| opendevreview | cid proposed openstack/ironic master: Self-Service via Runbooks https://review.opendev.org/c/openstack/ironic/+/922142 | 04:32 |
|---|---|---|
| opendevreview | cid proposed openstack/ironic master: Self-Service via Runbooks https://review.opendev.org/c/openstack/ironic/+/922142 | 04:44 |
| opendevreview | cid proposed openstack/ironic-python-agent master: Add support for burnin-gpu https://review.opendev.org/c/openstack/ironic-python-agent/+/925415 | 05:11 |
| opendevreview | cid proposed openstack/ironic master: docs-audit-2024: Use gerunds for task headings https://review.opendev.org/c/openstack/ironic/+/925129 | 05:46 |
| rpittau | good morning ironic! happy friday! o/ | 06:58 |
| janders | rpittau o/ | 07:21 |
| opendevreview | Riccardo Pittau proposed openstack/bifrost master: Uplift default Ansible version to 9.x https://review.opendev.org/c/openstack/bifrost/+/925563 | 07:23 |
| rpittau | hey janders :) | 07:23 |
| rpittau | cid: great job with the video on runbooks, really appreciated that :) | 08:09 |
| opendevreview | Maryna Savchenko proposed openstack/ironic master: Fixes config drive base64 encoding in kickstart file https://review.opendev.org/c/openstack/ironic/+/925569 | 09:37 |
| cid | ++ :) | 09:53 |
| jssfr | https://review.opendev.org/c/openstack/ironic-python-agent/+/925087 so this is ready for review and we've used it successfully in our deployment. do I need to poke someone specifically, set some flag, or will someone look at it eventually? | 10:06 |
| opendevreview | Himanshu Roy proposed openstack/ironic master: add virtual media GET api https://review.opendev.org/c/openstack/ironic/+/921657 | 10:22 |
| adam-metal3 | hellor Ironic! | 10:56 |
| iurygregory | I'm really wondering if 2017 bugs are valid or not omg XD | 11:20 |
| iurygregory | https://bugs.launchpad.net/networking-generic-switch/+bug/1737017 | 11:22 |
| iurygregory | ok it was fixed https://github.com/openstack/networking-generic-switch/commit/461def3e0b5a94bf7dbb6c08a263a4379ac9d2cf | 11:29 |
| iurygregory | rpittau, I've assigned you to https://bugs.launchpad.net/ironic-python-agent-builder/+bug/2069266 :D and marked as triaged lol | 12:03 |
| iurygregory | launchpad, why you don't sync things in bug when they have Closes-Bug? <facepalm> | 12:19 |
| iurygregory | adam-metal3, I was doing triage for some bugs and I've found https://bugs.launchpad.net/sushy-tools/+bug/2067405 that you have a patch for it, but I'm not sure if there was a discussion to approve the RFE or not, do you know? Tks | 12:20 |
| adam-metal3 | iurygregory: yes I have a patch for it https://review.opendev.org/c/openstack/sushy-tools/+/912113 | 12:24 |
| iurygregory | ack o/ | 12:24 |
| opendevreview | Himanshu Roy proposed openstack/ironic master: add virtual media GET api https://review.opendev.org/c/openstack/ironic/+/921657 | 12:25 |
| adam-metal3 | iurygregory: oh sorry I missread your question, I don't think there was any deeper discussion but it is not really an RFE or I am not sure it is somewhere between a bug and and RFE | 12:26 |
| iurygregory | adam-metal3, since it you mentioned RFE normally we try to discuss in the weekly meeting to make sure the community is ok with the proposal | 12:27 |
| * iurygregory updates the bug just in case | 12:27 | |
| opendevreview | Himanshu Roy proposed openstack/ironic master: add virtual media GET api https://review.opendev.org/c/openstack/ironic/+/921657 | 12:28 |
| adam-metal3 | iurygregory: should I add it to the next meeting agenda? | 12:30 |
| rpittau | iurygregory: not sure I'll have the time to work on that, I started it but currently JayF has taken over | 12:35 |
| iurygregory | adam-metal3, ++ | 12:37 |
| iurygregory | rpittau, oh ack o/ lets wait for him to be online before I assign to him | 12:38 |
| opendevreview | Himanshu Roy proposed openstack/ironic master: add virtual media GET api https://review.opendev.org/c/openstack/ironic/+/921657 | 12:40 |
