| opendevreview | Takashi Kajinami proposed openstack/ironic master: Skip functional tests on pre-commit config update https://review.opendev.org/c/openstack/ironic/+/933194 | 01:32 |
|---|---|---|
| opendevreview | Takashi Kajinami proposed openstack/ironic master: Skip functional tests on pre-commit config update https://review.opendev.org/c/openstack/ironic/+/933194 | 01:39 |
| opendevreview | Steve Baker proposed openstack/ironic-python-agent master: Capture and log sector sizes https://review.opendev.org/c/openstack/ironic-python-agent/+/933200 | 01:57 |
| opendevreview | Steve Baker proposed openstack/ironic-python-agent master: Capture and log sector sizes https://review.opendev.org/c/openstack/ironic-python-agent/+/933200 | 02:08 |
| rpittau | good morning ironic! o/ | 06:51 |
| rpittau | TheJulia: re baremetal SIG: thanks for that, looks good to me! | 07:13 |
| dtantsur | JayF: I did not mean last_error, I simply overlooked that we use LOG.exception here | 07:45 |
| opendevreview | Merged openstack/ironic-python-agent master: CI: Remove metalsmith legacy jobs https://review.opendev.org/c/openstack/ironic-python-agent/+/933153 | 08:10 |
| opendevreview | Takashi Kajinami proposed openstack/ironic master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic/+/933227 | 09:13 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-inspector master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic-inspector/+/933228 | 09:14 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-lib master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic-lib/+/933229 | 09:14 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-python-agent master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic-python-agent/+/933230 | 09:15 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironicclient master: Remove Python 3.8 support https://review.opendev.org/c/openstack/python-ironicclient/+/933231 | 09:16 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironic-inspector-client master: Remove Python 3.8 support https://review.opendev.org/c/openstack/python-ironic-inspector-client/+/933232 | 09:17 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-lib master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic-lib/+/933229 | 09:29 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-lib master: Replace outdated python job template https://review.opendev.org/c/openstack/ironic-lib/+/933242 | 09:29 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironicclient master: Remove Python 3.8 support https://review.opendev.org/c/openstack/python-ironicclient/+/933231 | 09:51 |
| opendevreview | Takashi Kajinami proposed openstack/python-ironicclient master: Replace outdated python job template https://review.opendev.org/c/openstack/python-ironicclient/+/933244 | 09:51 |
| rpittau | tkajinam: hey thank you for all the py38-removal patches! Are you planning to do that for all the ironic repos besides what you proposed so far? jlmk in case I can take care of the rest | 11:30 |
| tkajinam | rpittau, hey! let me try covering the others, too, so that you can vote your +2 | 11:38 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-ui master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic-ui/+/933258 | 11:39 |
| tkajinam | in case you find anything not covered then feel free to submit the change... it'd be helpful if the same topic name can be used | 11:39 |
| opendevreview | Takashi Kajinami proposed openstack/sushy-tools master: Remove Python 3.8 support https://review.opendev.org/c/openstack/sushy-tools/+/933259 | 11:41 |
| rpittau | tkajinam: thanks! will do in case | 11:42 |
| opendevreview | Takashi Kajinami proposed openstack/virtualpdu master: Remove Python 3.8 support https://review.opendev.org/c/openstack/virtualpdu/+/933260 | 11:42 |
