| opendevreview | Merged openstack/ironic master: Log when a periodic is completed https://review.opendev.org/c/openstack/ironic/+/888503 | 00:13 |
|---|---|---|
| opendevreview | Merged openstack/ironic master: DB: Select upon delete for allocations https://review.opendev.org/c/openstack/ironic/+/889890 | 00:38 |
| opendevreview | likui proposed openstack/python-ironicclient master: These should be indented 4 spaces to match the other things in this block https://review.opendev.org/c/openstack/python-ironicclient/+/794691 | 07:03 |
| opendevreview | likui proposed openstack/python-ironicclient master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/python-ironicclient/+/890050 | 07:16 |
| rpittau | good morning ironic! o/ | 07:16 |
| opendevreview | likui proposed openstack/ironic master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic/+/890051 | 07:19 |
| opendevreview | likui proposed openstack/ironic-inspector master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-inspector/+/890052 | 07:25 |
| opendevreview | likui proposed openstack/ironic-python-agent master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-python-agent/+/890053 | 07:26 |
| opendevreview | likui proposed openstack/ironic-prometheus-exporter master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/890054 | 07:28 |
| kubajj | Good morning Ironic o/ | 07:38 |
| masghar | o/ | 08:09 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: [WIP] Build tinyipa with tinycore 14.x https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/887754 | 08:11 |
| iurygregory | good morning Ironic | 11:19 |
| dtantsur | Oh, how much I love turning off my IRC bouncer during vacation: no pings, no scrollback to read! Looking at you, Slack!! | 11:20 |
| dtantsur | In other words, good morning folks, I'm back :) | 11:20 |
| kubajj | dtantsur: good morning | 11:23 |
| dtantsur | hey-hey kubajj, long time no see | 11:23 |
| kubajj | dtantsur: Yeah, exams got me a bit busy and then volunteered at a scout camp but now I'm back | 11:24 |
| dtantsur | \o/ | 11:26 |
| opendevreview | Jacob Anders proposed openstack/ironic master: [WIP] Wait for secure boot state change if it's not immediate https://review.opendev.org/c/openstack/ironic/+/863999 | 11:28 |
| kubajj | dtantsur: I'll be working on ironic part-time, mostly looking into https://specs.openstack.org/openstack/ironic-specs/priorities/2023-2-workitems.html#improving-deploy-kernel-ramdisk-config but can also help with anything that could help with the inspector | 11:28 |
| dtantsur | I think the kernel/ramdisk is a good and useful action items. | 11:29 |
| dtantsur | I need to wrap my head around the recent Inspector development myself (need to sync with masghar) | 11:29 |
| opendevreview | Merged openstack/ironic master: Add the initial skeleton of the agent inspect interface https://review.opendev.org/c/openstack/ironic/+/877814 | 11:51 |
| JayF | kubajj: you know, you can loudly state to dtantsur and others that you work with me :D kubajj is part of the GR-OSS team now helping us out for a while | 14:14 |
| dtantsur | w000t! | 14:14 |
| JayF | dtantsur: in the realm of things you might know/care about | 14:14 |
| JayF | dtantsur: I might have an opportunity for an MLH intern (think: outreachy but different brand and coming thru GR-OSS channels, not openstack) | 14:15 |
| JayF | dtantsur: Curious if you had ideas on a project, because I wanted to nail that down before I said yes :D | 14:15 |
| opendevreview | Dmitry Tantsur proposed openstack/ironic master: Fix minor grammar issues in the help for new inspector options https://review.opendev.org/c/openstack/ironic/+/890138 | 14:16 |
| dtantsur | JayF: ^^^ | 14:16 |
| dtantsur | follow-up you asked for (funnily, these constants are not used until the next patch) | 14:16 |
| JayF | yeah I just didn't want the grammar fix to get lost; +2 | 14:16 |
| dtantsur | JayF: how ambitious should the project be? I have quite a few ideas (or at least I had before my brain was reset during the vacation lol) | 14:17 |
| dtantsur | all the way up to IPv6 support in Bifrost ;) | 14:17 |
