Monday, 2026-05-18

elodillesstephenfin: don't you want to include this to os-vif's hibiscus-milestone-1 release? https://review.opendev.org/c/openstack/os-vif/+/94158609:10
sean-k-mooneyelodilles: i would love if that was in m110:33
sean-k-mooneyelodilles: stephenfin proably is ok with it too10:33
sean-k-mooneyelodilles: but we need https://review.opendev.org/c/openstack/nova/+/957040 to merge first10:34
sean-k-mooneyelodilles: so as long as it happens this cycle its fine if its not there for m110:34
elodillessean-k-mooney: ah, i see, thanks for the info. i see that the nova patch is on the gate already, so maybe it's OK to wait for that10:36
elodilles(Hibiscus-1 milestone deadline was last week, but actually i've started to merge the remaining open hibiscus-1 release patches today, so we can still wait a bit)10:37
sean-k-mooneyelodilles: we can alway do another release of os-vif10:37
sean-k-mooneystephen woudl liek the typehint to be in a release10:37
sean-k-mooneyand those are merged10:37
elodillessean-k-mooney: that's true, but if it's just a couple of hours waiting, then i'm OK to wait :]10:37
elodillessean-k-mooney: yepp, i saw that those have merged. +110:38
sean-k-mooneyi woudl sayu if its not merge by end fo day just proceed10:40
elodillesACK, will do that10:58
opendevreviewLajos Katona proposed openstack/nova master: Fix live migration with pinned VM and iothreads  https://review.opendev.org/c/openstack/nova/+/98877411:58
zigosean-k-mooney: Looks like we're affected by https://bugs.launchpad.net/nova/+bug/202851813:45
zigoIt feels like it would have been nice to merge this one: https://review.opendev.org/c/openstack/nova/+/90406013:45
zigo(rather than nit-picking about tests...)13:46
sean-k-mooneythat not a nit pick and you note i idd orginally approve this.13:47
sean-k-mooneywe generally dont accpet chagne with no testing13:47
sean-k-mooneyzigo: the reall issue however is artom moved on form an ic role and is now a manager of one of our internal ci teams13:48
sean-k-mooneyzigo: so the reason this has not made any progress is 1 we forgot about it and 2 no one has been working on it since13:48
zigoSure, but it's very frustrating to see a valid fix from 3 years ago still not merged.13:49
sean-k-mooneyplease dont look at my open reviews then :P13:49
sean-k-mooneylet me see if i can have ai rebase this and add the test coverage really quickly13:50
zigoIt applies cleanly on gazpacho, FYI.13:51
zigoGood chance it does on master.13:51
zigoAs for the policy to have tests: I very much support it and I thank you for making sure we follow the policy, no worries.13:51
sean-k-mooneyyep but we at the very least need a new ci run so ill do a trivial rebase add the requested unit test and push it up13:51
zigoThanks a lot ! :)13:52
sean-k-mooneythis si kind fo one of the perfect usecase for ai. we have a patch we knew fixed the problem but it jsut needs a rebase adn trivial testing added13:52
opendevreviewMerged openstack/nova master: Remove Linux bridge VIF support from Nova  https://review.opendev.org/c/openstack/nova/+/95704013:54
sean-k-mooneywoot!!13:55
sean-k-mooneyzigo: that ^ has been on my todolist for like over a year since neutron removed supprot in 2025.2?13:55
zigo:)13:56
opendevreviewsean mooney proposed openstack/nova master: weights: fix TypeError in weight debug log serialization  https://review.opendev.org/c/openstack/nova/+/90406014:08
sean-k-mooneyzigo: so i think ^ should be ok lets see what ci say14:11
zigoGreat ! Thanks.14:12
sean-k-mooneywe do use debug mode in ci so if that prints propelry and the output looks sane i may upgrade back to +2 but +1 for now since i have only run the unit test on that14:13
zigoI believe the issue is debug + json, not just debug.14:13
sean-k-mooneyhttps://review.opendev.org/c/openstack/nova/+/904060/4/nova/tests/unit/test_weights.py#8714:13
sean-k-mooneyshoudl hopefully validate it14:14
sean-k-mooneybut are you sayign its more that its a problem if you have debug logging + oslo log in json mode?14:14
sean-k-mooneythe orginal bug just mentions debug=true https://bugs.launchpad.net/nova/+bug/202851814:15
sean-k-mooneyso i think the patch shoudl be sufficent but we can alwasy extened it again if needed14:16
