted up on that at all | 19:22 |
| JayF | I'm happy to approve that change, it looks OK to me if that's the right hting to do? | 19:23 |
| cardoe | So Karan's change had an issue with call-arg cause it double imported things. | 19:23 |
| cardoe | But a bunch of the changes are to clean up typing in dead code. | 19:23 |
| TheJulia | https://github.com/openstack/ironic/commit/876a9d40275d21bf2a7889a2d864877292f18143 | 19:23 |
| cardoe | So I'm just removing the dead code first. | 19:23 |
| JayF | so you want that one approved? | 19:24 |
| cardoe | TheJulia: oh yes. That thing. Yeah we should include that in our ask to make it fail. | 19:24 |
| cardoe | JayF: yah | 19:24 |
| TheJulia | cardoe: ... oh yeah, if we see a hard fail then we should fail really too | 19:24 |
| TheJulia | yeah, one step at a time | 19:25 |
| cardoe | Unfortunately Karan set the types on all objects to the parent object via the typing so the types are currently all wrong | 19:25 |
| cardoe | https://review.opendev.org/c/openstack/python-ironicclient/+/989000 was just one example but I've found a bunch | 19:25 |
| cardoe | so basically that change to save you a click is on "class Node" there's a "list_ports()" method and it returns back a "list[Node]" now. | 19:26 |
| JayF | oh fun :| | 19:26 |
| cardoe | What should have happened is only "list()" returns "list[Node]" and "list_ports()" returns "list[Port]" | 19:26 |
| JayF | yep, that fits | 19:26 |
| TheJulia | THAT BUG! | 19:26 |
| * cardoe hides. | 19:27 | |
| TheJulia | sorry, back in the days of early bifrost it frustrated me | 19:27 |
| TheJulia | no, no hiding! | 19:27 |
| JayF | https://review.opendev.org/c/openstack/ironic/+/988357 should grenade still be running on 2025.1? | 19:29 |
| opendevreview | Merged openstack/python-ironicclient master: Remove dead code in apiclient module https://review.opendev.org/c/openstack/python-ironicclient/+/989016 | 19:33 |
| TheJulia | it can go | 19:34 |
| TheJulia | on 2025.1 that is | 19:34 |
| JayF | that's whatI thought, doing it | 19:37 |
| opendevreview | Jay Faulkner proposed openstack/ironic stable/2025.1: Remove non-voting, obsolete CI jobs https://review.opendev.org/c/openstack/ironic/+/989215 | 19:39 |
| opendevreview | Jay Faulkner proposed openstack/ironic stable/2025.1: [stable-only][ci] Remove non-voting, obsolete jobs https://review.opendev.org/c/openstack/ironic/+/989215 | 19:40 |
| JayF | TheJulia: dtantsur: RFR https://review.opendev.org/c/openstack/ossa/+/988344 (OSSA-2026-013) | 19:40 |
| cardoe | JayF: stop hacking ironic :( | 19:48 |
| TheJulia | muchas gracias | 19:49 |
| JayF | once 2025.1 CI is fived, we need to land 988357 | 19:54 |
| JayF | and cut releases | 19:54 |
| JayF | I will be announcing OSSA-2026-013 now() | 19:54 |
| TheJulia | k | 20:03 |
| JayF | OSSA-2026-013 released; thanks for the patch Julia | 20:10 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: Stop ignoring index in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985241 | 20:11 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: Stop ignoring dict-item in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985242 | 20:11 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: Stop ignoring func-returns-value in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985243 | 20:11 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: Stop ignoring list-item in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985245 | 20:11 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add deferred actions data model, DB API, and versioned object https://review.opendev.org/c/openstack/ironic/+/989224 | 20:24 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add config options, reschedule exception, and action creation helper https://review.opendev.org/c/openstack/ironic/+/989225 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add deferred action launcher periodic and executor https://review.opendev.org/c/openstack/ironic/+/989226 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add deferred action cleanup periodic and stuck recovery https://review.opendev.org/c/openstack/ironic/+/989227 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add notifications for deferred action terminal states https://review.opendev.org/c/openstack/ironic/+/989228 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add metrics gauges and enhanced logging for deferred actions https://review.opendev.org/c/openstack/ironic/+/989229 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Add upgrade check for deferred action backlog https://review.opendev.org/c/openstack/ironic/+/989230 | 20:25 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Decompose redfish firmware stability validation into deferred actions https://review.opendev.org/c/openstack/ironic/+/989231 | 20:25 |
| stevebaker[m] | Hey I'd like to re-enable graphical console testing, but in a way which doesn't require package installs during a container image build. We could do this by using a base image which already has the required packages. The questions for me would be what quay.io repo should this image be published in, how can we keep this image up to date? This would be of interest to cardoe too | 20:41 |
| JayF | might be a good question for #openstack-infra as well | 20:41 |
| cardoe | <insert Jack Nicholson grinning and nodding meme> | 20:42 |
| cardoe | I'd love for us to build that regularly and then just have in that in a way we can consume like IPA | 20:42 |
| stevebaker[m] | JayF: done | 20:49 |
| JayF | I'll note the actual opendev channel is on matrix if you wanna have a matrix-native chat :) I'm not in that one tho | 20:50 |
| stevebaker[m] | oh yeah I'm on matrix | 20:56 |
| opendevreview | Merged openstack/python-ironicclient master: Stop ignoring index in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985241 | 21:20 |
| opendevreview | Merged openstack/python-ironicclient master: Stop ignoring dict-item in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985242 | 21:25 |
| opendevreview | Merged openstack/python-ironicclient master: Stop ignoring func-returns-value in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985243 | 21:45 |
| opendevreview | Merged openstack/python-ironicclient master: Stop ignoring list-item in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985245 | 21:46 |
| opendevreview | Ivan Anfimov proposed openstack/ironic-ui master: Restore explicit package list for module detection https://review.opendev.org/c/openstack/ironic-ui/+/988997 | 21:50 |
| opendevreview | Ivan Anfimov proposed openstack/ironic-ui master: Restore explicit package list for module detection https://review.opendev.org/c/openstack/ironic-ui/+/988997 | 21:51 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: Stop ignoring override in tests https://review.opendev.org/c/openstack/python-ironicclient/+/985237 | 21:57 |
| opendevreview | Doug Goldstein proposed openstack/python-ironicclient master: clean up pyproject.toml deprecated and missing entries https://review.opendev.org/c/openstack/python-ironicclient/+/989241 | 22:15 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: DB-API methods for maintenance entries https://review.opendev.org/c/openstack/ironic/+/989242 | 22:19 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: Node maintenance exception classes https://review.opendev.org/c/openstack/ironic/+/989243 | 22:19 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: Introduce maintenance types https://review.opendev.org/c/openstack/ironic/+/989244 | 22:24 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: Node maintenance exception classes https://review.opendev.org/c/openstack/ironic/+/989243 | 23:33 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: Introduce maintenance types https://review.opendev.org/c/openstack/ironic/+/989244 | 23:33 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: DB-API methods for maintenance entries https://review.opendev.org/c/openstack/ironic/+/989242 | 23:33 |
| opendevreview | cid proposed openstack/ironic master: WIP [LOTO]: Schema for maintenance types https://review.opendev.org/c/openstack/ironic/+/989305 | 23:33 |
| opendevreview | Merged openstack/ironic unmaintained/2024.1: Update .gitreview for unmaintained/2024.1 https://review.opendev.org/c/openstack/ironic/+/965730 | 23:50 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!