| opendevreview | Adam McArthur proposed openstack/placement master: Placement spec for bulk read API of resource providers https://review.opendev.org/c/openstack/placement/+/990761 | 02:56 |
|---|---|---|
| *** ianw is now known as Guest4536 | 03:59 | |
| opendevreview | Bhavana proposed openstack/nova-specs master: Add spec for MIG-aware PCI Host Weigher https://review.opendev.org/c/openstack/nova-specs/+/1013456 | 04:58 |
| opendevreview | Lucas HENRY proposed openstack/nova master: tox: Fix test imports in placement's nova functional env https://review.opendev.org/c/openstack/nova/+/1013490 | 10:08 |
| *** bauzas0 is now known as bauzas | 10:09 | |
| opendevreview | Thomas Goirand proposed openstack/nova master: Clear the guestfs forceTCG flag in the libvirt fixture https://review.opendev.org/c/openstack/nova/+/1012543 | 10:10 |
| opendevreview | Kamil Sambor proposed openstack/nova master: Create an executor wrapper that has task limit per type https://review.opendev.org/c/openstack/nova/+/975924 | 11:12 |
| opendevreview | Elod Illes proposed openstack/nova stable/2026.1: Fix KeyError: 'body' in os-volumes_boot API proxy https://review.opendev.org/c/openstack/nova/+/1010162 | 13:57 |
| sean-k-mooney | stephenfin: have you seen this before https://paste.opendev.org/show/bSAOd8GzUtIEvBYl3kDj/ | 14:19 |
| stephenfin | sean-k-mooney: https://opendev.org/openstack/openstack-zuul-jobs/commit/bd16a13f0a4605e1d62d762d493787104e03ff23 | 14:20 |
| sean-k-mooney | i saw that in one of the coverage jobs | 14:20 |
| sean-k-mooney | ah | 14:20 |
| stephenfin | I spent some of my Sunday working on that 😅 | 14:20 |
| sean-k-mooney | cool so i cna recheck that | 14:20 |
| stephenfin | Yes, but I don't think it's a complete fix unfortuantely | 14:20 |
| stephenfin | I'm still seeing failures in other jobs, e.g. https://review.opendev.org/c/openstack/openstacksdk/+/952158 | 14:21 |
| stephenfin | I haven't debugged why but I'm guessing tox isn't being capped there or is being reinstalled | 14:21 |
| sean-k-mooney | we coudl tweak the umask for the process but we likely need to wait for tox to resolve it properly | 14:21 |
| stephenfin | jfyi in case it affects nova gates also | 14:21 |
| stephenfin | yeah, the tox maintainers are usually pretty responsive so I would expect to see this fixes in the next day or two | 14:22 |
| sean-k-mooney | tox is not in uc | 14:22 |
| sean-k-mooney | and not capped in our jobs in general | 14:22 |
| sean-k-mooney | but i think the failture predates the patch you linked | 14:22 |
| sean-k-mooney | 2026-11-02 10:16:01 | 14:23 |
| sean-k-mooney | so i guess not | 14:23 |
| sean-k-mooney | it was after the change | 14:23 |
| sean-k-mooney | i saw it on https://review.opendev.org/c/openstack/nova/+/1012543 | 14:23 |
| opendevreview | Merged openstack/nova stable/2026.1: Reproduce and document trusted cert bugs https://review.opendev.org/c/openstack/nova/+/1012883 | 14:27 |
| tkajinam | interesting that the issue affects only some of the targets (I saw -py3N passes but -cover fails) | 14:33 |
| tkajinam | https://zuul.opendev.org/t/openstack/builds?job_name=openstack-tox-cover&skip=0 | 14:35 |
| sean-k-mooney | tkajinam ya im working on a patch to fix it | 14:35 |
| sean-k-mooney | the coverage job inherits form tox-coverage form zuul-jobs not from openstack-tox | 14:35 |
| tkajinam | ahh ok | 14:36 |
| tkajinam | that makes sense | 14:36 |
| sean-k-mooney | im going to port the cap to the other jobs that dont inherit but we may want to condier making them all inhirit form a common job in the future | 14:37 |