| JayF | dtantsur: you've worked with that style of intern before; I usually try to make the project easily achievable and a gateway to further stuff if they knock it out | 14:17 |
| dtantsur | to rephrase my question: which level will the interns have and how much time (both per week and totally) are they going to spend? | 14:18 |
| JayF | you know, bifrost is fertile ground | 14:18 |
| JayF | v6 support is probably very fiddly for an intern | 14:18 |
| JayF | maybe similar for neutron integration | 14:18 |
| dtantsur | BIOS settings / RAID support. Actually, any manual cleaning support in the Bifrost playbooks. | 14:20 |
| dtantsur | Someone recently complained that we don't support node.network_data (as opposed to configdrive) in Bifrost. Can be a small "try it" work item. | 14:21 |
| JayF | plus I feel less bad tossing someone at a bifrost dev item | 14:21 |
| JayF | because the dev env is less painfail | 14:21 |
| JayF | you know, that's the most hilariously perfect typo of painful to describe devstack | 14:21 |
| dtantsur | :D | 14:21 |
| dtantsur | JayF: something people keep asking about: support for symlinks in root device hints | 14:22 |
| dtantsur | so that they can use /dev/disk/by-whatever/whatever instead of /dev/sda | 14:22 |
| JayF | I was surprised at first we didn't support that | 14:23 |
| JayF | then I remembered cleaning would change many identifiers there | 14:23 |
| JayF | UUID/PARTUUID/LABEL | 14:23 |
| dtantsur | the old item: https://storyboard.openstack.org/#!/story/2009814 | 14:23 |
| dtantsur | yeah, there are caveats | 14:23 |
| dtantsur | https://storyboard.openstack.org/#!/story/1718726 needs fixing if not already | 14:23 |
| JayF | 2009814 is a good idea, too | 14:24 |
| JayF | I think 2009814 or one of those bifrost enhancements | 14:24 |
| JayF | either way, thanks for giving me confidence we have enough low-hanging items | 14:24 |
| dtantsur | yeah, I have an enormous TODO list in my personal board under "Ideas and Low Priority" aka stuff I'll never get to :D | 14:25 |
| dtantsur | JayF: a pretty ambitious task would be to move ironicclient to openstacksdk internally | 14:26 |
| dtantsur | I think I should stop at this point :) | 14:26 |
| JayF | yeah I'm looking for not-ambitious | 14:26 |
| JayF | I'm looking for get the dev env to hello world | 14:26 |
| JayF | take one more step | 14:26 |
| JayF | great success | 14:26 |
| JayF | then we can add ambition | 14:26 |
| JayF | lol | 14:26 |
| dtantsur | Oh, last one. My personal pet peeve. Get rid of that stupid API for setting maintenance and make it normally updateable :) | 14:27 |
| dtantsur | it's API work, but it's not super sophisticated | 14:28 |
| JayF | s/make it noramlly updatable/and remove the vague concept which causes more confusion than anything else/ | 14:28 |
| dtantsur | heh | 14:29 |
| JayF | if I could go back in time and change anything about original ironic | 14:29 |
| JayF | it'd be maintenance | 14:29 |
| JayF | just make it online:bool just like the old school printers with an on-line light and button lol | 14:29 |
| dtantsur | I would have changed many things, but yeah. We've overloaded this one too much. | 14:31 |
| JayF | you only get one, what is it | 14:31 |
| dtantsur | free-form JSON fields | 14:32 |
| dtantsur | (except for node.extra, of course) | 14:32 |
| JayF | node.junk_drawer['lol' | 14:32 |
| JayF | :) | 14:32 |
| dtantsur | another dream of my is to gradually decompose and get rid of them | 14:32 |
| dtantsur | yeah | 14:32 |
| JayF | I think in a weird way, those fields were a strength early on | 14:33 |
| dtantsur | properties could be a good start | 14:33 |
| JayF | because it allowed us to express things in our data model we didn't understand when it was written | 14:33 |
| dtantsur | aka technical debt ;) | 14:33 |
| JayF | #startmeeting ironic | 15:00 |
| opendevmeet | Meeting started Mon Jul 31 15:00:05 2023 UTC and is due to finish in 60 minutes. The chair is JayF. 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 |
| iurygregory | o/ | 15:00 |
