Monday, 2026-06-01

opendevreviewChungWon Lee proposed openstack/nova master: Reduce node build fail value by time  https://review.opendev.org/c/openstack/nova/+/93992906:54
opendevreviewMerged openstack/nova-specs master: Repropose and update cyborg vGPU (mdev) support  https://review.opendev.org/c/openstack/nova-specs/+/96751508:24
jgilaberHi all! I have a question about the spec https://review.opendev.org/c/openstack/nova-specs/+/967515 which just merged08:35
jgilaberSince it's reproposed it had linked an existing blueprint, should I reuse that or create a new one?08:36
rm_work[m]ralonsoh: on https://review.opendev.org/c/openstack/neutron/+/990588 I believe I've got the correct AI tag (so won't change that), but while I don't really feel the preemptive optimization on comma line endings here makes a ton of sense, I am willing to do it if you care that much 08:46
rm_work[m]> If a substantial portion of a patch is generated in this way, it must be marked with a “Generated-By:” label.08:47
seyeongkimHi. My patch https://review.opendev.org/c/openstack/nova/+/963665 makes libvirt emit per-device <boot order/> so multiple boot_index disks work (SeaBIOS 1.14+ skips non-bootindex virtio disks otherwise). In review it was noted this is more a feature than a bug. In practice the boot disks with boot_index don't seem to have to be different types(although doc says so), so I'm wondering whether this would still need a blueprint — specless, or a fu09:24
ralonsohrm_work[m], it would be better to talk about this patch in the neutron channel10:00
opendevreviewElod Illes proposed openstack/nova stable/2025.1: [CI][stable-only] nova-grenade-multinode fix  https://review.opendev.org/c/openstack/nova/+/98957914:02
*** do is now known as Guest1055614:50
UgglaReminder: upstream meeting in ~20mn15:39
opendevreviewTakashi Kajinami proposed openstack/nova-specs master: libvirt: AMD SEV-SNP support  https://review.opendev.org/c/openstack/nova-specs/+/98337615:46
Uggla#startmeeting nova16:00
opendevmeetMeeting started Mon Jun  1 16:00:57 2026 UTC and is due to finish in 60 minutes.  The chair is Uggla. Information about MeetBot at http://wiki.debian.org/MeetBot.16:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:00
opendevmeetThe meeting name has been set to 'nova'16:00
gibio/16:01
UgglaHello everyone16:01
samborko/16:01
phildree[m]o/16:01
cw0306-lee[m]o/16:01
elodilleso/16:01
fwieselo/16:02
UgglaWaiting 1 more minute to let people join.16:03
bauzaso/16:03
UgglaLet's go16:04
Uggla#topic Bugs (stuck/critical) 16:04
Uggla#info No Critical bug16:04
Uggla#topic Gate status 16:04
Uggla#link https://bugs.launchpad.net/nova/+bugs?field.tag=gate-failure Nova gate bugs 16:04
Uggla#link https://etherpad.opendev.org/p/nova-ci-failures-minimal16:04
Uggla#link https://zuul.openstack.org/builds?project=openstack%2Fnova&project=openstack%2Fplacement&branch=stable%2F*&branch=master&pipeline=periodic-weekly&skip=0 Nova&Placement periodic jobs status16:04
Uggla#info Please look at the gate failures and file a bug report with the gate-failure tag.16:04
Uggla#info Please try to provide a meaningful comment when you recheck16:04
tkajinamo/16:05
UgglaWe have a point about kernel panics in the open discussion16:05
Ugglaother than that, I'm not aware of issues ? 16:06
Uggla#topic Release Planning 16:06
Uggla#link https://releases.openstack.org/hibiscus/schedule.html16:06
Uggla#info Nova deadlines are set in the above schedule16:06
Uggla#info PTG etherpad for 2026.2 is available: https://etherpad.opendev.org/p/nova-2026.2-ptg16:06
Uggla#info Last week Thursday was soft spec freeze.16:06
Uggla#info This week Thursday (4th June) is second spec review day.16:07
Uggla#info Next week Thursday (11st June) is spec freeze.16:07
Uggla#topic Review priorities16:07