| tkajinam | i mean I don't know why we don't use openstack-tox but that situation clearly explain why it fails | 14:37 |
| sean-k-mooney | i think this is a historical thing btu i belive tox-cover in zuul-jobs installs some extra things that is not in the base tox job | 14:37 |
| sean-k-mooney | this is not the first time we have been hit by the fact it does not inherit however | 14:38 |
| sean-k-mooney | i belive the functional with sudo job also does not inherit | 14:38 |
| sean-k-mooney | actully it does | 14:40 |
| sean-k-mooney | so just docs coverage and molecule | 14:40 |
| sean-k-mooney | ok https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013552 shoudl fix the rest | 14:43 |
| opendevreview | Takashi Kajinami proposed openstack/os-vif master: Fix setuptools deprecation warnings https://review.opendev.org/c/openstack/os-vif/+/1013555 | 14:50 |
| Uggla | Reminder: nova upstream meeting in ~10mn | 15:49 |
| melwitt | sean-k-mooney: just a reminder ... this is the step 1 patch for smartnic https://review.opendev.org/c/openstack/nova/+/1005623 | 15:52 |
| sean-k-mooney | ack ill review that again today. also ill see if i cna re reviwe the unified limits bug fix pathces | 15:53 |
| melwitt | sweet thank you | 15:56 |
| Uggla | #startmeeting nova | 16:00 |
| opendevmeet | Meeting started Mon Nov 2 16:00:10 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 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| opendevmeet | The meeting name has been set to 'nova' | 16:00 |
| Uggla | Hello everyone | 16:00 |
| sambork | o/ | 16:00 |
| elodilles | o/ | 16:00 |
| gmaan | o/ | 16:00 |
| dansmith | o/ | 16:00 |
| sean-k-mooney | o/ | 16:00 |
| tkajinam | o/ | 16:01 |
| Uggla | Let's start | 16:02 |
| Uggla | #topic Bugs (stuck/critical) | 16:02 |
| Uggla | #info No critical bugs (https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New&field.importance=Critical&orderby=-datecreated) were reported. | 16:02 |
| Uggla | #topic Gate status | 16:02 |
| Uggla | #link https://bugs.launchpad.net/nova/+bugs?field.tag=gate-failure — Nova gate bugs | 16:02 |
| Uggla | #link https://etherpad.opendev.org/p/nova-ci-failures-minimal | 16:02 |
| Uggla | #link https://zuul.openstack.org/builds?project=openstack%2Fnova&project=openstack%2Fplacement&branch=stable%2F*&branch=master&pipeline=periodic-weekly&skip=0 — Nova and Placement periodic jobs | 16:02 |
| Uggla | #info Investigate gate failures and report bugs tagged gate-failure. | 16:03 |
| Uggla | #info Include a useful explanation when requesting a recheck. | 16:03 |
| tkajinam | there are number of jobs affected by the latest release of tox. at this moment some jobs, particularly cover job is still broken and needs fix | 16:03 |
| tkajinam | https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013552 | 16:03 |
| tkajinam | if you scroll up the irc log then you may see related discussions there | 16:03 |
| Uggla | #link I did that report: https://nova-ci.uggla.fr/ it probably need some refinement, but that's a start. | 16:03 |
| sean-k-mooney | that or a new tox release whichever happens first | 16:03 |
| sean-k-mooney | #link https://github.com/tox-dev/tox/issues/4357 | 16:04 |
| tkajinam | Uggla, oh that's a nice dashboard. | 16:04 |
| sean-k-mooney | is the tox tracker for that | 16:04 |
| sean-k-mooney | https://github.com/tox-dev/tox/pull/4360 was jsut merged | 16:04 |
| Uggla | thx tkajinam | 16:04 |
| Uggla | I hope it will help to find job/test prio to work on. | 16:05 |
| * gibi joins a bit late | 16:05 | |
| * bauzas waves late | 16:06 | |
| Uggla | feedback welcome for improvements. | 16:06 |
| Uggla | can we move on ? | 16:07 |
| Uggla | #topic Release Planning | 16:07 |