| rpittau | o/ | 15:00 |
| kubajj | o/ | 15:00 |
| masghar | o/ | 15:00 |
| JayF | Good morning. Welcome to the Ironic meeting. Reminder this is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct | 15:00 |
| JayF | #topic Announcements/Reminder | 15:01 |
| JayF | #note If you have a patch ready for review; hashtag it with ironic-week-prio; and please use our review dashboard to prioritize patches tagged as such. | 15:01 |
| JayF | #note next PTL is 10/23-27 | 15:01 |
| JayF | I'll say generally, Bobcat release timeline is in the agenda. I'll only mention if we have a milestone coming up | 15:02 |
| JayF | next one is 8/24 for non-client library freeze | 15:02 |
| JayF | #topic Review action items from last meeting | 15:02 |
| JayF | pretty sure there were none; checking | 15:02 |
| * dtantsur is curious if "next PTL" is a typo or a prediction | 15:02 | |
| JayF | confirmed; no actions last meeting omving on | 15:02 |
| JayF | dtantsur: haha, typo for sure | 15:03 |
| JayF | dtantsur: but I will say I've been nudging rpittau that it's his turn coming up soon ;) | 15:03 |
| * rpittau hides | 15:03 | |
| JayF | #topic Review Ironic CI Status | 15:03 |
| dtantsur | the clock is ticking, rpittau! | 15:03 |
| JayF | How are things? They look OK? At least relative to the horrible place we were a month ago... | 15:03 |
| rpittau | less horrible than usual | 15:03 |
| iurygregory | need to agree, I didn't look that much during last week.. | 15:04 |
| JayF | Keep an eye on CI. If it gets too outta whack mention in here, and put it in the whiteboard | 15:04 |
| JayF | for now I assume things are happier | 15:04 |
| JayF | #topic Review ongoing 2023.2 workstreams | 15:04 |
| JayF | As a reminder if you're newer reading this log or in meeting, we have a doc we publish at the start of the cycle | 15:05 |
| JayF | which is entirely hopeful about features we wanna work on | 15:05 |
| JayF | when we do work on them, we ask contributors working on this to put a very high levl update here | 15:05 |
| JayF | #link https://etherpad.opendev.org/p/IronicWorkstreams2023.2 | 15:05 |
| JayF | tbh we've made quite a dent into our priorities list | 15:06 |
| JayF | we might actually complete most of the things \o/ | 15:06 |
| dtantsur | \o/ | 15:07 |
| JayF | There are no RFEs to review, skipping on to... | 15:07 |
| JayF | #topic Open Discussion | 15:07 |
| JayF | Floor is open if anyone has anything | 15:07 |
| JayF | Last call? | 15:09 |
| rpittau | I've struggling a bit understanding why the new tinycore 14.x is failing on standalone jobs on uefi | 15:09 |
| rpittau | it looks like efibootmgr is failing and the issue seems related to efivars moduile not present | 15:09 |
| rpittau | we're not in a rush with that, but if anyone has ideas on how to fix that or how to troubleshoot further help is appreciated | 15:10 |
| JayF | efivar is a separate binary, not a kernel module though | 15:10 |
| JayF | can you link the exact err? | 15:10 |
| JayF | We will troubleshoot this with the logger off :) TY all | 15:11 |
| JayF | ] | 15:11 |
| rpittau | efivar is a separate binary, but efivars is indeed a module | 15:11 |
| JayF | #endmeeting | 15:11 |
| opendevmeet | Meeting ended Mon Jul 31 15:11:07 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:11 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-07-31-15.00.html | 15:11 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-07-31-15.00.txt | 15:11 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/ironic/2023/ironic.2023-07-31-15.00.log.html | 15:11 |
| JayF | rpittau: not one I have loaded | 15:11 |
| JayF | rpittau: lemme see if it's built i | 15:11 |
| JayF | *ikn | 15:11 |
| iurygregory | maybe the module changed the name or we need something extra? <just thinking out loud> | 15:11 |
| rpittau | probably | 15:11 |
| JayF | CONFIG_EFIVAR_FS=y | 15:11 |
| JayF | likely all that's needed | 15:11 |
| JayF | or the module that comes out if =m | 15:11 |
| rpittau | it could be a change in the kernel | 15:12 |
| rpittau | I tried with efivars and efivarfs but no luck | 15:12 |
