| janders | good morning Ironic o/ | 07:51 |
|---|---|---|
| janders | Happy (Gregorian) New Year 2023 everyone! | 07:51 |
| rpittau | good morning ironic! o/ | 08:19 |
| rpittau | vmedia is not happy with 2023 already :/ | 08:19 |
| opendevreview | Riccardo Pittau proposed openstack/sushy-tools master: Fix tox4 error https://review.opendev.org/c/openstack/sushy-tools/+/868760 | 08:27 |
| rpittau | seeing this in the vmedia job logs https://8d73c19295e479ed277f-694cde954935b887fd0bda99c57e475e.ssl.cf2.rackcdn.com/867962/3/check/ipa-tempest-uefi-redfish-vmedia-src/d1fdf05/controller/logs/ironic-bm-logs/node-0_console_2023-01-02-17%3A57%3A49_log.txt | 09:07 |
| rpittau | could be an issue due to uefi+libvirt | 09:07 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 09:24 |
| ajya | Hi, in wallaby sushy docs job fails with "make is not allowed, use allowlist_externals to allow it". Looks like a change in newer tox. Not sure why it's wallaby only. Should add make to allowlist_externals? https://review.opendev.org/c/openstack/sushy/+/868788 | 09:25 |
| ajya | Or was PDF generation removed in newer branches? | 09:25 |
| rpittau | ajya: I think we added make to allowlist starting from xena, so that needs to be added | 09:36 |
| rpittau | or better, we need to change from whitelist to allowlist | 09:37 |
| ajya | rpittau: backport this https://review.opendev.org/c/openstack/sushy/+/796399? Or still need to keep 3.9.0 as min version in wallaby? | 09:47 |
| rpittau | ajya: that's perfect, thanks, go ahead with the backport, I'll approve it | 09:48 |
| rpittau | looks like we're using tox4 for all the stable branches too, so any change due to that needs to be backported down to wallaby | 09:56 |
| rpittau | or we cap tox in older branches, but I doubt we'll do that | 09:56 |
| opendevreview | Aija Jauntēva proposed openstack/sushy stable/wallaby: Update min version of tox to use allowlist https://review.opendev.org/c/openstack/sushy/+/869041 | 10:03 |
| opendevreview | Aija Jauntēva proposed openstack/sushy stable/wallaby: Update min version of tox to use allowlist https://review.opendev.org/c/openstack/sushy/+/869041 | 10:04 |
| ajya | ^ ready | 10:06 |
| rpittau | ajya: sorry, I think we need to include that in https://review.opendev.org/c/openstack/sushy/+/868788 otherwise other jobs won't pass :/ | 10:18 |
| ajya | rpittau: right, then dropping this and kamleshChauvhan will update his patch | 10:20 |
| rpittau | ajya: thanks, realized a bit late :/ | 10:20 |
| rpittau | TheJulia, JayF, when you have a moment please check https://review.opendev.org/c/openstack/ironic/+/868521 | 10:28 |
| ajya | rpittau: np, should have realized myself | 11:01 |
| rpittau | I wonder how the functional tests were passing until now if IRONIC_BAREMETAL_BASIC_OPS was set to False and the macs file was not there, the job should've failed at https://opendev.org/openstack/ironic/src/branch/master/devstack/lib/ironic#L2379 but the trap was not triggering until switching to jammy | 11:27 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: [WIP] Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 11:27 |
| opendevreview | kamlesh chauvhan proposed openstack/sushy stable/wallaby: Fix tox4 and setuptools errors, update tox version https://review.opendev.org/c/openstack/sushy/+/868788 | 11:45 |
| iurygregory | morning Ironic | 12:18 |
| janders | hey iurygregory o/ | 12:23 |
| iurygregory | janders, o/ happy new year =) | 12:23 |
| opendevreview | Merged openstack/ironic-python-agent master: Fix for tox4 and setuptools https://review.opendev.org/c/openstack/ironic-python-agent/+/867962 | 12:31 |
| ajya | Wallaby patch is passing, but newer branches have blockers. Should W+1 wallaby patch or wait for newer ones to merge first? https://review.opendev.org/c/openstack/sushy/+/868788 | 13:23 |
| janders | iurygregory thank you - Happy New Year 2023 to you, too! | 13:32 |