| opendevreview | Takashi Kajinami proposed openstack/virtualbmc master: Remove Python 3.8 support https://review.opendev.org/c/openstack/virtualbmc/+/933263 | 11:44 |
| opendevreview | Takashi Kajinami proposed openstack/metalsmith master: Replace outdated python job template https://review.opendev.org/c/openstack/metalsmith/+/933264 | 11:46 |
| opendevreview | Takashi Kajinami proposed openstack/metalsmith master: Bumm minimum python version https://review.opendev.org/c/openstack/metalsmith/+/933265 | 11:48 |
| opendevreview | Takashi Kajinami proposed openstack/ironic-tempest-plugin master: Remove support for Pyton 3.6 and 3.7 https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/933266 | 11:51 |
| opendevreview | Takashi Kajinami proposed openstack/virtualpdu master: Drop unnecessary 'x' bit from doc config file https://review.opendev.org/c/openstack/virtualpdu/+/933268 | 11:53 |
| opendevreview | Takashi Kajinami proposed openstack/virtualbmc master: Drop unnecessary 'x' bit from doc config file https://review.opendev.org/c/openstack/virtualbmc/+/933269 | 11:54 |
| opendevreview | Takashi Kajinami proposed openstack/networking-baremetal master: Remove Python 3.8 support https://review.opendev.org/c/openstack/networking-baremetal/+/933273 | 12:01 |
| opendevreview | Takashi Kajinami proposed openstack/networking-generic-switch master: Remove Python 3.8 support https://review.opendev.org/c/openstack/networking-generic-switch/+/933274 | 12:01 |
| opendevreview | Takashi Kajinami proposed openstack/tenks master: Add Python 3.12 job https://review.opendev.org/c/openstack/tenks/+/933275 | 12:03 |
| tkajinam | asking a quick question if anyone can answer. Is networking-generic-switch still alpha status or can we promote its status now ? | 12:08 |
| tkajinam | looking at "Development Status :: 3 - Alpha" in its setup.cfg | 12:08 |
| opendevreview | Takashi Kajinami proposed openstack/bifrost master: Remove Python 3.8 support https://review.opendev.org/c/openstack/bifrost/+/933288 | 12:21 |
| rpittau | tkajinam: that's from 9 years ago, I'm pretty sure we can remove it | 12:32 |
| tkajinam | yeah that's what I thought | 12:34 |
| opendevreview | Takashi Kajinami proposed openstack/networking-generic-switch master: Drop alpha status https://review.opendev.org/c/openstack/networking-generic-switch/+/933297 | 12:35 |
| dtantsur | tkajinam: people definitely use it in production | 12:38 |
| tkajinam | dtantsur, oh, nice :-) | 12:40 |
| tkajinam | so people does not really care about that metadata :-) | 12:40 |
| dtantsur | yeah, I suspect they don't :) | 12:44 |
| rpittau | ironic PTG part 4 is starting in 15 minutes with Nova joint session in austin room | 12:46 |
| dtantsur | rpittau: for people who forgot what Nova is: Ironic itself starts in 1:15, right? | 12:48 |
| rpittau | dtantsur: yeah :D | 12:48 |
| rpittau | well that was quick | 13:22 |
| rpittau | we'll be back with ironic PTG in 37 minutes in mitaka room | 13:23 |
| TheJulia | okay, thanks! | 13:23 |
| cardoe | rpittau: so stupid question about the depends-on. when I'm rebased on top and use git review, it says it wants to submit both patches. It won't resubmit the underlying one as current patch + 1 | 13:30 |
| rpittau | cardoe: it should not submit the first patch | 13:30 |
| cardoe | okay good | 13:31 |
| rpittau | "should not" :D | 13:31 |
| TheJulia | If anyone would like to keep steve baker awake, join the ironic chat :) | 13:32 |
| TheJulia | err | 13:32 |
| TheJulia | ironic ptg room | 13:32 |
| rpittau | cardoe: btw if you update your vote here https://review.opendev.org/c/openstack/sushy/+/931321 I will +W | 13:32 |
| rpittau | TheJulia: coming in 2 minutes :D | 13:33 |
| rpittau | 5 minutes to ironic PTG part 4 :) | 13:55 |
| opendevreview | Verification of a change to openstack/ironic master failed: Gracefully handle bad request exception https://review.opendev.org/c/openstack/ironic/+/931849 | 13:57 |