| Uggla | #link https://releases.openstack.org/indri/schedule.html | 16:07 |
| gmaan | sean-k-mooney: back to tox cover job, is it broken and need to be nv or we will wait for the fix to merge | 16:07 |
| Uggla | #info The release schedule includes Nova-specific deadlines. | 16:08 |
| Uggla | #link https://etherpad.opendev.org/p/nova-2027.1-ptg | 16:08 |
| Uggla | sorry gmaan I was a bit fast. | 16:08 |
| sean-k-mooney | gmaan: its broken and need either that patch or a new tox release | 16:08 |
| gmaan | no worry, i have not read full chat yet but just wanted to check in case we need to make it nv to unblock gate | 16:09 |
| sean-k-mooney | it was failing on changes this morning on nova | 16:09 |
| gmaan | k, i think in that case we can make it nv and revert once fix or tox new release | 16:09 |
| sean-k-mooney | sure i would expect this to be resolved today in either case | 16:10 |
| sean-k-mooney | either via the exclude or a new release | 16:10 |
| gmaan | ack | 16:10 |
| sean-k-mooney | so we may not need to make it non voting in nova if that happens | 16:10 |
| sean-k-mooney | but if tis still broken tomorrow i would be fien wiht that | 16:10 |
| tkajinam | +1 | 16:10 |
| gmaan | k, will push change to make it nv | 16:11 |
| gmaan | Uggla: please move, that is all i wanted to check | 16:12 |
| Uggla | thx gmaan | 16:13 |
| Uggla | there was nothing special in the review prio except that all the deadlines are now un the "schedule.html" doc. | 16:14 |
| Uggla | *review planning | 16:14 |
| Uggla | #topic Review priorities | 16:14 |
| Uggla | #info Starting: https://etherpad.opendev.org/p/nova-2026.2-status#L16 interesting bugs to review | 16:14 |
| Uggla | Request to review https://review.opendev.org/c/openstack/nova/+/1004890, https://review.opendev.org/c/openstack/nova/+/1004899 and https://review.opendev.org/c/openstack/nova/+/1005073 , which are ironic backports of the perf fixes. | 16:15 |
| Uggla | from ironic folks ^ | 16:15 |
| Uggla | We also have tkajinam patches about libvirt version bump | 16:15 |
| Uggla | 1008847: Bump MIN_{LIBVIRT,QEMU} for "Indri" | https://review.opendev.org/c/openstack/nova/+/1008847 | 16:15 |
| Uggla | 1008862: Remove version check for old libvirt/qemu | https://review.opendev.org/c/openstack/nova/+/1008862 | 16:15 |
| Uggla | 1008867: libvirt: Remove vTPM version detection for < 8.6.0 | https://review.opendev.org/c/openstack/nova/+/1008867 | 16:15 |
| Uggla | 1009084: libvirt: Stop testing missing sev capability in libvirt < 10.0.0 | https://review.opendev.org/c/openstack/nova/+/1009084 | 16:15 |
| opendevreview | Ghanshyam Maan proposed openstack/nova master: Make failing tox-cover job non voting https://review.opendev.org/c/openstack/nova/+/1013576 | 16:15 |
| sean-k-mooney | oh ya it would be good to get the bump done early | 16:16 |
| sean-k-mooney | ill open that serise for review later | 16:16 |
| Uggla | sean-k-mooney ++ | 16:17 |
| Uggla | #topic Stable Branches | 16:17 |
| * Uggla is giving the mic to elodilles | 16:17 | |
| elodilles | thx o/ | 16:17 |
| elodilles | #info No stable gate problems were known at the time of the meeting. | 16:17 |
| elodilles | well, i haven't seen tox issues on stable gate, but there might be some... | 16:18 |
| elodilles | #info unmaintained/2025.1 branches were cut, stable/2025.1 ones will be deleted soon (please do not propose patches to stable/2025.1 anymore, rather to unmaintained/2025.1) | 16:18 |
| elodilles | #info Track stable branch status and gate failures in the following pad: | 16:18 |
| elodilles | #link https://etherpad.opendev.org/p/nova-stable-branch-ci | 16:18 |