| rpittau | ajya: let's merge that and xena, for zed we need to wait for ironic patch to merge and I think we'll have to do some backports for yoga too | 13:36 |
| ajya | rpittau: ok | 13:38 |
| opendevreview | Riccardo Pittau proposed openstack/ironic stable/yoga: Fix CI https://review.opendev.org/c/openstack/ironic/+/869045 | 13:38 |
| rpittau | this ^ is the yoga backport, probably we can remove the increase memory part, we'll see | 13:39 |
| rpittau | still need to check why the image has doubled its size | 13:39 |
| rpittau | mmm I think I need a separate change to fix the file not found error for the macs file | 13:41 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Do not look for IRONIC_VM_MACS_CSV_FILE if we don't generate it https://review.opendev.org/c/openstack/ironic/+/869084 | 13:47 |
| rpittau | this ^ should avoid the issue with the missing macs file in jammy | 13:47 |
| rpittau | not sure why it does not trigger the trap until focal | 13:47 |
| rpittau | maybe a bug in the shell? ¯\_(ツ)_/¯ | 13:48 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: [WIP] Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 13:49 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 13:49 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 13:49 |
| opendevreview | Oleksandr Kozachenko proposed openstack/networking-generic-switch master: Add ArubaOS-CX switch support https://review.opendev.org/c/openstack/networking-generic-switch/+/868598 | 14:27 |
| opendevreview | Merged openstack/sushy stable/wallaby: Fix tox4 and setuptools errors, update tox version https://review.opendev.org/c/openstack/sushy/+/868788 | 14:35 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: [WIP] [PoC] A metal3 CI job https://review.opendev.org/c/openstack/ironic/+/863873 | 14:59 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Create IRONIC_VM_MACS_CSV_FILE if it does not exist https://review.opendev.org/c/openstack/ironic/+/869084 | 15:09 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Use jammy for base jobs https://review.opendev.org/c/openstack/ironic/+/869052 | 15:10 |
| opendevreview | Merged openstack/sushy stable/xena: Fix tox4 and setuptools errors https://review.opendev.org/c/openstack/sushy/+/868787 | 15:27 |
| TheJulia | Good morning | 15:37 |
| rpittau | good morning TheJulia :) | 15:40 |
| TheJulia | rpittau: regarding the last part of the sentence on https://review.opendev.org/c/openstack/ironic/+/868521/4/releasenotes/notes/fix-context-image-hardlink-16f452974abc7327.yaml#6 | 15:53 |
| TheJulia | does that mean the deploy will, or will not work | 15:53 |
| TheJulia | I ask because depending on how you read the release note, it in not entirely clear | 15:54 |
| TheJulia | oh wow :( Image sizes | 16:08 |
| TheJulia | ipa-centos9-stable-zed.tar.gz2022-12-15 20:13926M | 16:08 |
| TheJulia | [ ]ipa-centos9-master.tar.gz2023-01-03 12:41805M | 16:08 |
| TheJulia | that is 2x the size | 16:08 |
| * TheJulia downloads one to begin the examination | 16:09 | |
| samuelkunkel[m] | August in 2022 it was around 450M... | 16:09 |
| samuelkunkel[m] | Atleast thats the size of our ipa from back then | 16:10 |
| TheJulia | good dat apoint, I wonder if they moved firmware around or added new firmware | 16:10 |
| TheJulia | data point | 16:10 |
| TheJulia | Our yoga image is like 417 MB | 16:14 |
| TheJulia | I do remember during Zed we were hitting around 430mb or so | 16:14 |
| * TheJulia wonders if there could be an ipa image downloading song | 16:16 | |
| TheJulia | but not "222 megabytes of ipa, download one, 223 megabytes of ipa" | 16:17 |
| samuelkunkel[m] | that matches. Just checked - its even less of a size. | 16:18 |
| samuelkunkel[m] | 413M Aug 31 20:08 ipa.initramfs | 16:19 |
| TheJulia | rpittau: is the vmedia stuff from yesteday in relation to the macs hwinfo file? | 16:23 |
| *** dansmith_ is now known as dansmith | 16:24 | |