| opendevreview | Takashi Kajinami proposed openstack/sushy master: Drop unnecessary 'x' bit from doc config file https://review.opendev.org/c/openstack/sushy/+/933314 | 13:57 |
| opendevreview | Takashi Kajinami proposed openstack/sushy-tools master: Drop unnecessary 'x' bit from doc config file https://review.opendev.org/c/openstack/sushy-tools/+/933315 | 13:58 |
| cardoe | okay that's twice now a failure on the 4k for https://review.opendev.org/c/openstack/ironic/+/931849 | 14:01 |
| JayF | clarkb: you had a comment on the bootable containers on our etherpad, if you wanna jump we're in here | 14:07 |
| JayF | **talking about it now | 14:07 |
| clarkb | JayF: bit early for me but if you are still discussing it I am around ish now | 14:26 |
| JayF | we just got done | 14:26 |
| JayF | comments are in the etherpad but tl;dr he hought was your concern wasn't valid for bootc cases | 14:27 |
| JayF | because bootc just handles those details in ways that don't involve new package installs | 14:27 |
| clarkb | the main thing I wanted to call out is that you can have tools that are not centos and fedora specific and also that there may be security concerns depending on how involved the conversion is | 14:27 |
| JayF | so I went down that list, and it seems bootc *does* support non-fedora/coreos | 14:28 |
| clarkb | ok the readme definitely says "this is fedora/centos focused" | 14:28 |
| clarkb | in whatever tool was linked in the etherpad | 14:28 |
| JayF | or at least people in the issues there saying "I can't get debian to work" and getting happy at the end | 14:29 |
| JayF | yeah, I agree | 14:30 |
| JayF | but we already have an anaconda driver | 14:30 |
| JayF | which is "enterprise linux" specific, too | 14:30 |
| clarkb | a direct format conversion seems safest. But if you start needing to do package installs to add additioanl required bits (for example to switch from a minimal to a generic kernel capable of booting on hardware, or adding cloud-init, etc) then you need to be careful | 14:31 |
| JayF | yeah we specifically scoped out anything more than running bootc + putting down a configdrive | 14:31 |
| JayF | no package installs, no injected or dynamic configurations | 14:31 |
| clarkb | perfect | 14:31 |
| opendevreview | Merged openstack/sushy master: Remove Python 3.8 support https://review.opendev.org/c/openstack/sushy/+/931321 | 14:33 |
| rpittau | mmm lost connection for a moment | 14:52 |
| rpittau | JayF: looks like I have issues with meetpad, can't reconnect | 14:54 |
| TheJulia | turn off ipv6? | 14:54 |
| TheJulia | It is working just fine for me, I just reloaded... | 14:55 |
| clarkb | sounds like it broke for a couple of people. Wondering if it is a network issue for people sharing routes | 14:55 |
| TheJulia | I'm definitely using v4 | 14:55 |
| TheJulia | ++ | 14:55 |
| clarkb | it is reachable for me (also over v4) | 14:55 |
| JayF | I am v6 only | 14:55 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: Add vlan aware VMs support https://review.opendev.org/c/openstack/networking-generic-switch/+/928490 | 14:55 |
| rpittau | I'm having issues with other random sites too, probably some general internet issue | 14:57 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: Add vlan aware VMs support https://review.opendev.org/c/openstack/networking-generic-switch/+/928490 | 14:57 |
| JayF | s/v6/v4/ | 14:57 |
| JayF | I am v4 only | 14:57 |
| opendevreview | Michal Nasiadka proposed openstack/networking-generic-switch master: Add vlan aware VMs support https://review.opendev.org/c/openstack/networking-generic-switch/+/928490 | 14:58 |
| johnthetubaguy | rpittau: FWIW, I have having the same thing suddenly, review.opendev.org no dice, although its just started coming back now… | 14:58 |
| JayF | We are breaking until :10 after the hour | 14:58 |