Uggla#link New file for Hibiscus https://etherpad.opendev.org/p/nova-2026.2-status16:07
Uggla#info I have updated Launchpad and the above doc. Please ping me if you spot something missing.16:07
Uggla#info I have flagged the priorities in bold.16:08
Uggla#info Please review 471815: Expose vlan trunking in metadata/configdrive | https://review.opendev.org/c/openstack/nova-specs/+/47181516:08
Uggla#info Please review 984384: Add spec for bidirectional RPC liveness handshake | https://review.opendev.org/c/openstack/nova-specs/+/98438416:08
UgglaCore maintainers, I know you are busy, but it would be great to review the 2 above specs, especially the first one where Sean is already +216:09
UgglaThis spec is the one more awaited by cardoe, that we miss several time.16:10
Uggla#info Starting: https://etherpad.opendev.org/p/nova-2026.2-status#L16 interesting bugs to review.16:10
UgglaThat's all for that topic unless you have something else to share. Or any reaction ?16:11
Ugglaseems not so moving on.16:12
Uggla#topic Stable Branches 16:12
* Uggla giving the mic to elodilles16:12
elodillesthanks16:12
elodillesgrenade-skip-level on stable/2025.2 and grenade on stable/2025.1 are failing due to 2024.2 Dalmatian EOL16:12
elodilles#info grenade-skip-level job removed from stable/2025.2, gate is unblocked on the branch16:12
elodilles#info stable/2025.1 is still blocked, grenade on stable/2025.1 has a workaround: https://review.opendev.org/c/openstack/nova/+/98957916:12
elodillesthough it was agreed on QA channel that these grenade's are not relevant and can be removed,16:13
elodillesi've proposed the above fix, as it is still exercising the grenade16:13
elodilles(against 2024.2-eol)16:13
elodilleswhen it breaks -> we can remove for good16:14
elodilles#info stable branch status / gate failures tracking etherpad: https://etherpad.opendev.org/p/nova-stable-branch-ci16:14
UgglaI think I would remove it also for 2025.1.16:15
elodillesUggla: ACK, we can do that16:15
Ugglajust a personal opinion16:15
Ugglaother could not agree16:16
elodilleseither that, or merge the above fix will unblock the gate for now16:16
elodillesfor stable/2025.116:16
elodillesand that's all from me16:16
* elodilles gives back the microphone16:17
Ugglathanks elodilles16:17
elodillesnp16:17
Uggla#topic vmwareapi 3rd-party CI efforts Highlights16:17
Ugglafwiesel, do you have something to share? 16:17
fwieselHi, no updates from my side. I was on PTO.16:17
Ugglathanks, I hope they were good! 16:18
Uggla#topic Gibi/Kamil's news about eventlet removal16:18
* Uggla giving the mic to gibi16:18
* gibi passing to sambork :)16:19
Ugglaor sambork16:19
samborkEventlet update:16:19
samborkhttps://review.opendev.org/c/openstack/nova/+/983179 - This updates the nova-alt-configurations job to run with the noVNC proxy enabled over TLS. It previously had a +2, but after a rebase, it is now missing one.16:19
samborkhttps://review.opendev.org/c/openstack/nova/+/988583/7 - This switches fixtures from in-memory SQLite to per-test, file-backed SQLite with WAL mode to support native threading.16:19
samborkhttps://review.opendev.org/c/openstack/nova/+/987815 - I am still trying to figure out the Python 3.14 / websockify issue with pickling (specifically, how to fix it to avoid pickling entirely).16:19
samborkhttps://review.opendev.org/c/openstack/nova/+/986282 - gibi also update poison eventlet patch16:19
samborkso we progressing but slower in last weeks bigest progress was with testing16:20
samborkdo you have something to add gibi?16:20
samborkand thanks for passing mic :)16:20
gibiI have nothing to add16:21
gibithanks sambork for the summary16:21
Ugglaadmin question? all the eventlet work is covered by bugs right ? Or do we have a SLBP ?16:21
Ugglagibi , sambork ^?16:22
gibiUggla: that is a good question. 16:22
samborkI think most of it but definitely not all16:22