| rpittau | TheJulia: re https://review.opendev.org/c/openstack/ironic/+/868521/4/releasenotes/notes/fix-context-image-hardlink-16f452974abc7327.yaml#6 the deployment won't work if selinux is enabled and enforcing where the hardlink is created | 16:30 |
| TheJulia | That seems like a separate thing we need to address | 16:31 |
| rpittau | TheJulia: re vmedia stuff is not related to the macs hwinfo, I think. Puzzling enough, jobs should fail if that does not exist, but they don't, I noticed only in jammy where it correctly exits with a trap | 16:31 |
| TheJulia | is there a check we can run? | 16:31 |
| TheJulia | okay | 16:31 |
| TheJulia | w/r/t vmedia, is there a pristine example of a failure? | 16:31 |
| rpittau | TheJulia: for the selinux stuff you can check the bifrost CI | 16:31 |
| rpittau | for vmedia this is zed but the failure is the same https://zuul.opendev.org/t/openstack/build/fdd84facbd5f45a495483fd0c81a4ad0 | 16:33 |
| rpittau | I need to go now, but I will check msgs later o/ | 16:34 |
| JayF | That isn't like the failure I saw on master and was digging yesterday | 16:37 |
| JayF | that one was a single failure error that it never completed cleaning | 16:37 |
| TheJulia | w/r/t vmedia: https://b07845379824cfa9e48f-ea6c0ca013bce2ed83ca9ffa0031d5bd.ssl.cf1.rackcdn.com/868789/1/check/ironic-tempest-uefi-redfish-vmedia/fdd84fa/controller/logs/ironic-bm-logs/node-1_console_2023-01-03-09%3A35%3A53_log.txt | 16:41 |
| TheJulia | there is our problem | 16:41 |
| JayF | I've seen a lot of errors on server consoles | 16:41 |
| JayF | that's a new one for me :) | 16:41 |
| TheJulia | no oom or anything... although I think there is a case where grub can do something like that in a resource constrained VM | 16:42 |
| TheJulia | we should fix the ramdisk first, tbh | 16:42 |
| TheJulia | and see where things go from there | 16:42 |
| * TheJulia extracts the giant ramdisk of doom | 16:42 | |
| JayF | Did you find anything obviously outta whack (outside of "it's big" with the ramdisk?) | 16:42 |
| TheJulia | yup | 16:45 |
| JayF | path to firmware root dir changed? | 16:45 |
| TheJulia | https://www.irccloud.com/pastebin/kz56YovN/ | 16:45 |
| TheJulia | junk in /var/tmp.... | 16:45 |
| JayF | egad | 16:46 |
| TheJulia | looks like there is a slightly larger mellenox firmware image now | 16:47 |
| TheJulia | but only 66 mb, so not a huge gain and people do use those cards | 16:47 |
| TheJulia | (and each file is still reasonably sized) | 16:47 |
| JayF | I'm amazed we went this long without stripping /var/tmp | 16:47 |
| TheJulia | well, we expect to be empty | 16:50 |
| TheJulia | we don't expect dracut to be leaving things around | 16:50 |
| JayF | is /var/tmp a ramdisk on a booted machine? | 16:51 |
| JayF | It is on many distros; I've not used CentOS 9 stream though.... | 16:51 |
| TheJulia | yeah, in the ramdisk | 16:51 |
| TheJulia | in the base image, I believe it is just an empty folder | 16:51 |
| JayF | How do we end up with dracut in the var/tmp then? | 16:52 |
| TheJulia | dracut likely gets triggered in the image build | 16:53 |
| TheJulia | to include modules | 16:53 |
| TheJulia | I'll have to check the dib source | 16:53 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder master: Remove /var/tmp/* from images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869097 | 16:57 |
| JayF | TheJulia: I wonder if we should make CI fail if IPA jobs produce an image greater than "N" | 17:03 |
| TheJulia | so it looks like there should be only one or two folders based upon what I see in the dracut scripting and dib | 17:07 |
| TheJulia | looks like the last change was to drop py2 support from dib, but other things can also trigger dracut to regenerate like updated kernels | 17:08 |
| TheJulia | CI... maybe. But not the image build itself, some operators need the absurd ramdisk images :\ | 17:09 |
| JayF | yeah, of course | 17:10 |
