| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2024-08-06 | |||
| 16:16:22 | elodilles | if anyone finds any gate issue then please add it there ^^^ | |
| 16:16:35 | elodilles | that's all about stable from me | |
| 16:17:01 | gibi | elodilles: thanks! | |
| 16:17:08 | elodilles | np | |
| 16:17:16 | gibi | #topic vmwareapi 3rd-party CI efforts Highlights | |
| 16:17:20 | gibi | #info fwiesel on PTO until the 27th August. | |
| 16:17:33 | gibi | so I guess no news here | |
| 16:17:39 | gibi | #topic Open discussion | |
| 16:17:44 | gibi | (bauzas) Aug 6, 13 and 20 nova meetings | |
| 16:18:05 | gibi | so I took today's meeting but cannot do 13th or 20th | |
| 16:18:10 | gibi | any volunteer? | |
| 16:20:24 | gibi | then they are cancelled | |
| 16:20:47 | gibi | #info the meeting on the 13th and on the 20th are cancelled | |
| 16:21:00 | gibi | any other topic for today? | |
| 16:21:43 | dansmith | ++ for cancel :) | |
| 16:23:25 | gibi | I guess we are done for today then | |
| 16:23:31 | gibi | #endmeeting | |
| 16:23:31 | opendevmeet | Meeting ended Tue Aug 6 16:23:31 2024 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | |
| 16:23:31 | opendevmeet | Minutes: https://meetings.opendev.org/meetings/nova/2024/nova.2024-08-06-16.00.html | |
| 16:23:31 | opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/nova/2024/nova.2024-08-06-16.00.txt | |
| 16:23:31 | opendevmeet | Log: https://meetings.opendev.org/meetings/nova/2024/nova.2024-08-06-16.00.log.html | |
| 16:23:36 | elodilles | thanks o/ | |
| 16:23:37 | gibi | thanks folks | |
| 16:24:03 | tkajinam | thanks | |
| 16:24:22 | opendevreview | Merged openstack/nova-specs master: Follow up for "libvirt: AMD SEV-ES support" https://review.opendev.org/c/openstack/nova-specs/+/924563 | |
| 18:09:48 | opendevreview | Artom Lifshitz proposed openstack/nova master: POC/DNM: Remove vTPM live migration block https://review.opendev.org/c/openstack/nova/+/925771 | |
| 18:31:22 | opendevreview | Artom Lifshitz proposed openstack/nova master: POC/DNM: Remove vTPM live migration block https://review.opendev.org/c/openstack/nova/+/925771 | |
| 23:09:09 | opendevreview | melanie witt proposed openstack/nova master: libvirt: Check version for node device XML validate flags https://review.opendev.org/c/openstack/nova/+/925826 | |
| 23:14:06 | melwitt | fyi gibi and sean-k-mooney ^ thanks songwenping_ for reporting the issue | |
| 23:17:40 | sean-k-mooney | melwitt: that wont work | |
| 23:17:49 | sean-k-mooney | the api does not have a flags filed before 8.10.0 | |
| 23:18:04 | sean-k-mooney | melwitt: so you ned to change the function call rahter then the flags | |
| 23:18:48 | melwitt | sean-k-mooney: ugh dammit. thanks 😩 | |
| 23:19:18 | sean-k-mooney | actully | |
| 23:19:23 | sean-k-mooney | no i miss read the c code | |
| 23:19:28 | sean-k-mooney | https://github.com/libvirt/libvirt/commit/d8791c3c7caa6e3cadaf98a5a2c94b232ac30fed#diff-29f62ad20a5a2dffc62d8d09df5ef6e7b28c36c723ae7da56eb371c7547fbf98L70 | |
| 23:19:39 | sean-k-mooney | so it was passing a hardcoded 0 for flags before | |
| 23:19:48 | sean-k-mooney | so your change shoudl be good | |
| 23:20:03 | sean-k-mooney | ill review it properly tomorrow after we see what ci says | |
| 23:20:30 | sean-k-mooney | the docs change clarifies it | |
| 23:20:32 | sean-k-mooney | https://github.com/libvirt/libvirt/commit/d8791c3c7caa6e3cadaf98a5a2c94b232ac30fed#diff-7552129210db333b0c2e3d224d1ae11a674bbb0b56f6b2708452be4e0c023a98L697 | |
| 23:20:38 | sean-k-mooney | @flags: extra flags; not used yet, so callers should always pass 0 | |
| 23:20:46 | sean-k-mooney | to * @flags: bitwise-OR of supported virNodeDeviceCreateXMLFlags | |
| 23:22:31 | melwitt | ah ok, I have seen they have that for some APIs where it takes flags of 0 as sort of a placeholder | |
| 23:22:43 | sean-k-mooney | yep | |
| 23:23:01 | sean-k-mooney | the try to future proof may apis by allowing flags even if not supproted | |
| 23:23:16 | sean-k-mooney | i guess that semi reasonabel for c to not break abi compatiablity | |