| jlejeune | hello all, I have a quick question for you, I'm testing a "new" version of nova (2023.1) with a stein version of Ironic, and since the upgrade I can see in noa-compute ironic log a lot of stacktraces complaining about a get_host_uptime() method not implemented in ironic driver, that I didn't have with a Stein version of Nova. Does it ring a bell to someone ? | 12:48 |
| jlejeune | I see that opened bug which seems to be related: https://bugs.launchpad.net/nova/+bug/1910504 | 12:49 |
| jlejeune | the exception is caught on nova conductor side but I think it's strange to not catch it on compute side | 12:50 |
| jlejeune | what do you think ? | 12:50 |
| iurygregory | maybe something changed in the virt driver and you also need to upgrade ironic to have things working, I never used nova + ironic so I'm not sure | 13:06 |
| dtantsur | I cannot imagine how the uptime value would be related to Ironic version, but I"m also no nova expert | 13:06 |
| TheJulia | stein. Wow | 13:11 |
| TheJulia | Anyway, I do remember the get_host_uptime thing | 13:11 |
| TheJulia | I think either nova nuked it from high orbit, or nova changed the driver to silence errors on it from the ironic case | 13:11 |
| TheJulia | since, there is no hypervisor uptime to convey/relate and it really means nothing when the responsible entity is the service as a whole it is talking to | 13:12 |
| jlejeune | yeah Stein I know ;) indeed, on nova side, the NotImplementedError is caught and ignored but not on compute side | 13:14 |
| jlejeune | a easy workaround is to create a get_host_uptime() method in ironic driver to return None but I tried to understand why the error occurs now | 13:15 |
| jlejeune | maybe I should ask on nova chan, actually, they may have an explanation | 13:16 |
| TheJulia | jlejeune: They are going to likely point out you *really* need to upgrade | 14:02 |
| TheJulia | fwiw | 14:02 |
| jlejeune | I know but it's not so simple for us | 14:10 |
| cardoe | good morning ironic | 14:22 |
| rpittau | bye everyone, have a great weekend! o/ | 14:46 |
| opendevreview | Riccardo Pittau proposed openstack/bifrost master: Update min required pip version to 22.3.1 https://review.opendev.org/c/openstack/bifrost/+/925602 | 14:54 |
| opendevreview | Riccardo Pittau proposed openstack/bifrost master: Uplift default Ansible version to 9.x https://review.opendev.org/c/openstack/bifrost/+/925563 | 14:55 |
| cardoe | So I'm wondering on https://review.opendev.org/c/openstack/python-ironicclient/+/924895 if I should add the flag to the "service" command as well? | 16:06 |
| TheJulia | cardoe: huh? not sure I understand the context of the question | 17:23 |
| JayF | jlejeune: you really, really need to get on something you can security patch, especially in light of the recent nova/cinder/glance CVEs, if you're using openstack integrated | 17:24 |
| cardoe | So my change adds --disable-ramdisk to the "baremetal node clean" command. Should I also add it for "baremetal node service"? | 17:24 |
| JayF | +1 | 17:24 |
| TheJulia | cardoe: oh, I guess yeah | 17:25 |
| TheJulia | Has anyone submitted to the OpenInfra Days North America meetup? | 18:23 |
| TheJulia | well, 2 days | 18:23 |
| TheJulia | CFP is open until August 12th it looks | 18:24 |
| iurygregory | I haven't | 18:36 |
| iurygregory | and I probably won't submit anything, since I'm alrady going to the OIS | 18:38 |
| TheJulia | Yeah, the event is looking at 150-ish folks if memory serves | 18:44 |
| JayF | IIRC it's the weekend after my brothers' wedding, so I won't be able to attend | 19:08 |
| JayF | which is sad because I have a GREAT talk idea | 19:08 |
| TheJulia | awww | 19:10 |
| JayF | I've been toying for a while with "OpenStack is the bearer of bad news" as the idea at the center of a talk | 19:11 |