| JayF | the other thing I've pondered, if we should publish a CI-only image with more aggro firmware pruning | 17:10 |
| TheJulia | that would be tinyipa really :) | 17:37 |
| TheJulia | it has only cpu firmware | 17:37 |
| JayF | that is true | 17:48 |
| TheJulia | 869097 is looking like it is in good shape so far | 18:02 |
| TheJulia | Just need one more core | 18:03 |
| JayF | iurygregory: you around to land https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869097 | 18:04 |
| iurygregory | JayF, yes | 18:08 |
| iurygregory | CI looks green based on status, lgtm +W | 18:09 |
| TheJulia | so the vmedia issue *may* be ovmf | 18:12 |
| JayF | ovmf? | 18:12 |
| TheJulia | uefi firmware reference code | 18:19 |
| TheJulia | but... I'm going to run it past some grub folks I know to see what they say | 18:19 |
| JayF | What is our fix in the short term? | 18:21 |
| JayF | make the job nonvoting? | 18:21 |
| TheJulia | I'm not sure yet | 18:21 |
| TheJulia | grub maintainers prodded | 18:22 |
| TheJulia | "hey, does this cause deja vu?" | 18:22 |
| TheJulia | There is an open debian bug with the same invalid opcode from March | 18:22 |
| TheJulia | There is a super super fimilar RHCOS bug as well, but I think it was one of the things we already found and got fixed (unless something reverted) | 18:23 |
| TheJulia | err, similar | 18:24 |
| JayF | Even if that is correct, it'll take days for that to roll downstream to us, yeah? | 18:27 |
| TheJulia | possibly longer, but they they think it might be the wrong grub binary | 18:33 |
| * TheJulia waits() | 18:33 | |
| TheJulia | rhel recently did some simplification of the /boot/EFI structure becaues the wrong file was getting loaded in some cases and bad htings would happen | 18:34 |
| TheJulia | JayF: by chance have you submitted an operator feedback sort of session for the summit? | 18:38 |
| JayF | I've not submitted anything for summit | 18:38 |
| JayF | Forum stuff is open until April; I would think that kind of feedback would be forum? | 18:38 |
| TheJulia | yeah, but I know some folks are already submitting stuff, so I wouldn't wait until april | 18:39 |
| TheJulia | so the esp image is ubuntu artifacts | 18:43 |
| TheJulia | and the ramdisk is centos | 18:44 |
| TheJulia | I think that might be our issue :\ | 18:44 |
| JayF | Bluntly; that sounds like a bug in any event | 18:44 |
| JayF | Yeah? | 18:44 |
| JayF | like not our bug, like, their bug | 18:44 |
| JayF | Is there a valid reason for them to be incompatible? | 18:45 |
| TheJulia | well | 18:45 |
| TheJulia | I'm not sure, but I'm also wondering if this is the ovmf bug I stumbled upon in debian | 18:45 |
| TheJulia | which makes no sense | 18:45 |
| JayF | Do you have that bug handy for me to read up on? | 18:47 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Use centos grub artifacts with centos ramdisk for vmedia https://review.opendev.org/c/openstack/ironic/+/869103 | 19:14 |
| TheJulia | I hope that works | 19:15 |
| TheJulia | JayF: ^^^ | 19:15 |
| TheJulia | explaination why support of just "working" is iffy as well in the commit message | 19:15 |
| JayF | not a bad approach from our perspective | 19:16 |
| JayF | but I really ,really hope upstream isn't dropping this | 19:16 |
| TheJulia | what do you mean "dropping this"? | 19:21 |
| JayF | meaning like, this failure scenario is not OK | 19:22 |
| JayF | I should be able to mix and match those as a user | 19:22 |
| JayF | (upstream in this case being grub/ubu/centos) | 19:23 |
| TheJulia | it is not okay if secure boot is enforcing | 19:23 |
| TheJulia | because it shouldn't work with signed bits | 19:23 |
| TheJulia | but these are not | 19:23 |
| TheJulia | which is... depressing | 19:23 |
| TheJulia | we'll see | 19:23 |
| JayF | I mean, that's what I'm saying | 19:24 |
| JayF | if it's validating that the boot chain is the ssame with secure boot off, that's bad behavior | 19:24 |
