| opendevreview | LiZekun proposed openstack/bifrost master: Add pre-commit https://review.opendev.org/c/openstack/bifrost/+/896603 | 06:02 |
|---|---|---|
| opendevreview | LiZekun proposed openstack/bifrost master: Add pre-commit https://review.opendev.org/c/openstack/bifrost/+/896603 | 06:04 |
| rpittau | good morning ironic! o/ | 07:09 |
| kubajj | Good morning rpittau, and Ironic! o/ | 07:17 |
| rpittau | hey kubajj :) | 07:17 |
| opendevreview | Maksim Malchuk proposed openstack/bifrost master: Drop hardcode from the ironic-inspector and optimise defaults https://review.opendev.org/c/openstack/bifrost/+/896168 | 08:29 |
| arne_wiebalck | Good morning, Ironic! | 09:03 |
| arne_wiebalck | Any chance https://review.opendev.org/c/openstack/ironic-python-agent/+/891609 could get some look? | 09:05 |
| dtantsur | morning, putting on my queue for today | 09:07 |
| arne_wiebalck | thanks dtantsur! not super urgent, just wanted to make sure it does not fall through the cracks :) | 09:38 |
| mmalchuk | morning o/ | 09:40 |
| mmalchuk | dtantsur please add to your queue https://review.opendev.org/c/openstack/bifrost/+/896168 too | 09:40 |
| mmalchuk | thanks in advance | 09:40 |
| dtantsur | already there :) | 09:41 |
| mmalchuk | cool | 09:42 |
| iurygregory | good morning Ironic | 10:58 |
| iurygregory | arne_wiebalck, I've added the hashtag ironic-week-prio to give visibility to your patch =) | 11:00 |
| arne_wiebalck | iurygregory: thanks! | 12:13 |
| opendevreview | Merged openstack/ironic master: Update master for stable/2023.2 https://review.opendev.org/c/openstack/ironic/+/896243 | 12:20 |
| TheJulia | good morning | 13:24 |
| iurygregory | good morning TheJulia =) | 13:37 |
| TheJulia | how is everyone this wonderful morning | 13:44 |
| dtantsur | Annoyed by $certain_things, but generally okay. | 13:49 |
| dtantsur | How are you? | 13:50 |
| TheJulia | awake, starting to caffinate | 13:50 |
| TheJulia | Anything I can do w/r/t $certain_things ? | 13:51 |
| * TheJulia might have a mallot someplace | 13:51 | |
| rpittau | hey TheJulia :) | 13:53 |
| iurygregory | I'm tired | 13:53 |
| dtantsur | TheJulia, some old people keep pissing me off. Some new people start pissing me off. There is a lot of piss off for me in the coming months. | 13:54 |
| * TheJulia makes iurygregory coffee | 13:54 | |
| TheJulia | dtantsur: i... see... | 13:54 |
| iurygregory | my neighbor has 5 kittys and yesterday they decided to cry all night | 13:54 |
| mmalchuk | good morning TheJulia ) | 13:54 |
| TheJulia | ugh | 13:54 |
| iurygregory | I only slept 6 hrs and I was waking up from time to time | 13:55 |
| iurygregory | will probably get some sleep in the afternoon, and finish my work at night :D | 13:55 |
| mmalchuk | could you also take a look nice morning patch: https://review.opendev.org/c/openstack/bifrost/+/896168 ? thanks | 13:55 |
| dtantsur | mmalchuk, but do YOU have 4 kitties? | 13:56 |
| dtantsur | sorry, 5 apparently. you got an accidental kitty discount. | 13:56 |
| mmalchuk | 1 kitten, 1 dog, >20 fish | 13:56 |
| rpittau | how does a dog converts to cats ? | 13:57 |
| dtantsur | not bad! | 13:57 |
| dtantsur | pictures of the said animals may make us review things faster ;) | 13:57 |
| mmalchuk | how fish converts to cat) | 13:57 |
| iurygregory | LOL | 13:57 |
| dtantsur | mmalchuk, what if someone already overrides inspector_extra_kernel_options locally? | 13:58 |
| TheJulia | mmalchuk: I'm impressed by >20 fish! | 13:58 |
| iurygregory | in JS you can convert dog to cats and fish to cats | 13:58 |
| TheJulia | Well done! | 13:58 |
| * TheJulia raises an eyebrow and wonders if iurygregory needs more rest | 13:58 | |
| TheJulia | :) | 13:58 |
| mmalchuk | TheJulia 4 aquariums | 13:58 |
| iurygregory | TheJulia, I will get some rest after lunch =) | 13:59 |