| elodilles | and that's all from me about stable for now | 16:18 |
| Uggla | thx elodilles | 16:20 |
| elodilles | np | 16:20 |
| Uggla | #topic vmwareapi 3rd-party CI efforts Highlights | 16:20 |
| Uggla | fwiesel anything to share ? | 16:20 |
| fwiesel | Hi, no updates from my side. | 16:20 |
| Uggla | thx so moving. | 16:20 |
| Uggla | #topic Bug scrubbing | 16:21 |
| Uggla | #info The reported count is at 59 (+0). | 16:21 |
| Uggla | #link https://etherpad.opendev.org/p/nova-bug-triage-roster | 16:21 |
| Uggla | #link https://truc.uggla.fr/ — Bug trend tracking | 16:21 |
| Uggla | #info The next public upstream bug-triage meeting is scheduled for Wednesday, November 04, 2026. I will probably move it a little to adapt to daylight saving change. | 16:22 |
| Uggla | #link https://meet.google.com/zjr-rxus-hzj Video call | 16:22 |
| opendevreview | Merged openstack/os-vif master: Remove use of testscenarios https://review.opendev.org/c/openstack/os-vif/+/1008029 | 16:22 |
| Uggla | And sorry last week I had to cancel it, due to my laptop migration... | 16:22 |
| Uggla | #topic Open discussion | 16:23 |
| YushoYamaguchi | Could I speak? | 16:23 |
| Uggla | It seems we have 2 topic from YushoYamaguchi | 16:24 |
| Uggla | yes please go ahead the floor is yours | 16:24 |
| YushoYamaguchi | https://review.opendev.org/c/openstack/nova-specs/+/1009977 | 16:24 |
| YushoYamaguchi | I want to talk about the direction of this spec. | 16:24 |
| YushoYamaguchi | sean-k-mooney: I've read your comment. Thank you. | 16:25 |
| sean-k-mooney | i think the way to move it forward, IMO is to just drop the image proerty and use the computecapablities filter to filter based on supproted queue count as we did for the adress specas feature | 16:26 |
| YushoYamaguchi | For introducing virtio queue num limitation, scheduler fix is required, right? | 16:26 |
| sean-k-mooney | i belive so. the thing we want to avodi it sa user create a vm on host a and get 16Ques | 16:26 |
| sean-k-mooney | then later an admin moves it to host b | 16:26 |
| sean-k-mooney | keeping the 16Q since the xml is not update | 16:27 |
| sean-k-mooney | and later after a hard reboot it changes to say 8 | 16:27 |
| sean-k-mooney | the queue count shoudl ideally be stabel unless you resize | 16:27 |
| YushoYamaguchi | I think today's codebase has the same issue. | 16:28 |
| sean-k-mooney | today unless clamped by the host value withc is unlimited by defualt | 16:28 |
| sean-k-mooney | the queue count is equeal to the vcpu count | 16:28 |
| sean-k-mooney | so for most deployments its fixed for the life tiem of the instance | 16:28 |
| sean-k-mooney | but you are correct | 16:28 |
| sean-k-mooney | if you use the hsot value and its diffent on differnt hosts | 16:29 |
| sean-k-mooney | then it could change | 16:29 |
| tkajinam | sean-k-mooney, I think the discussion during the PTG lead to the agreement that the feature do not guarantee the requested queues but just do min(request queues, max queues in that compute host) | 16:29 |
| sean-k-mooney | yes | 16:29 |
| sean-k-mooney | but that does not mean we shoudl shoudl not shcudle based on it | 16:29 |
| tkajinam | ok | 16:30 |
| sean-k-mooney | the config value need to set the max as it is there to allow the operator to specify what is supproted by the host | 16:30 |
| sean-k-mooney | im jsut sayign we probaly shoudl select a host that can supprot the current queu count and probaly record that for the lifetime of the isntnace | 16:31 |
| sean-k-mooney | my concern is a user deploys an applcaition and tunes it for a given queue count | 16:31 |
| sean-k-mooney | then an admin moves it an breaks there applciation | 16:31 |