zigoThe bug tells about oslo.log in json mode, yeah. And that's why we're affected (we use debug + json in production, so it's easier to stream to elastic search).14:17
sean-k-mooneywell https://bugs.launchpad.net/nova/+bug/2028518 does not mention json mode, that just about normla debug mode with an object that we happen to be serisalise as json which is slightly diffent14:18
sean-k-mooneyzigo: do you have the ablity to  test this? 14:19
zigoOh ok.14:19
sean-k-mooneyyou mentioned seeing it in debian?14:19
zigoI mentioned we have the issue in our public cloud with the scheduler.log.14:19
sean-k-mooneyah ok14:19
sean-k-mooneyso we use debug logging in our downstream product as well14:19
zigoThat'd be on Caracal, for which I just backported the patch (not your revision though).14:19
sean-k-mooneyi.e. we have the schduler defautl to debug14:20
sean-k-mooneyim not aware of this impacting that but that based on antelope14:20
sean-k-mooneyhow does it manifest is it jsut a traceback but otherwise it works?14:20
sean-k-mooneyim trying to undersand why this isnt breakign us hence the question14:21
zigoJust traceback, VMs are still spawned, but it doesn't feel very right ! :)14:21
sean-k-mooneyack ok so this is likely a "silgnet issue" we are not aware of in the ci14:22
sean-k-mooneyand yes14:22
sean-k-mooneyi hate "standared tracebacks/errors" 14:22
sean-k-mooneyin logs14:22
sean-k-mooneyi.e. the ones form os-brick14:22
sean-k-mooneyso ya it woudl be good to fix and backport14:22
sean-k-mooneydansmith: oh we dont need to read the full stream for any format? i tought one of the vmdk or ova format need that but that is good to know14:53
sean-k-mooneyim probaly miss rememberign but i thoguh one of the format had a triailer we needed to look at14:53
sean-k-mooneyalthogh that really just an implamtion detial even in that case as the main point of it being stream based wehre we can reject it once we know its wrogn still holds14:54
sean-k-mooneyglance would still reject it at the first point it know tis wrong even if that is a trailer at the end of the format14:54
dansmithsean-k-mooney: nope :)14:55
sean-k-mooneycool14:55
sean-k-mooneywe do have EndCaptureRegion however  https://github.com/openstack/oslo.utils/blob/master/oslo_utils/imageutils/format_inspector.py#L106 14:56
sean-k-mooneythat the thing i was thinking of14:56
sean-k-mooneywhich is for vmdk https://github.com/openstack/oslo.utils/blob/master/oslo_utils/imageutils/format_inspector.py#L942-L94914:57
dansmithsean-k-mooney: we can capture the trailer yep14:57
dansmithbut we don't _have_ to do that, and we also don't need to store the image to do it14:58
sean-k-mooneyack ok that fits with my recolection then14:58
UgglaReminder, upstream meeting in ~30mn15:33
lajoskatonasean-k-mooney: Hi, I updated gibi's patch for iothreads (https://review.opendev.org/c/openstack/nova/+/988774 ) as the unit test failed i the right way :-)15:59
lajoskatonasean-k-mooney: first I can't see if the patch is related to livemigration post_failure, but it is suspicious....16:00
Uggla#startmeeting nova16:01
opendevmeetMeeting started Mon May 18 16:01:04 2026 UTC and is due to finish in 60 minutes.  The chair is Uggla. Information about MeetBot at http://wiki.debian.org/MeetBot.16:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:01
opendevmeetThe meeting name has been set to 'nova'16:01
UgglaHello everyone16:01
sean-k-mooneylajoskatona: we can chat after the meeting16:01
sean-k-mooneybut thanks16:01
tkajinamo/16:01
lajoskatonaack16:01
lajoskatonao/16:01
dansmitho/16:01
sean-k-mooneyo/16:02
elodilleso/16:02
UgglaLet's start16:03
Uggla#topic Bugs (stuck/critical) 16:03
Uggla#info No Critical bug16:03
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
UgglaI have not seen something bad around the gate. Please tell me if I'm wrong16:05
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 This week Thursday is first spec review day.16:06
Uggla#info Nova soft spec freeze is next week, please submit any new specs and BP before next Thursday EOD (28 Mai)16:06
Uggla#info No meeting next week due to public holiday. (most of European countries involved)16:06
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.16:07
Uggla#info Starting: https://etherpad.opendev.org/p/nova-2026.2-status#L16 interesting bugs to review.16:07