| #openstack-nova - 2024-08-07 | |||
| 01:00:53 | opendevreview | melanie witt proposed openstack/nova master: DNM testing https://review.opendev.org/c/openstack/nova/+/925635 | |
| 02:45:46 | opendevreview | Takashi Kajinami proposed openstack/nova master: Detect AMD SEV-ES support https://review.opendev.org/c/openstack/nova/+/925685 | |
| 06:57:49 | opendevreview | melanie witt proposed openstack/nova master: DNM testing cont'd https://review.opendev.org/c/openstack/nova/+/925838 | |
| 07:04:10 | opendevreview | melanie witt proposed openstack/nova master: DNM testing cont'd https://review.opendev.org/c/openstack/nova/+/925838 | |
| 07:10:02 | melwitt | sean-k-mooney: just for illustrating how the fetch_func stuff is not used at all in this poc ^ (hopefully I didn't mess up in deleting) | |
| 07:41:17 | gibi | melwitt: thanks for proposing the fix for the persistent mdev bug | |
| 10:06:51 | ykarel | can someone check the zed backport https://review.opendev.org/c/openstack/nova/+/925749 | |
| 10:11:56 | opendevreview | Takashi Kajinami proposed openstack/nova master: Detect AMD SEV-ES support https://review.opendev.org/c/openstack/nova/+/925685 | |
| 11:48:32 | elodilles | ykarel: clean cherry pick so +2+W'd it | |
| 11:48:47 | ykarel | thx elodilles | |
| 11:48:51 | gibi | ykarel: as that targets an unmaintained branch nova core team is not involved reviewing and landing it | |
| 11:51:45 | gibi | but I guess elodilles is in the global unmaintained core team :) | |
| 11:53:02 | ykarel | gibi, ack | |
| 11:53:03 | sean-k-mooney | so i have debated this in the past with my self but i think the unmatined branches woudl benifit form a dedicated channel that is seperate form the proejct channels to cordinate unmatined branch reviews | |
| 11:54:53 | elodilles | sean-k-mooney: there is a semi-official channel: #openstack-unmaintained o:) | |
| 11:57:49 | sean-k-mooney | ack provided the unmaintained pings are not excessive i dont entirly object to havign them here too but we were very clear in teh resolution that core teams would not review patches too or genericallly maintiane the unmainted branches as part fo the resolution | |
| 11:58:07 | elodilles | yepp | |
| 11:58:56 | sean-k-mooney | so unless an unmainted branch maitiner wants input on a patch they are unsure about i woudl prefer to avoid the day to day work of maintainign them here | |
| 11:59:34 | sean-k-mooney | elodilles: if you are unsure about somethign imporant liek a cve back port we will proably be fine with a quick skim over a patch you have prepared | |
| 12:00:07 | sean-k-mooney | or if your trying to resolve a merge conflcit and we are not too busy but in general for can we review x i would prefer to keep that in #openstack-unmaintained | |
| 12:13:47 | elodilles | sean-k-mooney: ACK, thanks in advance! o:) | |
| 12:23:59 | gibi | (not that I want it) | |
| 12:24:48 | sean-k-mooney | same i considerd puting my name forward but i have too much on my plate currently to really consier it | |
| 12:25:34 | sean-k-mooney | i know many are in that both but im jsut relecting reality that i cant spend time on the unmainited branches at present | |
| 12:35:48 | gibi | we set them unmaintaned as we don't want to / cannot maintain them. So it is fair to say that we dont care :) | |
| 15:34:23 | stephenfin | elodilles: Speaking of unmaintained, we have a variety of git-review proposal patches open against cliff, python-openstackclient and openstacksdk. Can you or someone abandon those since it's very unlikely anyone will use those branches for client tooling | |
| 16:34:19 | opendevreview | melanie witt proposed openstack/nova master: libvirt: Remove node device XML validate flags https://review.opendev.org/c/openstack/nova/+/925826 | |
| 16:39:34 | JayF | sean-k-mooney: & other cores; https://review.opendev.org/q/topic:ironic-guest-metadata is tested working in devstack, I wanted some review feedback before moving beyond this point into migration/tempest. I've asked Ironic cores to look at them as well. | |