| sean-k-mooney | that why im suggesting we recored the calulated queue count in the instnace_system_metadata or similar and schdule based on that | 16:32 |
| sean-k-mooney | if other are not concerned with hat we could perhaps weaken the guarenteed but it make the feature less user friedly if we do | 16:33 |
| dansmith | you want to expose the min queues the host has configured so the scheduler would not send the instance there? | 16:33 |
| YushoYamaguchi | Sorry, disconnected. | 16:34 |
| dansmith | that would affect the initial scheduling as well and sort of eliminate the min() behavior we talked about | 16:34 |
| sean-k-mooney | dansmith: i want to expsoe the max queues a host supprot | 16:34 |
| sean-k-mooney | and make sure we dont schdule to a host that cant supprot it but ya more or less | 16:35 |
| sean-k-mooney | that woudl remove the need for the min(conf,flavor) yes | 16:35 |
| YushoYamaguchi | Sorry. Maybe I missed some message. | 16:36 |
| sean-k-mooney | https://specs.openstack.org/openstack/nova-specs/specs/2023.2/implemented/libvirt-maxphysaddr-support.html#nova-scheduler-changes | 16:36 |
| sean-k-mooney | ^ basiclly do what we did for that | 16:36 |
| tkajinam | and makes the queue count consistent during migration, I think | 16:36 |
| opendevreview | Mike Lowe proposed openstack/nova master: Pass image to unified limits resize/restore checks https://review.opendev.org/c/openstack/nova/+/1012988 | 16:37 |
| sean-k-mooney | tkajinam: yep it woudl as a side effect | 16:37 |
| sean-k-mooney | althoguh we cant change it in the xml regardless | 16:37 |
| sean-k-mooney | that woudl be a guest visabel hardware change and i dont think qemu supprots that | 16:37 |
| sean-k-mooney | (changing queue count live) | 16:37 |
| YushoYamaguchi | I understand. | 16:38 |
| YushoYamaguchi | I'll fix my spec. Thank you. | 16:38 |
| sean-k-mooney | dansmith: tkajinam are ye ok with that direction ^ | 16:39 |
| sean-k-mooney | count woudl become the actual value | 16:39 |
| sean-k-mooney | and the schduler via the compute cpabalities filter woudl ensure the host supprot that count | 16:39 |
| YushoYamaguchi | Do you think we should address that change and my spec together? | 16:40 |
| YushoYamaguchi | Should I wait for that change before moving forward with my proposal? | 16:41 |
| sean-k-mooney | yes i thinik it shoudl be in your spec personaly. that woudl also mean removing the image property | 16:41 |
| YushoYamaguchi | okay. | 16:41 |
| sean-k-mooney | bsiclly the change would be "if hw:vif_count is defiend filter based on that" | 16:41 |
| sean-k-mooney | *hw:vif_queue_count | 16:41 |
| tkajinam | I would personally prefer strict approach (if a user requests 4 then give 4). | 16:41 |
| sean-k-mooney | yep ^ is my prefence as well | 16:42 |
| opendevreview | Merged openstack/os-vif master: tox: Add functional-py* targets https://review.opendev.org/c/openstack/os-vif/+/1008097 | 16:43 |
| tkajinam | I think the good first step would be to document the behavior during migration (min vs guarantee) , and then discuss further about the preferred behavior. | 16:43 |
| tkajinam | in spec review | 16:43 |
| sean-k-mooney | we shoudll capature the non slected option in the alternitives regardless | 16:44 |
| tkajinam | yup | 16:44 |
| YushoYamaguchi | I don't fully understand yet, so could you explain it in more detail in the review comments? Sorry | 16:45 |
| sean-k-mooney | we can try yes | 16:46 |
| YushoYamaguchi | Thank you so much. Could I move on to next topic? | 16:46 |
| sean-k-mooney | +1 | 16:47 |
| YushoYamaguchi | https://review.opendev.org/c/openstack/nova-specs/+/1011067 | 16:47 |