| JayF | to hopefully give time for the internet to recombobulate itself | 14:58 |
| rpittau | JayF: ack | 14:58 |
| opendevreview | Derek Higgins proposed openstack/ironic master: Eject vmedia if fast track is enabled https://review.opendev.org/c/openstack/ironic/+/933326 | 14:59 |
| opendevreview | Merged openstack/sushy master: Drop unnecessary 'x' bit from doc config file https://review.opendev.org/c/openstack/sushy/+/933314 | 15:27 |
| opendevreview | Merged openstack/networking-generic-switch master: Drop alpha status https://review.opendev.org/c/openstack/networking-generic-switch/+/933297 | 15:29 |
| pas-ha[m] | did I miss the ansible deploy discussion? .. ☹️ looks like I miscalculated the time 😭 | 15:33 |
| JayF | yes, but I suspect if you jump in we can revisit | 15:33 |
| JayF | TheJulia: rpittau ^ | 15:33 |
| rpittau | pas-ha[m]: we can defintely have a second take if you join | 15:34 |
| pas-ha[m] | gimme 10 min and I can join | 15:34 |
| opendevreview | Merged openstack/networking-generic-switch master: Remove Python 3.8 support https://review.opendev.org/c/openstack/networking-generic-switch/+/933274 | 15:34 |
| opendevreview | Doug Goldstein proposed openstack/ironic master: Enable WSGI module entrypoint for Ironic https://review.opendev.org/c/openstack/ironic/+/927635 | 15:37 |
| pas-ha[m] | rpittau: JayF TheJulia whenever you are ready, I'm in the call | 15:45 |
| JayF | ++ this topic is winding down | 15:45 |
| rpittau | pas-ha[m]: jsut edning the currnet topic | 15:45 |
| pas-ha[m] | sure, I can hear that 🙂 no rush | 15:45 |
| TheJulia | That was an excellent PTG! | 16:18 |
| TheJulia | so, who is doing the follow-up ironicbaremetal.org blog post?! | 16:18 |
| cardoe | rpittau: so I made a wsgi module for https://review.opendev.org/c/openstack/ironic/+/927635 and that still didn't work with autodoc. | 16:24 |
| opendevreview | Derek Higgins proposed openstack/sushy master: Provide vmedia username and password if required https://review.opendev.org/c/openstack/sushy/+/923524 | 16:24 |
| JayF | TheJulia: I will likely do a summary *video*, so I'll leave the blogging to someone else, but I will explicitly authorize anyone to point an AI to that video as the starts of a blogpost | 16:25 |
| cardoe | Where's devnull when ya need him. He likes doing blog posts. | 16:29 |
| cardoe | If anyone wants to go +W happy on the ironic-week-prio list, there's a handful in there. | 16:52 |
| opendevreview | Doug Goldstein proposed openstack/metalsmith master: Bump minimum python version https://review.opendev.org/c/openstack/metalsmith/+/933265 | 16:54 |
| cardoe | cid: I didn't peek at what was wrong with https://review.opendev.org/c/openstack/ironic/+/931849 | 16:59 |
| cardoe | JayF: I'll defer to your review for https://review.opendev.org/c/openstack/ironic/+/930944?usp=search since you asked for a change there. | 17:01 |
| cardoe | https://review.opendev.org/c/openstack/ironic/+/931061 is a bit outside of my comfort to +2 | 17:02 |
| JayF | cardoe: if you wanna +2 930944 I'll land it | 17:03 |
| cardoe | and that covers most of the ironic-week-prio items. There's a couple more that are a bit older than have been banging around so I won't mention those since they're getting worked. | 17:03 |
| JayF | I'm mostly in the same boat re 931061 | 17:03 |
| JayF | was going to wait for steve's review there tbh | 17:03 |
| cardoe | done on 930944 | 17:03 |
| cid | cardoe: I just checked, no specific failures, "a unit test exited without returning a status" | 17:04 |
| cardoe | yeah. twice on the 4k job | 17:04 |
| cardoe | I dunno what that means. | 17:04 |
| JayF | > ironic.tests.unit.common.test_format_inspector.TestFormatInspectors.test_iso_9660 | 17:04 |
| JayF | I would +2 a change to remove that whole damned test module | 17:05 |
| JayF | since it's going to be in oslo.utils this cycle anyway | 17:05 |