| 16:40:14 | sean-k-mooney | i was ment to loop back to those after the cve | |
| 16:40:50 | JayF | I figured :). No real rush just sometime in the next few days would be nice | |
| 16:41:10 | JayF | I'm pretty crushed right now too and wouldn't be able to do anything immediately with the feedback, but I know the queue is long so wanted to take my number ;) | |
| 16:42:40 | sean-k-mooney | JayF: does the ironic patch depend on the nova one and is that tested in ci | |
| 16:42:48 | sean-k-mooney | i.e. can we use that to demonstrate thsi end to end | |
| 16:43:08 | sean-k-mooney | we do have results form https://zuul.opendev.org/t/openstack/build/0c7147a1aea541eb88ae387cb5487fe0 also | |
| 16:43:35 | sean-k-mooney | which i guess we can reference im just not sure if the metadta will be visable in the ironic logs but i woudl hope we can see it properly there | |
| 16:43:46 | JayF | sean-k-mooney: I tested it manually in devstack. automatic_lessee defaults to false in Ironic so it won't show up in CI until we explicitly test it; and I'm not going to add to the chain until I get some directional feedback (i.e. this is good if tested) | |
| 16:44:17 | JayF | sean-k-mooney: mainly just trying to avoid a thing where I write 5 patches across 5 repos and get feedback to change a single word that takes an hour because of the stacking :P | |
| 16:45:02 | JayF | I knew I shoulda done a quick+dirty demo video; CID did that for his recent Ironic API change and it was WONDERFUL as a reviewer to see | |
| 16:45:37 | sean-k-mooney | ack we can likely turn it on in that job in a follwo up testing patch | |
| 16:45:40 | sean-k-mooney | by the way | |
| 16:45:42 | sean-k-mooney | Jul 25 20:34:42.508653 np0038062841 nova-compute[106292]: ERROR nova.network.neutron [req-a7169636-31a7-4d11-9017-a6865cd16b59 req-102a3eac-094a-42f8-a114-8ac7a841a33c service nova] [instance: 929bb573-a06a-4fa7-a77a-c643bd9655ce] The vnic_type of the bound port 88d554aa-8c7a-46ac-b54b-efeac135c643 has been changed in neutron from "normal" to "baremetal". Changing vnic_type of | |
| 16:45:44 | sean-k-mooney | a bound port is not supported by Nova. To avoid breaking the connectivity of the instance please change the port vnic_type back to "normal". | |
| 16:46:28 | sean-k-mooney | that is a sideffect of ironic doing somethign that is not allowed that related to a cve previously | |
| 16:47:02 | JayF | Can you document that in a bug somewhere please? | |
| 16:47:04 | sean-k-mooney | in this speicic case its not problematic form a secuirty point of view but is a side effect of the exisitng tech debt | |
| 16:47:52 | sean-k-mooney | sure my guess is the job is not configure properly to set the tempest config option for the vnic_type to baremental | |
| 16:48:38 | sean-k-mooney | speicicaly https://github.com/openstack/tempest/blob/master/tempest/config.py#L784-L791 | |
| 16:52:42 | sean-k-mooney | if you need a prot type other the normal nova only supprot that by passing a precreated port as part of the server create request | |
| 16:53:03 | sean-k-mooney | so ironic just like sriov ports required tempest to be configued to use the relevent vnic type in the job | |
| 17:08:16 | sean-k-mooney | JayF: https://bugs.launchpad.net/ironic/+bug/2076265 os for the issue in the job config althoguh i think the larger issue of nova/ironic/neutron interaction is really a seprate problem | |
| 17:09:15 | JayF | sean-k-mooney: I'd suggest sending an email to the mailing list proposing a joint PTG session then. It's not a topic I can address well in the moment as networking is far from my area of expertise in Ironic and I'd need to get more context | |
| 17:10:35 | elodilles | stephenfin: i guess those are mainly the .gitreview patches. i'll rather check if we could merge them now. (i also don't think that we will get (many) backports to these projects, but at least the .gitreview file should be correct) | |
| #openstack-nova - 2024-08-08 | |||
| 03:14:59 | opendevreview | melanie witt proposed openstack/nova master: DNM testing https://review.opendev.org/c/openstack/nova/+/925635 | |