| JayF | and will be breaky to folks in some rescue/recovery situations | 19:24 |
| TheJulia | woot, that change blew up horribly | 19:28 |
| TheJulia | like everything is red | 19:28 |
| * TheJulia feels like this is success for today | 19:28 | |
| TheJulia | i *think* there is another fundamental challenge at play, which is the fragmentation of grub2 overall. | 19:30 |
| TheJulia | like there are config differences between dialects of grub2 now | 19:30 |
| TheJulia | which... *insert scream here* | 19:30 |
| opendevreview | Julia Kreger proposed openstack/ironic master: Use centos grub artifacts with centos ramdisk for vmedia https://review.opendev.org/c/openstack/ironic/+/869103 | 19:34 |
| TheJulia | now we wait() | 19:34 |
| opendevreview | Julia Kreger proposed openstack/ironic-python-agent-builder stable/zed: Remove /var/tmp/* from images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869118 | 19:44 |
| opendevreview | Merged openstack/ironic-python-agent-builder master: Remove /var/tmp/* from images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869097 | 20:12 |
| JayF | fungi: how do I manually kick off a "post" job? We need ironic-python-agent-build-image-dib-centos9 to run and publish a new IPA image based on changes we just put into ironic-python-agent-builder | 20:41 |
| fungi | JayF: normally the way you do it is to merge something in the repo. worst case a sysadmin can re-enqueue the last thing that merged to the branch | 20:42 |
| JayF | If we merge a change to that repo, it'd be 100% just to kick that job off | 20:43 |
| JayF | which I can do if it's the preferred method... | 20:43 |
| fungi | we do it all the time for our container image builds. some of our dockerfiles have comment lines that literally say something like "update to trigger a new image upload yyyy-mm-dd" | 20:43 |
| fungi | i take it the ironic-python-agent-build-image-dib-centos9 job isn't triggered by merges to ironic-python-agent-builder... should it be? | 20:45 |
| opendevreview | Merged openstack/ironic-inspector master: Remove lib/neutron-legacy leftovers https://review.opendev.org/c/openstack/ironic-inspector/+/868192 | 20:45 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent master: Remove old, unused proxy.sh file https://review.opendev.org/c/openstack/ironic-python-agent/+/869105 | 20:46 |
| JayF | fungi: I was thinking about that, it's a possibility but we'd need to review to make sure we CI enough in that repo | 20:46 |
| JayF | iurygregory: TheJulia: https://review.opendev.org/c/openstack/ironic-python-agent/+/869105 exists to help us get a new IPA build out for CI, you wanna take a look? | 20:47 |
| iurygregory | JayF, I'm ok with it +2 | 20:52 |
| TheJulia | approved the ipa change | 20:57 |
| TheJulia | looks like inspector's merge fail on the 21st was memory related, so hopefully that should make the world much happier once it merges | 20:58 |
| TheJulia | I'm sensing vmedia is still broken :( | 20:59 |
| TheJulia | do we know when vmedia failures started? | 20:59 |
| JayF | I saw them emerge along the same timeline as tox 4.0 | 20:59 |
| JayF | so like, christmas week? maybe earlier and I didn't notice? | 20:59 |
| TheJulia | so 22nd possibly? | 20:59 |
| JayF | I mean, in that ballpark yeah | 21:00 |
| TheJulia | stevebaker[m]: ^^ I'm sensing the kernel might be our headache :\ | 21:00 |
| TheJulia | I'll look at the finished job | 21:00 |
| TheJulia | once it fails | 21:00 |
| JayF | I think it might be wise at this point to temporarily make this job nonvoting? | 21:00 |
| JayF | I'm happy to write that change if you'd +2 it | 21:00 |
| TheJulia | lets see what the current change up there comes back with since it is still running (a bad sign) | 21:01 |
| JayF | that comment is based on the presumption that fails :) | 21:02 |
| JayF | if it doesn't fail, celebrate instead :D | 21:02 |
| TheJulia | I'm going to packup here and head down the mountain | 21:03 |