| rpittau | this is the lat result https://zuul.opendev.org/t/openstack/build/463f1cfac2d0420e9e1a3b7a1c258c7d | 15:12 |
| rpittau | adn I found this https://michael-prokop.at/blog/2021/06/09/efivars-is-gone-with-debian-bullseye-newinbullseye/ | 15:12 |
| rpittau | I will give that a try since the kernel seems compatible | 15:13 |
| iurygregory | interesting | 15:13 |
| JayF | efivarfs is the module name | 15:13 |
| JayF | yeah that blog looks correct | 15:13 |
| opendevreview | Riccardo Pittau proposed openstack/ironic-python-agent-builder master: [WIP] Build tinyipa with tinycore 14.x https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/887754 | 15:15 |
| opendevreview | Merged openstack/ironic-prometheus-exporter master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-prometheus-exporter/+/890054 | 15:36 |
| opendevreview | Merged openstack/python-ironicclient master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/python-ironicclient/+/890050 | 15:36 |
| * dtantsur has forgotten that updating his NAS involves bring the IRC bouncer offline | 15:39 | |
| opendevreview | Merged openstack/ironic-python-agent master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-python-agent/+/890053 | 15:39 |
| dtantsur | oh, and now I need to update ZNC itself, so offline again :D | 15:39 |
| dtantsur | So, folks, with virtual media support growing.. WDYT about promoting virtual media attach/detach to the top-level API? | 15:43 |
| JayF | Can you help with a use case? | 15:44 |
| dtantsur | I'm aware of two actually. 1st seems pretty common: custom installers. Ironic is pretty decent in this regard: you can connect a CD, boot from it (aka ramdisk deploy), then set boot device to disk. BUT you end up with the CD still connected. | 15:45 |
| JayF | dtantsur: For that use case; we should genericize the kickstart driver to provide that value to all hardware IMO | 15:45 |
| dtantsur | 2nd is something new for me, but people want to be able to just connect/disconnect a thing to/from a machine. | 15:45 |
| JayF | #2 is the case I was thinking of the whole time | 15:45 |
| dtantsur | Think, provide some configuration. | 15:45 |
| dtantsur | I'm not sure how the kickstart drive helps with custom installers.. | 15:46 |
| JayF | I'm saying the kickstart driver right now | 15:46 |
| JayF | is boot an install ramdisk + some metadata | 15:46 |
| JayF | custom installers are: boot an install ramdisk + some metadata | 15:46 |
| dtantsur | how is it different from the ramdisk deploy? | 15:47 |
| JayF | Does the kickstart deploy allow us to pass arbitrary stuff to the booted ramdisk? | 15:47 |
| JayF | s/kickstart/ramdisk/ | 15:47 |
| JayF | e.g. ks=https://server/my.ks | 15:47 |
| dtantsur | Neither do? | 15:47 |
| JayF | except probably more like preseed=https://server/my.preseed | 15:48 |
| dtantsur | it's not arbitrary | 15:48 |
| JayF | I'm saying if we take ramdisk interface + some of the bits added on top for kickstart driver | 15:48 |
| dtantsur | not everything is done through kernel parameters, but the ramdisk deploy does support kernel parameters | 15:48 |
| opendevreview | Merged openstack/ironic master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic/+/890051 | 15:48 |
| opendevreview | Merged openstack/ironic stable/victoria: [stable-only] Cap virtualenv/setuptools https://review.opendev.org/c/openstack/ironic/+/888701 | 15:48 |
| JayF | OK; so we're basically there then | 15:48 |
| opendevreview | Merged openstack/ironic-inspector master: Add python3.10 support in testing runtime https://review.opendev.org/c/openstack/ironic-inspector/+/890052 | 15:48 |
| JayF | you just have to self-host your preseed file | 15:48 |
| dtantsur | JayF: it does not anyhow remove the need for disconnecting the ISO | 15:48 |
| JayF | that's... true | 15:49 |
| JayF | same with kickstart, actually | 15:49 |
| JayF | heh | 15:49 |
| JayF | custom installer deploy steps can now end with hold | 15:49 |
| JayF | and you can call resume on the node when the install is done | 15:49 |
| JayF | to get the cleanup actions done, like removing the disk | 15:49 |
