Wednesday, 2026-04-29

opendevreviewGhanshyam Maan proposed openstack/nova master: DNM: Test removal of enforce_scope  https://review.opendev.org/c/openstack/nova/+/98647901:18
opendevreviewMerged openstack/nova master: performance: reduce calls to libvirt / add caching  https://review.opendev.org/c/openstack/nova/+/92285501:24
opendevreviewMerged openstack/nova master: ironic: Abort build on fatal image validation errors  https://review.opendev.org/c/openstack/nova/+/98375301:24
opendevreviewMerged openstack/nova stable/2025.2: fix: device_by_alias should respect config type  https://review.opendev.org/c/openstack/nova/+/98492001:24
opendevreviewMerged openstack/nova stable/2025.1: api: Handle empty imageRef alongside null for local BDM check  https://review.opendev.org/c/openstack/nova/+/98054101:24
opendevreviewMerged openstack/nova stable/2024.2: Reproduce bug/2115905  https://review.opendev.org/c/openstack/nova/+/96024401:24
opendevreviewMerged openstack/nova stable/2024.2: [PCI tracker]Remove non configured devs when freed  https://review.opendev.org/c/openstack/nova/+/96024501:25
opendevreviewMerged openstack/nova stable/2024.2: [pci]Keep used dev in Placement regardless of dev_spec  https://review.opendev.org/c/openstack/nova/+/96024601:25
opendevreviewGhanshyam Maan proposed openstack/nova master: Standardize the service-to-service communication  https://review.opendev.org/c/openstack/nova/+/98659004:02
opendevreviewGhanshyam Maan proposed openstack/nova master: Standardize the service-to-service communication  https://review.opendev.org/c/openstack/nova/+/98659004:05
opendevreviewNilesh Thathagar proposed openstack/nova master: Dell PowerFlex: Pass requested_size to os-brick on extend  https://review.opendev.org/c/openstack/nova/+/98660406:24
opendevreviewPhilipp Dreesens proposed openstack/nova-specs master: Add spec for bidirectional RPC liveness handshake  https://review.opendev.org/c/openstack/nova-specs/+/98438407:12
opendevreviewLajos Katona proposed openstack/nova stable/2026.1: Doc: add short doc section for mdev_class  https://review.opendev.org/c/openstack/nova/+/98661807:59
opendevreviewAshish Gupta proposed openstack/nova master: Disable greenlet-based notifier sender check under native threading temporarily  https://review.opendev.org/c/openstack/nova/+/98562108:30
elodillesUggla: hi, this is just a heads up, that i might not be able to join to today's bug roster meeting, i've added comments/question to my bugs to those i had time to look at.11:18
opendevreviewKamil Sambor proposed openstack/nova master: Change nova-alt-configurations job  https://review.opendev.org/c/openstack/nova/+/98317911:36
gibiUggla: have to say the same unfortunately 11:52
opendevreviewMerged openstack/nova master: Test nova CLI commands with native threading  https://review.opendev.org/c/openstack/nova/+/98403611:55
opendevreviewribaudr proposed openstack/nova master: Doc: fix misleading PCI address in multi-VF example  https://review.opendev.org/c/openstack/nova/+/98665013:36
sambork@Uggla: Hey, Same here (in case of bug triage) i have updated the etherpad; two ofthe bugs are duplicates, and i'm currently trying to create a reproducer for the third one13:57
Ugglasambork ok no problem.13:57
opendevreviewArtem Goncharov proposed openstack/nova master: Fix invalid jsonschema for keypair list response  https://review.opendev.org/c/openstack/nova/+/98666014:12
opendevreviewribaudr proposed openstack/nova-specs master: WIP: Support different resource classes for VFs on the same PF  https://review.opendev.org/c/openstack/nova-specs/+/98666114:17
opendevreviewArtem Goncharov proposed openstack/nova master: Fix invalid jsonschema for keypair list response  https://review.opendev.org/c/openstack/nova/+/98666014:36
opendevreviewMerged openstack/nova master: Drop redundant N371 check  https://review.opendev.org/c/openstack/nova/+/97230214:37
stephenfingmaan: Looks like gtema found a bug in the keypair JSON schema spec, if you've time to take a look https://review.opendev.org/c/openstack/nova/+/98666015:08
melwittbauzas: hey do you think you might be able to take a look at the bottom 3 patches for vtpm prep? it's a follow up and two small things that got split out during code review. dansmith has reviewed them https://review.opendev.org/c/openstack/nova/+/978836 and https://review.opendev.org/c/openstack/nova/+/983504 and https://review.opendev.org/c/openstack/nova/+/98350515:18
gmaanstephenfin: done, saw from gerrit email15:18
bauzasmelwitt: not sure I'll have time, but I'll try15:18
stephenfingmaan++ ty 🙏15:19
melwittbauzas: ack thanks15:20
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Use native IntOpt for tx_queue_size and rx_queue_size  https://review.opendev.org/c/openstack/nova/+/96901415:21
gmaanmelwitt: stephenfin sean-k-mooney gibi bauzas dansmith Uggla, As we were not able to discuss the 'standardization of service-to-service communication' topic in PTG, I am proposing it in gerrit, please check when you have time  https://review.opendev.org/c/openstack/nova/+/98659015:32
gmaanI am not in hurry to merge it but would like to get a kind of team agreement on it 15:32
sean-k-mooneygmaan: ack. to soften my stand a bit. it may be ok to use the service_user section as a fall back ot create teh user token for requst made by nova on behalvf of a user. but it idally shoudl not be the only way15:33
gmaansean-k-mooney: I think it is aligned what you were suggesting but I am adding a fallback to service_user which you objected to use for service calling15:33
gmaansean-k-mooney: yeah15:34
sean-k-mooneythe reason im brinign that up is we know that cern and other have one nova that talks to neutron in 3 regions15:34
sean-k-mooneyso we can assume that all service a given compute host talks to are in the same region or use the same CA for tls ectra15:34
sean-k-mooneyso each service nova talsk to shoudl ahve its onw cofnig section with the common keystone auth optiosn in it15:35
gmaansean-k-mooney: and i ack your concern for devstack bug for mess up on configuring service users which I will take up as a separate things to solve, I can work on that when get time15:35
sean-k-mooneyand then and only then fallback to service_user15:35
bauzasdo people know that we have upstream bug triage discussion at the moment ? 15:37
bauzasI don't see Uggla telling about it15:37
bauzashttps://meet.google.com/fdz-phux-cnc15:37
opendevreviewJay Faulkner proposed openstack/nova stable/2026.1: ironic: Abort build on fatal image validation errors  https://review.opendev.org/c/openstack/nova/+/98667415:37
sean-k-mooneygmaan: by the way i mentioned this in the agentic coding irc channle. its entirly ok for proejct or teams to do one of ptg style meetign that are open for everyoen to join if they need high bandwith com like the midcycle. ill review your proposal when i have time but if we need oen we can poll the list for a time that works an do liek a 1 hour one off on this topic15:37
gmaan++15:38
sean-k-mooneybauzas: oh ok vagely i can join in a mainly liste mode15:38
sean-k-mooneyim doint somethign else in parallel15:38
gmaanlet's see how feedback goes in gerrit and accordingly we can decide 15:38
gmaan' agentic coding irc channle' can you link name, I can join it15:38
sean-k-mooney#openstack-agentic-workflows15:40
sean-k-mooneywe need to get that added to ease drop15:40
sean-k-mooneybut that will happen once the tc vote happens on the governace repo15:40
gmaanthanks15:50
gmaansean-k-mooney: Uggla bauzas as we spoke in bug triage meeting,. this is functional test for live migration cleanup things https://review.opendev.org/c/openstack/nova/+/98647715:51
cardoeSo for cyborg+nova, nova is still making a KVM VM for example and attaching the accelerator to the VM?16:11
sean-k-mooneycardoe: yes16:13
sean-k-mooneycardoe: nova is entrily respocneble for the vm defintion and lifecyle management16:14
cardoeokay cool. I wish the docs actually said that plainly on the intro page. :D16:14
sean-k-mooneycardoe: cyborg just dose the inventory of the accelerator adn tells nova here is the pci adress16:14
sean-k-mooneyin the future cybrog may to other thing such as cleanign the device it manages for example16:15
sean-k-mooneynot firmware flashing16:15
sean-k-mooneybut secure rease of nvme ssds16:15
sean-k-mooneythat sort of thing16:15
sean-k-mooneyeventually its is hoped that cyborg woudl be able to mange other devices beyond pci like usb or cxl or block devices16:16
sean-k-mooneynova woudl still need to eb able to add those to a vm16:16
sean-k-mooneybut that woudl be the limit of what nova woudl need to do and cybrog woudl do the enurmation and prepreation fo the device as requreid16:17
sean-k-mooneyat least that is the current split of responsiblitys, who knwo what ti will actully become16:17
opendevreviewSylvain Bauza proposed openstack/nova master: Add functional test for NUMA live migration overcommit bug  https://review.opendev.org/c/openstack/nova/+/98669316:26
melwittaw man, microversion 2.98 (server show image props) in openstackclient and openstacksdk have still not happened 😭 https://review.opendev.org/c/openstack/python-openstackclient/+/940799 and https://review.opendev.org/c/openstack/openstacksdk/+/94079817:23
sean-k-mooneyi also have a ptch for server backup i need ot respin at some point17:53
sean-k-mooneymelwitt: server backup works exactly once17:53
sean-k-mooneywhich is what we test17:53
sean-k-mooneymelwitt: stephenfin is gone for today but if we think those are ready to go we shoudl refresh them and ask stephen to then take a look17:55
melwitthah, awesome17:55
melwittyeah that would be good. I would usually assume there is something wrong with the proposed patches and that's why they've not been reviewed and merged yet17:55
melwittso if we can find out what that is, I or you or whoever can update them17:56
sean-k-mooneyyep we have +2 right there but not +w bug stpehn as both so if we can get them into a state we think is good to go we cna then ask stephn to take a look17:57
melwittoh I didn't even notice that it had been +2 in the past17:58
opendevreviewMerged openstack/nova master: Stop asking bug report to upstream  https://review.opendev.org/c/openstack/nova/+/93338118:49
opendevreviewMerged openstack/nova master: Drop workaround for old keystoneauth  https://review.opendev.org/c/openstack/nova/+/92766919:16
opendevreviewMerged openstack/nova master: Fix typo in Caracal prelude  https://review.opendev.org/c/openstack/nova/+/91503619:16
opendevreviewMerged openstack/nova master: Remove old warning log about UEFI support  https://review.opendev.org/c/openstack/nova/+/90856319:31
opendevreviewMerged openstack/nova master: Delete duplicate unit test  https://review.opendev.org/c/openstack/nova/+/98460720:24
opendevreviewMerged openstack/nova master: TPM: fixup for bump service version documentation  https://review.opendev.org/c/openstack/nova/+/97883620:25
opendevreviewMerged openstack/nova master: Remove unused gabbi  https://review.opendev.org/c/openstack/nova/+/97800620:25
opendevreviewMerged openstack/nova master: libvirt: Use native IntOpt for tx_queue_size and rx_queue_size  https://review.opendev.org/c/openstack/nova/+/96901422:57

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