| mmalchuk | dtantsur no problem look deeper the code | 13:59 |
| dtantsur | mmalchuk, questions/comments left. | 14:01 |
| mmalchuk | thanks, will see | 14:01 |
| opendevreview | Harald Jensås proposed openstack/metalsmith stable/yoga: Set role default metalsmith_provisioning_timeout https://review.opendev.org/c/openstack/metalsmith/+/893957 | 14:42 |
| opendevreview | Merged openstack/ironic-inspector stable/2023.1: Handle bracketed IPv6 redfish_address https://review.opendev.org/c/openstack/ironic-inspector/+/895906 | 14:49 |
| JayF | rpittau: dogs and cats are just measured by weight; so if you have a big dog like me, he's like at least 5 cats worth of trouble | 14:57 |
| dtantsur | LOL | 14:59 |
| dtantsur | Cats are black holes, they have infinite troubles regardless of their size | 14:59 |
| TheJulia | Variable gravity wells.... I often wonder what magic they possess! | 15:00 |
| rpittau | :D | 15:03 |
| TheJulia | brraaaaains | 15:14 |
| TheJulia | when did it become 10 AM?!? | 15:14 |
| iurygregory | 15min ago | 15:14 |
| iurygregory | almost lunch time for me :D | 15:15 |
| rpittau | mmm almost ice-cream time here | 15:16 |
| TheJulia | heh | 15:18 |
| iurygregory | it's always ice-cream time | 15:21 |
| iurygregory | you don't need a specific time rpittau :D | 15:21 |
| TheJulia | iurygregory has an excellent point! | 15:23 |
| rpittau | oh I just set a specific time to avoid eating it continuously :D | 15:24 |
| TheJulia | ahh! | 15:24 |
| TheJulia | Likely a wise plan! | 15:24 |
| iurygregory | ok, that makes sense :D | 15:27 |
| rpittau | with this sweet note I'll wish you a great end of the week, see you next monday! o/ | 15:28 |
| TheJulia | excellent! Have a wonderful weekend! | 15:29 |
| dtantsur | It's incredibly annoying that network_data is not documented anywhere.. should we just document it in ironic? | 15:30 |
| JayF | It's documented in a nova spec, that's the only official docs for it aiui | 15:30 |
| TheJulia | The schema? | 15:30 |
| TheJulia | We... should likely augment/mirror if the comment is the schema | 15:30 |
| JayF | It seems almost like that should be in API docs for Ironic since we take --network-data | 15:30 |
| dtantsur | yep | 15:31 |
| JayF | which is the correct approach since we're reflecting what *we* support and make no promises about other stuff | 15:31 |
| TheJulia | ++ | 15:31 |
| opendevreview | Merged openstack/ironic-specs master: Migrate completed specs to release version https://review.opendev.org/c/openstack/ironic-specs/+/896597 | 15:44 |
| JayF | Hey, just letting you all know we're going to be having an MLH fellow, Boushra, joining us soon (this week) and working on Ironic through the first week of December. | 15:46 |
| JayF | This is a fellowship organized by G-Research OSS team, so if there are any issues or concerns please direct them my way. Thank you! | 15:47 |
| dtantsur | nice! | 15:47 |
| JayF | After doing an orientation trip through devstack, her first project will be getting proper support for bugfix branch release/cut/retirement in release automation. | 15:47 |
| dtantsur | FYI Mahnoor and I are not planning to mentor an ironic project in outreachy this time. If someone wants to - please talk to me soon. | 15:48 |
| dtantsur | JayF++ | 15:48 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 15:59 |
| TheJulia | i sincerely hope that things just work | 15:59 |
| * TheJulia sets hope to maximum | 15:59 | |
| TheJulia | Has anyone looked at the weird ipmi failures we've had appear as of recent? | 16:19 |
| JayF | I haven't looked at it; tbh I am a bit overwhelmed with 'soft' non-tech work right now and trying to get through it | 16:21 |
| TheJulia | ack | 16:21 |
| TheJulia | challenge is I think it might be pyghmi | 16:21 |
| TheJulia | .... I'm wondering if we need to just dial back ipmi anyway | 16:21 |