| JayF | attach ISO -> change boot device -> boot an installer -> hold -> change boot device to local -> detach iso | 15:49 |
| JayF | (BTW; I'm sold on the API, this is just a nice thought experiment to walk down) | 15:50 |
| JayF | use cases for julia's new hold step | 15:50 |
| dtantsur | I totally agree, and I see value in both | 15:50 |
| dtantsur | the trigger for my question were people with "hey, can we use metal3/ironic to connect a CD with configuration" | 15:51 |
| dtantsur | I'm trying to move our people away from the ramdisk deploy :D | 15:51 |
| JayF | I suspected as mnuch, which is why I started asking more use case questions | 15:51 |
| dtantsur | (so, case #2 more than #1 for me personally) | 15:51 |
| JayF | I feel like a "just attach a thing and boot a server" might have people use Ironic more like a mechanical turk than a full orchestrator | 15:51 |
| dtantsur | (but I do know that case #1 is popular, also outside of Red Hat) | 15:51 |
| dtantsur | well.. remember we discussed the dual nature of our API before my PTO? | 15:51 |
| JayF | like, we can't both be the primary orchestration AND give all the hooks to the outside world for us to be the target of orchestration | 15:51 |
| JayF | ...can we? | 15:52 |
| dtantsur | you literally argued that we can :D | 15:52 |
| dtantsur | it was about having stuff like set_boot_device in the API | 15:52 |
| JayF | hmm | 15:52 |
| JayF | I just realized you might be the only person who would get onboard with my insaneo idea | 15:52 |
| JayF | I tried to sell Julia on it IRL and got furrowed brows back as a response | 15:52 |
| JayF | Ironic could potentially add a lot of value to the bare metal ecosystem | 15:53 |
| JayF | if we split our service from the library pieces in the back | 15:53 |
| JayF | so that other tools could use our *_interfaces directly as a libraryu | 15:53 |
| dtantsur | "Can we have Ironic as a library" was roughly the 1st question we got when Metal3 started | 15:53 |
| dtantsur | The fatal roadblock was Python | 15:53 |
| JayF | I mean, yeah, it'd be in python | 15:54 |
| JayF | I'm not saying just for metal3 use case, to be clear | 15:54 |
| JayF | can you imagine the value of | 15:54 |
| JayF | ironit --redfish=http://myserverbmc power on | 15:54 |
| dtantsur | That being said, yes, I can see value of that, and I've been toying with this idea in my head for at least 2 years | 15:54 |
| JayF | (yes, iron - it, not ironic) | 15:54 |
| dtantsur | The amount of work is enormous. We now don't distinguish between "hardware interfaces" (power, management) and "workflow interfaces" (deploy, inspect, your future clean). | 15:56 |
| dtantsur | Stuff like BIOS/RAID is theoretically "hardware only" but in practice relies on rebooting into IPA | 15:56 |
| dtantsur | etc etc | 15:56 |
| JayF | yeah, factoring agent out is probably a huge % of the difficulty | 15:56 |
| JayF | just a thought though | 15:57 |
| dtantsur | and TaskManager | 15:57 |
| JayF | I want to make sure that even if ... Ironic's view of the world doesn't win over the long term | 15:57 |
| JayF | that the work we've done to make these things easier to provision is in a place where others can use it and not have to start over from scratch | 15:57 |
| JayF | that's the real motivation; for me personally; behind this idea | 15:57 |
| dtantsur | Sushy is part of this effort | 15:57 |
| dtantsur | Maybe we should move more fishy logic there | 15:58 |
| dtantsur | Maybe we should finish the late Ilya's sushy-cli | 15:58 |
| JayF | sushy-cli | 15:58 |
| JayF | plus moving some of our old IPMI magic into a library | 15:58 |
| JayF | would probably be 50% of the value with 10% of the work | 15:58 |
| dtantsur | Possibly, but I'd focus on Redfish since 1) it's the future, 2) it requires much more hacking (who could imagine!) | 15:59 |
| JayF | the problem I have is that while this is really cool | 16:00 |
| JayF | zero of this use case crosses over with GR | 16:00 |
| JayF | or is even in the general direction they care about | 16:00 |