| YushoYamaguchi | I failed to talk this in vPTG. Sorry. | 16:47 |
| YushoYamaguchi | I want to disable checksum offload in nova.conf on compute nodes to fix an issue when OVS version up. | 16:48 |
| sean-k-mooney | so in the past we have condier thi to be a qemu/dpdk bug | 16:48 |
| sean-k-mooney | and decised not to fix or expose this in nova | 16:48 |
| sean-k-mooney | basiclly qemu and dpdk are ment to negoicate the supproted feature set | 16:49 |
| sean-k-mooney | libvirt does not have an xml api to supprot diablign host offload i bleive | 16:49 |
| sean-k-mooney | https://opendev.org/x/networking-ovs-dpdk/src/branch/master/devstack/libs/ovs-dpdk#L81 | 16:50 |
| sean-k-mooney | in the very very early days of ovs-dpdk supprot we sued to need to use a qemu wrapper script to disable this offload | 16:50 |
| YushoYamaguchi | Sorry. This time, guest offload | 16:50 |
| sean-k-mooney | but that was fixed in ovs and qemu a very long time ago | 16:50 |
| sean-k-mooney | YushoYamaguchi: this may be ok as a workaround but im not confived it shoudl be a top level feature | 16:52 |
| YushoYamaguchi | Umm. I see. | 16:52 |
| tkajinam | sean-k-mooney, I think the problem is not negotiation between qemu and dpdk but is rather guest and dpdk/qemu | 16:52 |
| sean-k-mooney | the fact that nova does nto schdule based on the negicated virtio feature is a long know issue | 16:53 |
| sean-k-mooney | tkajinam: yep and that has been a problem since effectivly day one | 16:53 |
| sean-k-mooney | its not the first time this has happend | 16:53 |
| sean-k-mooney | i explciltly assed the ovs core maintiaenr and libvirt teams to expose this so we could schduler on it | 16:53 |
| sean-k-mooney | and they did not | 16:53 |
| tkajinam | ok now I understand what you mean | 16:54 |
| sean-k-mooney | so what happign in your env is the call to libvirts migrateUri3 to kick of the migration is raisign an error right | 16:55 |
| Uggla | guys we need to close, are we good ? | 16:55 |
| sean-k-mooney | Uggla: i think this need more discsion as a possibel bugfix/workaround | 16:55 |
| sean-k-mooney | btu not as a spec/feature | 16:55 |
| sean-k-mooney | im not convicned this should be in nova's scope | 16:55 |
| YushoYamaguchi | I understand. | 16:55 |
| Uggla | we can discuss further next week if needed. | 16:56 |
| YushoYamaguchi | Where could I talk about bugfix/workaround? This IRC or other ? | 16:56 |
| YushoYamaguchi | okay. | 16:56 |
| Uggla | This IRC | 16:56 |
| stephenfin | gmaan: when you've a moment, could you confirm that you're happy to backport that API fix you reviewed last week? https://review.opendev.org/c/openstack/nova/+/1012924 | 16:56 |
| sean-k-mooney | irc is also ok altheoru i have to run to another meeting after | 16:56 |
| YushoYamaguchi | okay, thank you | 16:56 |
| stephenfin | (elodilles is looking for a second opinion on same) | 16:57 |
| YushoYamaguchi | oh really? thank you. | 16:57 |
| sean-k-mooney | YushoYamaguchi: we can maybe tray and follow up tomorrow fi that works | 16:57 |
| Uggla | YushoYamaguchi, sorry to interupt, but you can discuss in this IRC and/or next week | 16:57 |
| sean-k-mooney | gmaan: https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013552/1 is merged so coverge jobs shoudl be unblocked | 16:58 |
| YushoYamaguchi | Ok. | 16:58 |
| Uggla | ok time to close. | 16:58 |
| Uggla | Thanks for joining. Have a nice day/evening. | 16:58 |
| YushoYamaguchi | Thank you. | 16:58 |
| Uggla | #endmeeting | 16:58 |