| JayF | oh as in, virtualbmc itself is becoming less reliable? | 16:22 |
| TheJulia | dunno, it just seems super weird | 16:23 |
| JayF | well I'm thinking two things: 1) u-c or 2) python version | 16:23 |
| JayF | if it's pgyhmi probably one of those things broke it and might be trackable via stats | 16:23 |
| arne_wiebalck | thanks for the reviews kubajj rpittau and JayF ! | 16:23 |
| TheJulia | Looks like jared has started to wire some redfish stuffs into pyghmi too 8| | 16:25 |
| JayF | arne_wiebalck: https://bugs.launchpad.net/ironic-python-agent/+bug/2037572 for the followup; but feel no obligation to implement it | 16:26 |
| JayF | arne_wiebalck: I think it's extremely valuable to have low-hanging-fruit stuff to onboard people with; this is not urgent and super straightforward and unit tested | 16:26 |
| JayF | so it's basically the ideal low hanging fruit bug :D | 16:26 |
| JayF | I'm not going to be upset if you fix it; just letting you know I have zero expectation | 16:27 |
| TheJulia | Jared also seems to be trying to keep python2 support | 16:27 |
| JayF | that is ... a choice indeed | 16:28 |
| JayF | I learned something the other day: an Ironic project has a python 2.7 supporting branch | 16:28 |
| JayF | that stable/1.0 branch of tenks is for py2.7 support per mgoddard | 16:29 |
| * TheJulia twitches | 16:33 | |
| hjensas | TheJulia: | 16:37 |
| hjensas | :stderr: ERROR: Invalid requirement: 'openstacksdk=0.99.0' | 16:37 |
| hjensas | Hint: = is not a valid operator. Did you mean == ? | 16:37 |
| JayF | = != == ? | 16:37 |
| hjensas | ^^ error on - https://review.opendev.org/c/openstack/metalsmith/+/896684 | 16:38 |
| TheJulia | The example in the ansible docs is name=version | 16:39 |
| TheJulia | but there is a separate flag | 16:39 |
| TheJulia | I guess to try == next :) | 16:39 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 16:59 |
| JayF | TheJulia: dtantsur: rpittau: Would you all be +1 to updating our bugfix branch policy to explicitly state we never release from bugfix branches? | 17:44 |
| TheJulia | + would be | 17:44 |
| JayF | TheJulia: dtantsur: rpittau: I'm fairly certain if we do that, we can avoid any scoping questions with releases team (we just update their automation to retire/eol branches and don't worry about releases) | 17:44 |
| TheJulia | err, +1 | 17:44 |
| TheJulia | well, just forcing the sdk into place didn't work | 17:49 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 18:04 |
| TheJulia | I'm fairly sure that is not permitted in requirements, but maybe we make a separate file, dunno right now | 18:05 |
| TheJulia | woot https://review.opendev.org/c/openstack/ironic/+/885087 passed cI | 18:08 |
| opendevreview | Verification of a change to openstack/ironic-python-agent master failed: Conditional creation of RAIDed ESP for UEFI Software RAID https://review.opendev.org/c/openstack/ironic-python-agent/+/891609 | 18:11 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 18:17 |
| * TheJulia sits on the edge of the chair and hopes | 18:39 | |
| JayF | I've spent the last 90 minutes doing a big brain dump into an etherpad for my MLH fellow | 18:41 |
| JayF | you never think you know that much about a topic, then you start writing and all the sudden it's like a 3 page essay | 18:41 |
| JayF | (not really an essay in this case; but I'm amazed at the amount of context there is around so much that we do) | 18:41 |
| TheJulia | Okay, that works, we're going to need to disable the jobs and we can fix them in the patch to fix the contents | 19:07 |
| TheJulia | easy peasy | 19:07 |
| opendevreview | Verification of a change to openstack/ironic-python-agent master failed: Add mlnx deploy_step entry to enable deploy time firmware https://review.opendev.org/c/openstack/ironic-python-agent/+/893371 | 19:18 |
| *** JayF is now known as Guest1505 | 20:10 | |