| JayF | it means these jobs are IO intensive and fail sometimes | 17:05 |
| TheJulia | yeah, uhh... Did we move to the oslo module yet? | 17:33 |
| TheJulia | if we did, nuke it from high orbit | 17:33 |
| TheJulia | ... like tungston rod from orbit | 17:33 |
| JayF | not yet, but it's causing CI failures anyway | 17:50 |
| opendevreview | Julia Kreger proposed openstack/ironic master: CI: Change 4k block device job to use LVM https://review.opendev.org/c/openstack/ironic/+/931061 | 17:56 |
| JayF | we still use that module but: 1) I don't anticipate us changing it 2) We have canaries in other projects (oslo.utils itself) and 3) we'll have the replacement in place before we do a release this cycle (I hope) | 18:10 |
| JayF | I'm just tired of seeing timeouts on tests in that module | 18:10 |
| opendevreview | Merged openstack/ironic master: Add support for node name in port creation https://review.opendev.org/c/openstack/ironic/+/930944 | 18:18 |
| opendevreview | Merged openstack/ironic master: Remove Python 3.8 support https://review.opendev.org/c/openstack/ironic/+/933227 | 18:19 |
| TheJulia | Regarding supporting an OCI image url, would it make to always cache it conductor side? The alternative is also supporting the download in the agent but that might be un-necessary complexity | 18:47 |
| cardoe | So I think we'd probably say no caching on the conductor for that. | 18:50 |
| cardoe | At least that was my thought. | 18:50 |
| cardoe | Cause the point was reducing complexity / improving ergonomics for running Ironic in Kubernetes. | 18:51 |
| cardoe | Because today we've gotta create a Persistent Volume Claim, mount it to conductor and then mount it to the http server. We need to craft the Service to share the IP between conductor and the http server. There by defeating the ability to use a Load Balancer. | 18:52 |
| cardoe | I capitalized Kubernetes terms. | 18:52 |
| cardoe | I'll let dtantsur weigh in his thoughts here. | 18:53 |
| TheJulia | Well, just thinking in a fully integrated context, the usage is going to be mixed regardless | 19:00 |
| TheJulia | a smaller first step would be just the conductor side | 19:00 |
| TheJulia | conductor's http server was never intended to need a load balancer in front of it, fwiw | 19:00 |
| cardoe | yeah not saying conductor needs a load balancer. Sorry bad words. | 19:16 |
| cardoe | So usually in k8s you're not messing with exposed IPs. When you do its via a Load Balancer. | 19:17 |
| cardoe | But with the conductor is managing an http server, you're needing each conductor to have an IP | 19:17 |
| JayF | TheJulia: I'll take that question downstream to my team in context of steps to see what they think. My gut feel is to not cache at conductor and have agent directly interact with the container registry, but I'm not 100% sure | 19:38 |
| clarkb | a lot of container registries rate limit pretty aggressively these days. Distributing the fetches can also mitigate those limits assuming you don't end up going through a shared IP/NAT/Proxy | 20:03 |
| JayF | I *strongly* assume these will be connecting to private registries. | 20:16 |
| JayF | at least in any reasonable production environment | 20:16 |
| jamesdenton | ass-u-me | 20:20 |
| JayF | I mean, Ironic standalone you can provide an http url to anywhere in the world, public or private | 20:20 |
| JayF | any tool can be used badly :) | 20:20 |
| JayF | containers are just easier to use badly ;) | 20:21 |
| jamesdenton | yes, i also set selinux to permissive. | 20:21 |
| JayF | but it's SO EASY to create custom rules as you need, just follow these steps [seventeen page blog post] /s | 20:21 |
| jamesdenton | haha | 20:22 |
| cardoe | he laughs and quits. | 21:00 |
| cardoe | Little does he know I signed him up to write some neutron RFEs. | 21:00 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!