Uggla#info Please review 988501: Do not log metadata proxy shared secret | https://review.opendev.org/c/openstack/nova/+/98850116:07
Uggla^ please review this one from elodilles as related to security16:08
elodillesthx o:)16:08
Uggla#info Please review 471815: Expose vlan trunking in metadata/configdrive | https://review.opendev.org/c/openstack/nova-specs/+/47181516:08
Uggla^ is +2 from Sean, so it would be great to have it as we missed it several cycles.16:09
Uggla#topic Stable Branches16:10
* Uggla giving the mic to elodilles16:10
elodilleso/16:10
elodilles#info stable gates should be OK16:10
elodilles#info stable branch status / gate failures tracking etherpad: https://etherpad.opendev.org/p/nova-stable-branch-ci16:10
elodillesplease report here ^^^ in case you see any issue16:10
* elodilles gives back the mic to Uggla 16:10
Ugglathx elodilles16:11
Uggla#topic vmwareapi 3rd-party CI efforts Highlights16:11
Ugglafwiesel do you have something for us ?16:11
fwieselHi, no. Nothing from my side.16:12
fwieselUggla: Back to you.16:12
Ugglathx fwiesel16:12
Ugglaskipping gibi's topic because he is on PTO this week.16:12
Uggla#topic Nova using openstack sdk for neutron16:12
Ugglalajoskatona, anything to share ?16:12
lajoskatonaNo news from me16:12
Ugglathx lajoskatona16:13
Uggla#topic Ironic review priorities16:13
Ugglacardoe, something to share this week ?16:13
cardoeJust an ask about the trunk spec if folks can review it.16:13
Ugglaby the way I have seen your request to backport https://review.opendev.org/c/openstack/nova/+/95568516:14
cardoeWe landed https://review.opendev.org/c/openstack/nova/+/955685 but would be curious to know if that is backportable to see how much that improves things for users rather than waiting for 2026.216:14
Ugglathat sounds possible, I'll try to propose it. You would like it on 2026.1 right ?16:15
cardoeI'm out on vacation for a little bit coming up but once I return I'll be working on more performance items between nova/ironic and will have more to share.16:15
cardoeYes. 2026.116:15
Ugglaok so I'll try to do it, so I hope, you will have it back from vacation.16:16
elodilles^^^ feel free to add me as reviewer to the backport 16:16
Ugglaelodilles sure, as it will avoid me to do something wrong.16:16
sean-k-mooneyi think it likely can be backported yes but ya it would be good to get elods input for usre16:16
cardoeI can create it.16:16
elodilles+116:16
opendevreviewDoug Goldstein proposed openstack/nova stable/2026.1: Implementing get_num_instances for ironic virt driver  https://review.opendev.org/c/openstack/nova/+/98900216:17
Ugglacardoe ok, also the FUP doc patch would be good.16:18
UgglaI think we can move on.16:18
sean-k-mooneywe might want to squah the two when backporting16:18
sean-k-mooneyas its just comments16:18
sean-k-mooneybut sure i woudl be ok with or without the comment fixup16:19
Uggla#topic Bug scrubbing16:19
Uggla#info up to 189 (-4)16:19
Uggla#link https://etherpad.opendev.org/p/nova-bug-triage-roster16:19
Uggla#link https://etherpad.opendev.org/p/nova-bug-triage-roster16:19
Uggla#link https://truc.uggla.fr/ to follow the trend.16:19
cardoeI would need the topic removed from https://review.opendev.org/c/openstack/nova/+/985785 for gerrit to be happy squashing the backport.16:20
UgglaNext meeting (this week): [public] Upstream bug triage. Wednesday, May 20 · 15:30 – 16:00 UTC. Video call link: meet.google.com/zjr-rxus-hzj16:20
Uggla^ I made this one recurrent (every week). Note I could cancel from time to time if our internal bug triage is taking too long 16:21
Uggla#topic Open discussion16:21
UgglaWe have one topic from sean:16:21
Uggla(sean) libvirt driver support for managed flag via cyborg pci attachment handle. 16:21
* Uggla giving t he mic to sean-k-mooney16:22
sean-k-mooneyso the main question is simple16:22
sean-k-mooneyi am currnetly writing a cyborg spec to supprto vfio-varint driver for pci passthough in cyborg16:22
sean-k-mooneythe same as we did in nova a few cycles ago16:22
sean-k-mooneythis requires a very trivial change to nova in the libvirt driver to pass teh managed=True|False flag16:23
sean-k-mooneyare we ok with detailign this only in the cybrog spec and haveing a whishlist bug ro specless bluepirnt on the nova side16:23