| opendevmeet | Meeting ended Mon Nov 2 16:58:46 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:58 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/nova/2026/nova.2026-11-02-16.00.html | 16:58 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/nova/2026/nova.2026-11-02-16.00.txt | 16:58 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/nova/2026/nova.2026-11-02-16.00.log.html | 16:58 |
| gmaan | sean-k-mooney: nice, i abandon mine | 16:58 |
| gmaan | stephenfin: ack | 16:58 |
| elodilles | thanks o/ | 16:58 |
| * Uggla trying to respect schedule, mostly due to another meeting now. | 16:59 | |
| opendevreview | Merged openstack/nova master: tests: Functional reproducer for bug 2088831 https://review.opendev.org/c/openstack/nova/+/935565 | 17:14 |
| opendevreview | Takashi Kajinami proposed openstack/os-vif master: Resolve deprecation warnings from setuptools https://review.opendev.org/c/openstack/os-vif/+/1013555 | 17:14 |
| zigo | May I bring the attention to this one easy to review patch? | 18:39 |
| zigo | https://review.opendev.org/c/openstack/placement/+/1010823/3/placement/wsgi/__init__.py | 18:39 |
| zigo | It's kind of super annoying for downstream or operators not to be able to pass any argument to placement. | 18:39 |
| sean-k-mooney | zigo: passing command line argument to wsgi applcaiton is not complient with the wsgi spec | 18:42 |
| sean-k-mooney | while we could propaagete it it will onley work for uwsgi i think and maybe gunicorn? | 18:43 |
| zigo | sean-k-mooney: What ?!? It works like this absolutely everywhere else in OpenStack ! | 18:43 |
| sean-k-mooney | it wont be possibel on mod_wsgi | 18:43 |
| zigo | Of course, yes, that's expected. | 18:43 |
| sean-k-mooney | zigo: it does and it voiltes hwo wsgi applciaiton are ment to be wriitne | 18:43 |
| sean-k-mooney | im not against addint hits but if there is something that can only be done via a command line flag we likely shoudl fix that | 18:44 |
| sean-k-mooney | are you usign it for the config? | 18:44 |
| zigo | How do you pass a different --log-file to different daemons then? That, combined with the fact that oslo.log defaults to some bullshit '<frozen importlib._bootstrap>' ... | 18:45 |
| sean-k-mooney | you can set the log file in oslo.confg | 18:45 |
| sean-k-mooney | i.e. in the placement.conf | 18:45 |
| zigo | Then you get stuff logging to /var/log/FOO/<frozen importlib._bootstrap>.log <--- Literally ! | 18:45 |
| zigo | sean-k-mooney: Works when a project has *ONE* daemon, not when it has multiple. | 18:45 |
| sean-k-mooney | retryyireally | 18:46 |
| sean-k-mooney | ... | 18:46 |
| sean-k-mooney | https://github.com/openstack-k8s-operators/nova-operator/blob/main/templates/placement/api/config/placement.conf#L7 | 18:46 |
| sean-k-mooney | well by default we only have 1 logfile for placement | 18:46 |
| zigo | Placement is an exception because it has only one API though, but why would we want it to be different that everthing else in OpenStack? | 18:46 |
| sean-k-mooney | this is howe we do it everywehre else | 18:46 |
| sean-k-mooney | via config | 18:46 |
| zigo | No. | 18:46 |
| zigo | I strongly don't agree. | 18:47 |
| zigo | You can't have multiple log_file= differenciating daemons. | 18:47 |
| sean-k-mooney | yes you can | 18:47 |
| zigo | Anyways, got to go for dinner. | 18:47 |
| zigo | You mean with the env var? | 18:47 |
| sean-k-mooney | each deamon is ment ot have its own config file | 18:47 |
| zigo | That's not what people do. | 18:48 |
| sean-k-mooney | its what most instllers do | 18:48 |
| sean-k-mooney | or the log to the journal by default | 18:48 |
| zigo | That's not what *I* then... | 18:48 |
| zigo | Oh, you mean that thing from systemd that never was written to scale? :) | 18:48 |
| zigo | