Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2026-08-24
16:17:07 Uggla Btw currently testing the virtiofs shares and cold migration from gouthamr.
16:17:26 Uggla #topic Stable Branches
16:17:39 elodilles thx
16:17:43 elodilles #info stable/2026.2 branching will start soon, first for (client) libraries next week, then for services and others
16:17:50 elodilles #info osc-placement's stable/2025.2 and stable/2025.1 gate is blocked by tox-docs job - fix: https://review.opendev.org/q/I57d5e77056e643b821f1c26ae371e26c6df2f20a
16:18:09 elodilles hmm, i mean these have merged today ^^^
16:18:14 elodilles thanks to stephenfin o/
16:18:18 sean-k-mooney :)
16:18:22 elodilles #info other stable gates should be OK
16:18:47 elodilles i'm not aware of other stable gate issues
16:18:49 gibi elodilles: shoud we backport some instabilities form master? :)
16:18:56 gibi *from
16:19:00 elodilles gibi: noooo /o\
16:19:03 elodilles :)
16:19:11 elodilles #info stable branch status / gate failures tracking etherpad: https://etherpad.opendev.org/p/nova-stable-branch-ci
16:19:25 elodilles so that's all from me
16:19:35 sean-k-mooney speaking of mster i would like use to add a ubuntu 26.04 job before the end of the cycle is there is no objection
16:19:37 Uggla thx elodilles
16:19:50 sean-k-mooney we can dicuss that at the end of the meeting however
16:20:12 Uggla skipping vmware topic as fwiesel is on pto.
16:20:23 Uggla #topic Kamil's news about eventlet removal
16:20:32 Uggla sambork, gibi, something to share ?
16:20:42 sambork Unify the thread pool executors into a central factory" merged, and the follow-up is under review: https://review.opendev.org/c/openstack/nova/+/1001572. I rebased "Replace StaticallyDelayingWrapper with futurist DelayingExecutor" on top of it, but I want to retest and adopt a similar factory pattern. After that, we can proceed with this patch since Futurist was released last week. Functional test stabilization is progressing
16:20:43 sambork (https://review.opendev.org/c/openstack/nova/+/999121) the second of the two excluded tests is now under review: https://review.opendev.org/c/openstack/nova/+/1000780
16:21:38 sambork I also got a new set of performance tests with some config changes. The only thing I noticed was a slight reduction in memory consumption, though I'm not sure if that's because there were fewer replicas or if it's a genuine reduction in memory usage
16:22:33 sambork I should have some more news on wednesday sync
16:22:57 sambork something more gibi?
16:24:04 sambork if to, move back to you Uggla
16:24:05 gmaan one question, this not tied to FF right? "Replace StaticallyDelayingWrapper with futurist DelayingExecutor"
16:24:18 gmaan just want to make sure i prioritise the reviews accordingly
16:24:27 sambork imo no
16:24:30 sean-k-mooney that could merge before RC1
16:24:39 sean-k-mooney its just tech debt reduction
16:24:46 gmaan yeah, ++
16:24:46 sean-k-mooney not a feature
16:26:00 Uggla great thx sambork, moving on.
16:26:06 gibi I agree with the above and no additions
16:26:46 Uggla gibi you don't want to add instabilities in Kamil's patches too? :)
16:27:06 opendevreview Takashi Kajinami proposed openstack/nova master: doc: Add releases information for SEV-ES and SEV-SNP https://review.opendev.org/c/openstack/nova/+/1002164
16:27:11 Uggla #topic Nova using openstack sdk for neutron
16:27:24 Uggla lajoskatona, something you'd like to share?
16:27:31 lajoskatona I will be quick, I had no time since last week to work on this
16:28:00 lajoskatona The SDK patch is merged, and I hope it can be part of the coming release (with my other SDK patches together :-))
16:28:40 lajoskatona I mean this SDK patch for fileds filtering : https://review.opendev.org/c/openstack/openstacksdk/+/995724
16:28:49 lajoskatona so that's it for the SDK topic
16:28:58 Uggla :+1
16:29:09 Uggla thx lajoskatona
16:29:12 gibi Uggla: hehh those instabilities are expensive I cannot just give them away as candy :)
16:29:21 sean-k-mooney that shoudl be in the sdk release next week
16:29:52 Uggla #topic Confidential computing, status of SNP and TDX features
16:30:01 elodilles sean-k-mooney: https://review.opendev.org/c/openstack/releases/+/1001307
16:30:16 elodilles (the patch is already there ^^^)
16:30:23 Uggla tkajinam, antia, gibi, gmaan something something new ?