| * TheJulia wonders if centos's rpm builder is public | 21:04 | |
| TheJulia | yeah, it failed | 21:05 |
| TheJulia | :( | 21:05 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Temporarily mark redfish vmedia as non-voting https://review.opendev.org/c/openstack/ironic/+/869107 | 21:09 |
| TheJulia | JayF: you have a syntax issue | 21:12 |
| TheJulia | the job name now needs to end with a : | 21:12 |
| opendevreview | Jay Faulkner proposed openstack/ironic master: Temporarily mark redfish vmedia as non-voting https://review.opendev.org/c/openstack/ironic/+/869107 | 21:13 |
| JayF | thanks | 21:13 |
| TheJulia | np | 21:15 |
| TheJulia | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_598/869103/2/check/ironic-tempest-uefi-redfish-vmedia/5981ece/controller/logs/ironic-bm-logs/node-0_console_2023-01-03-20%3A19%3A02_log.txt le-sigh | 21:17 |
| JayF | you mentioned a "kernel change" | 21:18 |
| JayF | did centos9 ship the stuff removing support for 32 bit systems or similar? | 21:18 |
| TheJulia | I don | 21:18 |
| TheJulia | I don't think so, it was a build rev difference | 21:19 |
| TheJulia | like 210 vs 212 | 21:19 |
| TheJulia | I think that is grub differences, with food I'll be able to think through it | 21:19 |
| JayF | yeah, like pathing differences | 21:20 |
| TheJulia | a *ton* of wifi stuff merged | 21:24 |
| TheJulia | the change log is insane | 21:29 |
| TheJulia | lots of stuff dropped in the 21st and 22nd, but none of it looks really related | 21:30 |
| JayF | I mean, in my gentoo experience | 21:30 |
| JayF | invalid opcode can also mean like, cflags changed | 21:30 |
| JayF | e.g. they compiled it with options that make it not work on our VM | 21:30 |
| JayF | like enabling avx or something similar | 21:30 |
| JayF | TheJulia: can you look at a config diff? | 21:31 |
| JayF | TheJulia: (or link me one) | 21:31 |
| TheJulia | https://kojihub.stream.centos.org/koji/buildinfo?buildID=28004 | 21:32 |
| TheJulia | yeah, it could | 21:32 |
| TheJulia | hmm | 21:34 |
| JayF | TheJulia: | 21:40 |
| JayF | > - redhat: configs: disable vDPA on all archs except x86_64 (Laurent Vivier) [2140885] | 21:40 |
| JayF | 12/14 | 21:40 |
| JayF | that is sus | 21:40 |
| JayF | are we running x86_64 vms? | 21:40 |
| TheJulia | we are | 21:41 |
| TheJulia | we do host passthrough | 21:41 |
| JayF | ah | 21:41 |
| TheJulia | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_598/869103/2/check/ironic-tempest-uefi-redfish-vmedia/5981ece/controller/logs/libvirt/libvirt/qemu/node-0_log.txt | 21:46 |
| JayF | TheJulia: not sure what I should see there | 21:49 |
| TheJulia | just what is there and what the processor gets passed in as | 21:49 |
| JayF | ah, okay | 21:50 |
| TheJulia | stevebaker[m]: you mentioned you had some issues when you built an ESP image on rhel recently, do you have some notes? | 21:50 |
| stevebaker[m] | TheJulia: I think it was all environmental, it had to be a privileged container for the mount to work. But I'll be switching to mtools when the dependencies are in the zed container images https://github.com/steveb/ironic-operator/commit/220a07224281b2181acdcb1bee560515b788bed8 | 21:54 |
| TheJulia | hmm | 21:55 |
| TheJulia | okay | 21:55 |
| TheJulia | Anyway, I'm going to look later, lunch and drive time now | 21:55 |
| stevebaker[m] | ok | 21:58 |
| opendevreview | Merged openstack/ironic-python-agent-builder stable/zed: Remove /var/tmp/* from images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869118 | 23:12 |
| opendevreview | Jay Faulkner proposed openstack/ironic-python-agent-builder stable/yoga: Remove /var/tmp/* from images https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/869120 | 23:21 |
| JayF | We need one more core review on https://review.opendev.org/c/openstack/ironic/+/869107 to temporarily unstick the gate | 23:23 |
| stevebaker[m] | done | 23:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!