| JayF | and I think I finally found the way to tell a positive instead of a negative story with it | 19:12 |
| JayF | maybe I'll just youtube video it once I get a bit more time | 19:12 |
| TheJulia | everything is complex and blah blah blah ? | 19:12 |
| JayF | more like, we need to tell the full story with metrics | 19:12 |
| TheJulia | oh, heh | 19:12 |
| TheJulia | yeah | 19:12 |
| TheJulia | physics ftw? | 19:12 |
| JayF | like "don't just measure failures, measure how often good things happen, too" | 19:13 |
| TheJulia | ++++++++++ | 19:13 |
| JayF | basically around the idea that nobody thinks about infra when it works, so you have to push the narrative that things are working when they do work | 19:13 |
| TheJulia | that is the downside of metrics, when you focus on just the bad, you never see good | 19:13 |
| TheJulia | and then you never really know if your thresholds are good because your just hyper-focusing | 19:13 |
| JayF | and this leads in a way to openstack and openstack-ops teams getting a bad rap as openstack basically reports *any* issues (networking/power/hardware/human) | 19:13 |
| TheJulia | We enable a lot of complexity | 19:14 |
| TheJulia | and we simplify a lot | 19:14 |
| TheJulia | and if you make things overly complex further down the stack, the more risk of issues/complexity and whatnot breaking makes OpenStack the thing you blame when really, it was decisions made that may not have been good ideas | 19:14 |
| JayF | that's why you gotta point at the hidden seams so people see it BEFORE those seams might start to fray due to environmental issues | 19:14 |
| TheJulia | yeah | 19:14 |
| TheJulia | I've sort of been tempted to do a variation on the vancouver RBAC talk | 19:15 |
| JayF | "look at this massive pile of pain we're preventing; maybe the 2 hours you couldn't deploy a server because it's ram failed is not a big deal" | 19:15 |
| JayF | I don't know if you'd want to do it, but there's a talk around exposing Ironic API to integrated OpenStack | 19:15 |
| JayF | like, we talk about RBAC a lot and I think people see it thru a standalone lens | 19:15 |
| JayF | when in reality if you have power users, the combo of nova's provisioning API + ironic's maintenance API enabling self-service is a really good story | 19:16 |
| TheJulia | hmmm | 19:16 |
| JayF | especially considering runbooks and ironic-guest-metadata should land this cycle | 19:16 |
| JayF | this was on my "eventually" list for a talk, or to talk my downstream into doing one if/when they complete a deployment of this vision | 19:16 |
| JayF | but I'd be THRILLED if anyone told that story | 19:17 |
| TheJulia | That is, and I know there is a class of user that knows it, and gets it, and has just been doing it. But for many, it might not be obvious | 19:20 |
| JayF | well I think we have some of the edges smoothed over enough that smaller clusters can use it more easily now | 19:21 |
| JayF | we've removed quite a few decoder rings :D | 19:21 |
| cid | o// | 19:59 |
| cardoe | I like the talk idea around metrics btw. | 20:34 |
| JayF | You shouldn't need it if you kept all the previous dashboard setups I did last decade ;) /s | 20:36 |
| *** clarkb is now known as Guest1974 | 21:33 | |
| cardoe | These Dell servers are very spiteful. | 21:37 |
| JayF | I consider myself lucky that my downstream mostly uses HP iLO gear. | 21:38 |
| cardoe | So I addressed https://review.opendev.org/c/openstack/ironic/+/924943 differently based on feedback and the new change got merged https://review.opendev.org/c/openstack/sushy/+/925130 How do I correctly close the original and reference the real fix? | 21:51 |
| cardoe | Well finally made the bug around the ethernet interfaces for the Dells at... https://bugs.launchpad.net/sushy/+bug/2075979 | 21:57 |
| JayF | abandon | 21:59 |
| *** Guest1974 is now known as clarkb | 22:44 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!