gibiIf we want we can file a SLBP just to be sae16:22
gibisafe16:22
gibii.e. if we have the chance to go back to the rally based perf testing that is hardly qualifys as a bug16:23
* Uggla +116:23
gibiUggla: could you file a simple SLBP for us?16:23
Ugglayes sure although, I will put a "draft" content, then feel free to adapt if I missed or you need to add stuffs?16:24
Ugglas/?/./16:25
UgglaSkipping next topic coz Layos is not available.16:25
gibiUggla: thanks16:25
Ugglaas well as Ironic, Doug is on PTO.16:25
Uggla#topic Bug scrubbing 16:26
Uggla#info up to 185 (-4)16:26
Uggla#link https://etherpad.opendev.org/p/nova-bug-triage-roster16:26
Uggla#link https://truc.uggla.fr/ to follow the trend.16:26
Uggla#info I'm gonna assign new bugs.16:26
UgglaNext meeting (this week): [public] Upstream bug triage. Wednesday, June 03 · 15:30 – 16:00 UTC. Video call link: meet.google.com/zjr-rxus-hzj16:26
UgglaI will also close the bug we discussed last week. I'll do that probably tomorrow16:27
Uggla#topic Open discussion 16:27
Uggla3 topics, I think we can start with (cw0306-lee) blueprint for reduce build fail value by time base 16:27
Ugglacw0306-lee[m], please go ahead16:28
cw0306-lee[m]Thank you!16:28
cw0306-lee[m]I want to get approve for adding periodic job to reduce build failed value by time based.16:29
cw0306-lee[m]That is used for scheduling16:29
cw0306-lee[m]Now, it can be reduced only when vm success to build on that host.16:30
cw0306-lee[m]So i made this blueprint16:31
cw0306-lee[m]Should i have to do something else to get approve?16:32
Ugglacw0306-lee[m] do you have the link to the BP ?16:32
gibicw0306-lee[m]: why do we need another periodic? Isn't the stats flushed to the DB by the update_available_resources periodic? Can that periodic due the aging before flushing?16:33
gibi/due/do16:33
cw0306-lee[m]blueprints.launchpad.net/nova/+spec/reduce-build-fail-time-base16:33
tkajinamgibi, maybe the proposal is not flushing (resetting the value to 0) but gradually decreasing it16:35
Ugglacw0306-lee[m], so I think you want this as a spec less blue print.16:36
tkajinamthe decision may depend on how we control that "graduality"16:36
gibitkajinam: I mean https://github.com/openstack/nova/blob/380f657b5102707a3da676478e72fd96691e966b/nova/compute/manager.py#L2490-L2492 says we only update the stats in the DB at a the update_available_resources periodic 16:37
Ugglagibi, bauzas, tkajinam do you think this can be a SLBP ?16:37
tkajinamgibi, ah, ok so we use separate mechanism to maintain the counter and report it to placement16:38
gibithe counter ends up in the stats. I'm not sure where we update the placement trait16:38
tkajinamI'm inclined to say we need a spec to discuss the mechanism to configure this feature.16:39
cw0306-lee[m]https://github.com/openstack/nova/blob/380f657b5102707a3da676478e72fd96691e966b/nova/compute/stats.py#L15316:40
cw0306-lee[m]So one reason i made this blueprint was that todo16:40
cw0306-lee[m]I thought this is small topic that can be spec less blueprint, but if i need to write spec, i'll add spec until next time16:42
cw0306-lee[m]cw0306-lee[m]: Todo -> fixme16:43
* gibi_ network glitched16:44
Ugglacw0306-lee[m], problem from an admin perspective we are after soft spec freeze so we should not accept new specs. But if others are ok, maybe we can make a small exception.16:44
gibi_sure lets file a spec and see how complicated it is. I'm lost in finding how we manage the trait based on the stats16:45
*** gibi_ is now known as gibi16:45
Ugglayes, fill a spec ASAP, and maybe if it is simple we can re-discuss/approve it next meeting/week.16:46
Ugglabut no promises it will be accepted.16:46
Uggladoes it sounds good to everobody ?16:47
cw0306-lee[m]Sure. Thank you and i'll write the spec until next irc meeting!16:48