| JayF | that doesn't mean I can't work on it; but it is a canary that maybe this is work that'd be fun for us and not useful to anyone | 16:00 |
| dtantsur | I know janders will work on improving the logic around picking the vmedia device. We cannot move it to sushy right away (backports!), but maybe we should encourage him to follow-up with moving this logic to sushy. | 16:01 |
| dtantsur | Otherwise, I hear ya. But then, I'm stubbornly pushing the Inspector merger for years now, while nobody seems to care much :D | 16:02 |
| JayF | I think two things: | 16:02 |
| rpittau | see ya tomorrow folks! o/ | 16:02 |
| JayF | 1) I care about Inspector merge; it's good and it makes our model cleaner | 16:02 |
| JayF | BUT that's all mitigated by | 16:03 |
| JayF | 2) Deployment automation has improved to where an average operator DGAF it takes an extra service to do a thing | 16:03 |
| JayF | so like, good work, it's needed, but most of the pain it resolves is hidden from opers | 16:03 |
| dtantsur | It's not just installing a service. It's performance overhead on the ops side and a lot of maintenance overhead on our side. | 16:06 |
| JayF | I agree with that from our, deeper perspective | 16:07 |
| JayF | but it's hidden from operators just looking at it from a surface level | 16:08 |
| opendevreview | Merged openstack/ironic-python-agent-builder stable/2023.1: Add libaio engine for fio on Centos/Rhel9 https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/886488 | 16:29 |
| JayF | dtantsur: iurygregory: Isn't one of you on the RH Openshift team? | 16:47 |
| dtantsur | JayF: both (and Riccardo too) | 16:47 |
| JayF | DM | 16:47 |
| dtantsur | and Mahnoor, and Derek, and Jacob :) | 16:47 |
| JayF | heh, was telling dtantsur to be nice to my friend starting in openshift | 16:51 |
| JayF | but apparently you all are a small army | 16:51 |
| dtantsur | :D | 16:51 |
| dtantsur | speaking of inspector, could I get a 2nd +2 on https://review.opendev.org/c/openstack/ironic/+/887202 and https://review.opendev.org/c/openstack/ironic/+/885450 please? | 17:13 |
| opendevreview | Verification of a change to openstack/bifrost master failed: Fix ipaddr deprecation warning https://review.opendev.org/c/openstack/bifrost/+/889195 | 18:08 |
| JayF | lots and lots of TIMED_OUT on that one | 18:09 |
| opendevreview | Merged openstack/ironic master: Correct two mistakes in the /continue_inspection API https://review.opendev.org/c/openstack/ironic/+/887202 | 18:43 |
| opendevreview | Merged openstack/ironic master: Very basic in-band inspection with the "agent" interface https://review.opendev.org/c/openstack/ironic/+/885450 | 18:51 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Fix typo in deploy_templates docs https://review.opendev.org/c/openstack/ironic/+/890163 | 21:35 |
| JayF | Extremely trivial; spotted it while drafting a spec | 21:36 |
| opendevreview | Jay Faulkner proposed openstack/ironic-specs master: Self-Service via Templates https://review.opendev.org/c/openstack/ironic-specs/+/890164 | 21:56 |
| * JayF notes that 890164 was partially drafted by ChatGPT 4.0 + link reader plugin | 21:56 | |
| JayF | It did ... okay. Not great, but OK. Gave me a starting point for almost everywhere when fed our spec template, code repo, and the bug with the feature summary I already wrote | 21:57 |
| JayF | (The posted copy is heavily edited by my own hand; that was just a starting point) | 21:57 |
| opendevreview | Jay Faulkner proposed openstack/ironic-specs master: Self-Service via Templates https://review.opendev.org/c/openstack/ironic-specs/+/890164 | 22:14 |
| JayF | I have a friend from another IRC channel I was pointing at bifrost; I was telling them about discovery but aside from the release note ( https://github.com/openstack/bifrost/blob/de39802cdb75e5eeba0895824f83e6aba7420ffd/releasenotes/notes/enable-inspector-discovery-77aa6f5c1f1e42c6.yaml ) I don't see any docs on it in bifrost | 23:04 |
| JayF | is there any magic to it? | 23:04 |
| JayF | If awb joins, that's the friend, help 'em out :D | 23:04 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!