| *** JasonF is now known as jayf | 20:10 | |
| *** jayf is now known as JayF | 20:11 | |
| opendevreview | Julia Kreger proposed openstack/ironic master: Reset parent_node values to uuids instead of names https://review.opendev.org/c/openstack/ironic/+/889750 | 20:40 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 20:47 |
| TheJulia | I suspect that should do the magic, now lets see | 20:48 |
| opendevreview | Julia Kreger proposed openstack/ironic-specs master: Add HTTPBoot support https://review.opendev.org/c/openstack/ironic-specs/+/896474 | 21:07 |
| TheJulia | .... | 21:27 |
| TheJulia | .... the whole outcome of ansible and openstacksdk is just a nightmare | 21:28 |
| JayF | hard to disagree with that :( | 21:39 |
| TheJulia | I’m sitting here trying to figure a path forward and all I can think is downgrading ansible forcibly and hoping or just installing latest ack | 21:42 |
| JayF | I guess my confusion is that if this is version-related, shouldn't that stable/zed version of MS have the versions locked? | 21:43 |
| JayF | Are we trying to track current ansible support in a stable branch and that's the pain? or is SDK version not constrained? | 21:43 |
| JayF | I don't have context on why you're digging here, I'm sorry if it's been posted before but I have been slammed busy | 21:43 |
| TheJulia | We track what we get from Zuul | 21:43 |
| TheJulia | So basically, we get what we’re handed | 21:44 |
| JayF | metalsmith jobs use zuul ansible | 21:44 |
| JayF | zuul ansible for openstack upgraded recently | 21:44 |
| TheJulia | And don’t have options there | 21:44 |
| TheJulia | At least, that I’m aware of | 21:44 |
| JayF | so you can force old version, and hope infra isn't going to retire it before stable/zed is retired, as one option | 21:44 |
| TheJulia | The Sdk is constrained on the stable branch | 21:44 |
| TheJulia | Which is how we end up with 0.103.0 | 21:44 |
| TheJulia | Well, we are already in the play’s execution, so I’m not sure there is a way to swap it out | 21:45 |
| JayF | Yeah, this all makes sense. The only fixes that seem to fit with actually keeping the deliverable working as shipped + keeping CI working would be: force zuul ansible version lower for those jobs or somehow refactor the job to use metalsmith-defined ansible versions | 21:45 |
| TheJulia | And even then, it is the callers ansible, not what is present on the cli | 21:45 |
| JayF | TheJulia: there's a zuul config option to flip the ansible version used for the job | 21:45 |
| TheJulia | Ohhh ahhhh | 21:46 |
| JayF | yeah, to give a release valve for this kinda break | 21:46 |
| TheJulia | If you know the name or got a link, that would be awesome | 21:46 |
| JayF | clarkb: are there plans to remove Ansible N-1 version support from Zuul? | 21:46 |
| * TheJulia checks in with contractor | 21:46 | |
| * JayF looks for details on how that is structured | 21:46 | |
| JayF | https://github.com/openstack/project-config/blob/master/zuul/main.yaml#L132 is where the default is set, now 8 was 7 | 21:48 |
| JayF | I gotta find how the override is done | 21:48 |
| JayF | scratch that, I think it was 6 now is 8 | 21:49 |
| JayF | 6 is at least the documented other option | 21:49 |
| JayF | TheJulia: https://zuul-ci.org/docs/zuul/latest/config/job.html#attr-job.ansible-version | 21:49 |
| JayF | we may want clarkb to confirm that they aren't going to pull ansible 6 support anytime soon | 21:49 |
| JayF | but I suspect we'll need to somehow figure out how to decouple ansible versions in jobs, which is likely going to be gross (I suspect there *has* to be prior art in this direction from the rest of openstack) | 21:50 |
| JayF | maybe another item to add to the "What about metalsmith?" item for vPTG | 21:50 |
| clarkb | JayF: the plan is to do so soonish iirc. Everything is currently ansible 8 by default right now in zuul | 21:58 |