Ugglaalso maybe, we will have Dan, he might explain what he wanted to do behind the fixme.16:48
Ugglacw0306-lee[m], please ping me as when you have the spec. I'll add it to the tracking file16:49
cw0306-lee[m]Ok, when i write spec, i'll ping you in irc! Thank you!16:50
Ugglaas we are almost at the top of the hour, I will not cover the 2 topics from Sean and left them for next week. 16:50
UgglaAnything you want to discuss in 5mn ?16:51
gibi-16:51
Ugglaseems not, so I think we are done for today16:51
UgglaTime to close, thanks for joining this meeting. Have a nice day/evening.16:51
gibithanks Uggla 16:51
Uggla#endmeeting16:51
opendevmeetMeeting ended Mon Jun  1 16:51:54 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:51
opendevmeetMinutes:        https://meetings.opendev.org/meetings/nova/2026/nova.2026-06-01-16.00.html16:51
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/nova/2026/nova.2026-06-01-16.00.txt16:51
opendevmeetLog:            https://meetings.opendev.org/meetings/nova/2026/nova.2026-06-01-16.00.log.html16:51
phildree[m]I wanted to ask if i could have a review on https://review.opendev.org/c/openstack/nova-specs/+/984384 but Uggla already wrote about this16:51
elodillesthanks o/16:52
Ugglaphildree[m] I have mentionned it in the review point above16:52
phildree[m]Thank you Uggla  is there something i can do to it or do i just have to wait and see?16:52
phildree[m]If there is nothing i can do then i might start implementing it for our infrastructure and if something happens down the line then i might push the code to the openstack repo as well16:53
UgglaI think Dan was interested by this, so I suggest you add him as a reviewer.16:54
UgglaI'll try to have a look too. So maybe it could ease the review for a last core reviewer.16:55
phildree[m]Ok, thank you very much! :)16:55
Ugglaalso try to ping people around Thursday, as this is a review day.16:56
phildree[m]Good to know, will do!16:56
UgglaThey have a lot to review. But maybe they will find a slot before the spec freeze.16:57
phildree[m]Understood :)16:57
opendevreviewAnton Iacobaeus proposed openstack/nova-specs master: Intel TDX support in libvirt driver  https://review.opendev.org/c/openstack/nova-specs/+/97960817:09
Ugglagibi, sambork --> https://blueprints.launchpad.net/nova/+spec/complete-eventlet-removal   SLBP17:29
cardoeUggla: thanks. I’m on PTO so I missed the meeting. I’ll be out next week.17:30
Ugglacardoe, no worries I was aware of it, so I have just skip your topic. But keep mentioning the spec to have it approved.17:32
cardoeThank you.17:39
*** Unknown123 is now known as Mike--17:43
*** Unknown123 is now known as Mike--17:52
gmaanmelwitt: graceful shutdown spec is ready for 2nd core review https://review.opendev.org/c/openstack/nova-specs/+/986447 18:55
melwittgmaan: I am reading through it now :) thanks for reminding18:55
gmaanmelwitt: thanks 18:56
opendevreviewMerged openstack/nova-specs master: Spec for the graceful shutdown part2: Task tracking  https://review.opendev.org/c/openstack/nova-specs/+/98644719:29
melwittgmaan: we could use some help reviewing sean-k-mooney fix for unified limits quota enforcement https://review.opendev.org/c/openstack/nova/+/975872 the fix is basically to incorporate RequestSpec into the quota checking to handle cases where the flavor extra spec(s) is not a 1-to-1 mapping to the placement resource classes. if you might be able to take a look19:45
gmaanmelwitt: ack, will check.19:46
melwittthx19:47
melwittdansmith: I wondered if you might be able to help review this bug fix about stale file existence (false positive) on NFS https://review.opendev.org/c/openstack/nova/+/988547 this has come up in downstream CI intermittent failures on NFS and then more recently in a customer case. the fix itself is very small19:53
dansmithokay will look in a sec20:04