16:30:31 sean-k-mooney elodilles: yep
16:31:02 sean-k-mooney Uggla: SNP is merged TDX is +2'd all the way
16:31:09 sean-k-mooney i think it has +w on most of the patches
16:31:22 sean-k-mooney gibi gmaan and i have reviewd teh series
16:31:29 gmaan i too reviewed the TDX series on friday and most of it was in good shape. I think I had only one question on one of the changed
16:31:37 gibi yeah about secure boot
16:31:38 sean-k-mooney antia posted teste resutls for secure boot as a comment today
16:31:42 gibi yepp
16:31:49 gmaan perfect
16:32:00 gmaan I will check those today
16:32:06 gibi thanks
16:32:10 antia thanks all
16:32:37 gmaan gibi: FYI, sean-k-mooney antia and I talked about publishing the testing of TDX in upstream
16:32:39 sean-k-mooney antia: there may be some merge conflcit that will require rebases
16:33:02 gmaan here
16:33:05 gmaan #link https://etherpad.opendev.org/p/nova-ci-failures-minimal#L400
16:33:15 sean-k-mooney so we will need to wtach the patches until they finally land
16:33:18 antia sean-k-mooney: I rebased with main last friday, but I'll keep an eye on it!
16:33:43 gmaan #link https://github.com/openstack/nova/tree/master/doc/source/contributor/testing
16:34:11 gibi gmaan: I would say write a whitebox tempest test
16:34:17 gmaan as gibi also done the testing, i think it will be good to add your scenario if anything special
16:34:21 sean-k-mooney ya os i dont think its good to commit results in the repo
16:34:38 sean-k-mooney but we can have a basic how to test doc
16:34:38 gibi tests can serves as executable documentation
16:34:46 sean-k-mooney and ya tempest/whitebox were planned to follow
16:34:51 gmaan whitebox testing is good but as there are many things to take care for TDX configuration, having those in testing guide can be useful
16:35:05 gibi most of the compllication of testing this is HW setup, the flavor and image setup is pretty standard
16:35:22 gibi but I would not start carrying HW config doc in openstack tree
16:35:44 gmaan yeah, HW setup which will not be covered by tempest/whitebox tests and it will be helpful for someone who want to run those tests
16:36:10 gibi I don't like that as we cannot validate those HW configs and they tend to change over time
16:36:22 tkajinam I have to reach LY team to understand their current status about whitebox tempest plugin tests for snp but I suspect they haven't touched it yet.
16:36:28 gibi I'm fine linking to intel and amd docs
16:36:36 sean-k-mooney same, i woudl focus on just the nova settings
16:37:03 sean-k-mooney perhaps the generic kernel options but not toch any bios/hardware cofnig that is specific to a given severver
16:37:08 gibi yeah
16:37:30 sean-k-mooney thing like how to check the capcity or kernel parms is fine
16:37:43 tkajinam I was thinking of adding a few useful check points to determine whether OS and firmware are configured properly (for example checking sysfs output and libvirt domcapabilities)
16:37:46 gmaan well I was thinking of full guide somewhere whihc can tell HW setup (at leat generic one), flavor/image setup, and tests to run (manual or tepmest/whitebox)
16:37:59 tkajinam which I think sean-k-mooney is also suggesting
16:38:02 sean-k-mooney gmaan: in that case i think that hsoudl live outside the nova repo
16:38:06 gibi yeah
16:38:08 gibi maybe a blogpost
16:38:25 tkajinam blogpost may also work, yeah
16:38:26 sean-k-mooney tkajinam: yep more or less
16:38:33 gmaan thats fine, or in whitebox doc where tests will live or so but blogpost also works
16:38:45 sean-k-mooney virsh cap/domcaps or cat /sys is fine
16:39:05 gibi I would really like if amd and intel publish nice docs (and intel did) so we can refer them

Earlier   Later