| JayF | clarkb: that migration broke openstack/metalsmith stable jobs, because it uses ansible from zuul, apparently | 21:59 |
| clarkb | the problem is ansible has a very quick release then eol cadence and we're trying to avoid dragging along ancient versions more than necessary | 21:59 |
| JayF | clarkb: and because of the openstacksdk breaking changes against ansible, we can't really support newer ansible in metalsmith without significantly changing the deliverable | 21:59 |
| clarkb | (ancient in ansible terms) | 21:59 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 21:59 |
| clarkb | JayF: you should use nested ansible if the version of ansible mattres to you then | 21:59 |
| TheJulia | JayF: well, we may not have to change the deliverable | 22:00 |
| TheJulia | we may just need to change the testing to work and be functional | 22:00 |
| JayF | clarkb: I have that as a line item to discuss re: metalsmith for our PTG; but do you know of prior art we could reference for that? | 22:00 |
| TheJulia | time will tell | 22:00 |
| clarkb | if you need exampels of this look at the opendev/system-config system-config-run-* jobs we do that for all of the opendev stuff because opendev production management wants to control that | 22:00 |
| JayF | clarkb: to be explicit; like this? https://opendev.org/opendev/system-config/src/branch/master/zuul.d/system-config-run.yaml#L109 | 22:02 |
| JayF | seems like: 1) put ansible and modules in required projects with override checkout and install_ansible_collections as required as well (?) | 22:02 |
| TheJulia | JayF: that is code for the job, not code for execution, ansible version where executed is asserted/used on the remote host | 22:03 |
| JayF | I don't understand how that could be the case when this is also the syntax we'd use for ironic versions that are running inside the node? | 22:04 |
| * JayF is missing something | 22:04 | |
| JayF | like, are we using the distro ansible from the image or something? no b/c if that was the case, it wouldn't have changed (?) | 22:04 |
| TheJulia | we don't actually execute the ansible installed locally | 22:04 |
| clarkb | JayF: no that example is if you want to test ansible devel which you don't want you just want old releases. | 22:04 |
| TheJulia | it is part of the mechanics of how ansible actually works | 22:05 |
| JayF | TheJulia: I am like a 0.5/10 on ansible mechanics; it's not something I've ever used professionally outside of bifrost | 22:05 |
| clarkb | JayF: we bootstrap our bastion https://opendev.org/opendev/system-config/src/branch/master/playbooks/bootstrap-bridge.yaml#L20-L22 which includes an ansible isntall | 22:05 |
| clarkb | https://opendev.org/opendev/system-config/src/branch/master/playbooks/roles/install-ansible/tasks/main.yaml that role installs an ansible and configures it then later in the jobs we can execute ansible from that location | 22:06 |
| clarkb | its no different than say running tox in a test job | 22:06 |
| JayF | assuming that works (I don't understand what TheJulia is concerned about, but I trust she knows more bits about this than I do by a lot); that's a hell of a lot easier than I would've expected | 22:07 |
| JayF | thank you | 22:07 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Reset parent_node values to uuids instead of names https://review.opendev.org/c/openstack/ironic/+/889750 | 22:29 |
| TheJulia | hmm.. wondering if my most recent metalsmith change is doing the needful | 22:47 |
| TheJulia | err, nope. | 22:48 |
| TheJulia | oh, it is out of order, what was I thinking | 22:53 |
| opendevreview | Julia Kreger proposed openstack/metalsmith stable/zed: Stable-Only: CI: Force install an appropriate openstacsdk version https://review.opendev.org/c/openstack/metalsmith/+/896684 | 22:55 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!