melwittthx20:05
dansmithmelwitt: hmm, I get the hack but.. could we not do something similar by just trying to cache the disk and fail if it exists (using the same exclusive create)?20:08
dansmithmeaning,20:09
dansmithinstead of trying to create it, and removing it if we succeed,20:09
melwittdansmith: oh, maybe. the exclusive create is the key (obviously) I was kinda torn between putting it for everything vs knowing the swap disk is the uniquely problematic case20:09
dansmithjust create it (empty) and if we succeeded, continue to use that file for create_image()20:09
dansmithand if we failed to create because EEXIST, then we don't call create_image20:10
dansmithI mean, I think you can do this in _create_swap() right?20:10
dansmithjust always call that, and if we fail to create it because it exists, then return, else continue with create_image() ?20:10
melwittI think maybe not bc _create_swap() is only for creating the backing file20:11
melwitt(and I know, I know, we shouldn't have backing files for swap and eph but we do .... that's a separate change we have talked about making)20:11
dansmith..._create_swap() i the thing you're calling right after, right?20:11
dansmithohhh20:12
melwittthat's the fetch func20:12
dansmithswap.cache() "fetches" the backing image by calling _create_swap() I guess20:12
melwittyeah. hah20:12
dansmithokay gross but anyway20:12
melwittyes, _create_swap() is only for creating the backing file if it doesn't exist, then the file existence of the actual child disk is a separate thing20:12
melwittyes, it is gross20:13
dansmithI guess my point is.. do we have to create the fake thing, then remove it, then re-create with swap.cache()?20:13
dansmithcan we not just leave it?20:13
* melwitt thinks20:13
dansmiththe os.path.exists that gets fooled is the one in cache() right?20:15
melwittyes20:16
dansmithand you didn't want to do this in cache (replace os.exists with a test create) just because it will affect more than swap?20:17
melwittor in create_image possibly20:17
dansmiththis just seems like a really really specific thing where we tickle a bit of behavior that could change (or be different on some other NFS implementation) and making it more direct/in-the-critical-path would be better20:18
melwittkinda. was torn about it. bc this is kind of an NFS-only problem (i.e. shared storage that is not ceph, I guess) and a swap-only problem. to do the exclusive probe20:18
dansmithright, but like, CIFS could behave similarly but not exactly the same, for example20:18
melwittyeah ... I'm not opposed, just wasn't sure about adding a more expensive thing to all paths. I think your point is fair 20:20
dansmithbecause we could use os.posix_fallocate() on the fd we get from open instead of spawning the fallocate command, and then it would be way less expensive than it is today20:21
melwittoh, that's interesting20:21
dansmithanyway, I'm just not jazzed about the obscurity of it.. but I'll just approve with a not-so-passive-aggressive comment about maybe following up and exploring ^20:21
melwittI knew you would be good for this review :) I was blown away by the false positive, period. and my knowledge limited compared to what you're talking about. I am ok with changing the approach, I'm sure sean-k-mooney will understand it too20:23
dansmithNFS has this sort of behavior all the time, and it has changed over the years,20:26
dansmithand I suspect also differs based on your mount arguments20:26
dansmithcommented and +W'd20:27
melwittok yeah confirmed it's in cache() where it does the wrong thing, I just re-read the comments in https://redhat.atlassian.net/browse/OSPRH-12784 where I first saw this issue20:27
melwittthanks dansmith 20:28
*** erlon4 is now known as erlon23:47

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!