sean-k-mooneyor do we need somethign behiond that?16:23
dansmithmanaged=true is a pci whitelist flag right now for static devices yeah?16:24
sean-k-mooneyi.e. we coudl incorperate this isnto the nova spec for cybrog manged timesliced gpus (mdevs)16:24
sean-k-mooneydansmith: yep16:24
dansmithand you want to be able to let cyborg control that flag for things it provides via accelerator handle (whatever we call them)?16:24
sean-k-mooneyyep16:24
sean-k-mooneyjust adding a boolean to that respocne object16:25
dansmithwhich basically just controls what we tell libvirt about the device right?16:25
sean-k-mooneywith a new microversion of coucse on the cybrog api side16:25
sean-k-mooneyyep16:25
dansmithmakes sense to me16:25
sean-k-mooneycurrently it jsut has the pci adress16:25
* Uggla ok with a specless BP.16:25
sean-k-mooneyso we woudl jsut be addeding a managed=True|false boolean ot it as well16:25
sean-k-mooneyi have the cybrog spec locally with the minimal change to the libvirt driver detailed16:27
sean-k-mooneyill push it shortly16:27
sean-k-mooneyah 16:27
sean-k-mooney#link https://review.opendev.org/c/openstack/cyborg-specs/+/98900316:27
sean-k-mooneyhttps://review.opendev.org/c/openstack/cyborg-specs/+/989003/1/specs/2026.2/approved/support-vfio-variant-driver-managed-mode.rst#16116:28
*** haleyb|out is now known as haleyb16:28
sean-k-mooneyso that is the minimal change for nova16:28
Ugglasean-k-mooney, can you ping me as soon as you have done the BP, I'll update the status etherpad.16:28
sean-k-mooneysure i can create it after the meeting16:29
sean-k-mooneybut i just wanted to confirm we were ok wwith that for trackign if we are then ill go do the thing16:29
Ugglano hurries. (at least for me)  dansmith are you ok ?16:29
dansmithyeah16:30
Ugglado we need another core to approve or we are good ?16:31
sean-k-mooneywe can loop back to this next week if we like where there are more around16:32
dansmithI think PTL can make the call about BP/spec/etc,16:32
dansmithbut I'd say proceed as if we're good and let someone else complain if they want16:32
sean-k-mooneyill create the blueprint now and we can see if we are happy with the text16:32
dansmithI doubt anyone will16:32
UgglaWe could ask forgiveness as pythonista.  ;)16:34
Ugglaanything else to discuss ?16:34
Ugglaseems not, so I think we are done for today16:35
UgglaTime to close, thanks for joining this meeting. Have a nice day/evening.16:35
Uggla#endmeeting16:35
opendevmeetMeeting ended Mon May 18 16:35:50 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:35
opendevmeetMinutes:        https://meetings.opendev.org/meetings/nova/2026/nova.2026-05-18-16.01.html16:35
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/nova/2026/nova.2026-05-18-16.01.txt16:35
opendevmeetLog:            https://meetings.opendev.org/meetings/nova/2026/nova.2026-05-18-16.01.log.html16:35
lajoskatonao/16:35
elodillesthanks o/16:36
Uggladansmith ok with the pace ^ 16:36
dansmithUggla: A+16:38
dansmithespecially with light attendance, good to keep it moving, thanks16:39
sean-k-mooneyUggla: https://blueprints.launchpad.net/nova/+spec/support-vfio-variant-driver-managed-mode-via-cyborg16:40
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Consolidate vTPM and ephemeral encryption secret creation  https://review.opendev.org/c/openstack/nova/+/91209417:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: libvirt: Add optional 'description' kwarg to create_secret()  https://review.opendev.org/c/openstack/nova/+/91998617:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Follow up for encryption secret create consolidation  https://review.opendev.org/c/openstack/nova/+/92158617:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Remove use of (hw:|hw_)ephemeral_encryption_format  https://review.opendev.org/c/openstack/nova/+/92134317:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Add EncryptDetails object for ephemeral encryption  https://review.opendev.org/c/openstack/nova/+/92134417:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Add database migration to ALTER encryption_options  https://review.opendev.org/c/openstack/nova/+/92158717:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support create with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/87093217:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Validate key manager create access in API  https://review.opendev.org/c/openstack/nova/+/91998917:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Clean up unused ephemeral encryption libvirt secrets  https://review.opendev.org/c/openstack/nova/+/91999017:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: libvirt: "Auto heal" ephemeral encryption secrets on guest launch  https://review.opendev.org/c/openstack/nova/+/91999117:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support (resize|cold migration) with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/87093317:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support live migration with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/90551217:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support rebuild with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/87093917:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Add Glance standardized encryption image properties  https://review.opendev.org/c/openstack/nova/+/92134517:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support encrypted source images for qcow2  https://review.opendev.org/c/openstack/nova/+/91999217:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Add encryption support to qemu-img rebase  https://review.opendev.org/c/openstack/nova/+/87093617:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Add backing_encryption_secret_uuid to BlockDeviceMapping  https://review.opendev.org/c/openstack/nova/+/90796017:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support encrypted backing files for qcow2  https://review.opendev.org/c/openstack/nova/+/90796117:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support rescue with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/87367517:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support snapshot with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/87093717:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: Support cross cell resize with ephemeral encryption for qcow2  https://review.opendev.org/c/openstack/nova/+/90959517:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: libvirt: Introduce support for raw with LUKS  https://review.opendev.org/c/openstack/nova/+/88431317:13
opendevreviewPavlo Shchelokovskyy proposed openstack/nova master: libvirt: Introduce support for rbd with LUKS  https://review.opendev.org/c/openstack/nova/+/88991217:13
sean-k-mooneymelwitt: ^ :) i guess that the rebase of your series that ye were chatting about on the mailing list17:14
sean-k-mooneyalthough i dont think there is a reporposal of the spec is there?17:14
lajoskatonasean-k-mooney: o/ , so for the iothread patch from gibi (https://review.opendev.org/c/openstack/nova/+/988774 ) which I updated to fix the failing unit test17:31
lajoskatonasean-k-mooney: the nova-multi-cell failure (test_instance_boot_after_snapshot_deletion) seems unrelated, but I have to dig the nova-live-migration as to tell the truth I am not sure what that job is doing, the tempest part is green, but the the job-output seems like has much more than tempest....17:33
sean-k-mooneylajoskatona: so the post failure in the live migraiton job was not releated17:33
sean-k-mooneythe test actully passed but the scp of the logs fialed17:34
sean-k-mooneylajoskatona: right os after the tempest test runs in the live mgiration job17:34
sean-k-mooneywe also have playbooks that test evaucate and a few other things17:34
sean-k-mooneythose also appared to have passed so i rechecked the job to confirm it was just a log upload issue17:34
sean-k-mooney2026-05-18 13:15:38.440961 | compute1 | ERROR17:35
sean-k-mooney2026-05-18 13:15:38.441225 | compute1 | {17:35
sean-k-mooney2026-05-18 13:15:38.441296 | compute1 | "msg": "ssh: connect to host 23.253.56.19 port 22: Connection timed out\r\nrsync: connection unexpectedly closed (0 bytes received so far) [Receiver]\nrsync error: unexplained error (code 255) at io.c(232) [Receiver=3.2.7]\n",17:35
sean-k-mooney2026-05-18 13:15:38.441357 | compute1 | "rc": 25517:35
sean-k-mooney2026-05-18 13:15:38.441417 | compute1 | }17:35
sean-k-mooneythat i think was the actual reason it fiald but ill check again when it reports back17:35
sean-k-mooneyah the live migration job in the recheck has already passed https://zuul.opendev.org/t/openstack/build/ea322beec7b145c49f2316d6b1b5ec7e17:36
sean-k-mooneyso ya it was just a network blip on teh ci vms17:37
sean-k-mooneylajoskatona: if your intreested in whta th job was doing we run this as a post playbook in the job https://github.com/openstack/nova/blob/master/playbooks/nova-live-migration/post-run.yaml which invokes this role to test evacuate https://github.com/openstack/nova/blob/master/roles/run-evacuate-hook/tasks/main.yaml17:41
lajoskatonasean-k-mooney: thanks, I check these good to look exotic non-neutron jobs sometimes :-)17:52
sean-k-mooneyso form a neutron perspective17:52
sean-k-mooneynova-live-migration tests live migration with ovn17:52
sean-k-mooneynova-next tests live migraiton with ml2/ovs and the ovs firewall driver17:53
sean-k-mooneyand nova-alt-config tests ml2/ovs with the iptabels firewall driver17:53
sean-k-mooneyall with local nova storage17:53
sean-k-mooneyand nova-live-migration-ceph is ovn + cpeh for the vm storage17:53
sean-k-mooneyso we try to cover the most common cofnigurations17:54
blanson[m]Hello guys, quick question. I'm seeing a seemingly "normal" alert on my rabbitmq cluster for nova because scheduler_fanout stream queue il pilling up with nova_instance_info messages ? from what I could read in the code this is normal, expected behavior ? Is it up to the operator to tweaks retention on this queue to not have it pile up to infinity ? 18:20
blanson[m]https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L2428 <- this specifically from my understanding produces those messages which are never consumed ? or is something broken on my end ? 18:21
sean-k-mooneysoh those are consumed by the schduelr if you have that enabled19:03
blanson[m]I see, but it's not consumed by default ? in which case retention policy should be fine ?19:08
sean-k-mooneyblanson[m]: sorry was disctrcted19:35
sean-k-mooneyblanson[m]: i woudl have to confirm but i belive this is contole by https://docs.openstack.org/nova/latest/configuration/config.html#filter_scheduler.track_instance_changes19:36
sean-k-mooneyblanson[m]: and by defautl these are both sent by the compute nodes and consubed by the schduler19:36
blanson[m]I arrived to the same bit of code by digging, I will give that a try to see if that indeed starts consumming messages, thank you ! 19:37
sean-k-mooneyhowever if you have multiple cells then it may not work19:37
sean-k-mooneyblanson[m]: the other thing that is related to this si you shoudl only enable notificaion if there is a consumer in general19:40
sean-k-mooneyif you have notificaion enabel dits ecpected tha tyou either have a consumer for those or you have implemetin a retaion policy on rabbit ectra to discard them after a given treashold19:41
sean-k-mooneyblanson[m]: actully looking again https://docs.openstack.org/nova/latest/configuration/config.html#DEFAULT.scheduler_instance_sync_interval i sthe confgi to contol this and you can disble it by seting that to -119:42
blanson[m]I will have a look into the track_instance_change, but my next plan was to indeed disable the whole thing otherwise ! thank you very much ! 19:48
blanson[m]well, apparently that's an oslo_messaging/rabbitmq issue. I had the correct consumer configuration on the scheduler side, but messages in stream fanout queues are never ack'd. as per https://docs.openstack.org/openstack-ansible/latest/user/messaging/messaging.html#managing-rabbitmq-stream-policy tthose keep piling up until 500MB unless there is a retention policy on the rabbitmq side. TIL ! 20:28
opendevreviewsean mooney proposed openstack/nova master: weights: fix TypeError in weight debug log serialization  https://review.opendev.org/c/openstack/nova/+/90406022:30
opendevreviewsean mooney proposed openstack/nova master: weights: add regression test for JSON log TypeError (bug 2028518)  https://review.opendev.org/c/openstack/nova/+/98902922:30
opendevreviewMerged openstack/nova master: Add regression test for bug #2134375  https://review.opendev.org/c/openstack/nova/+/98490023:37

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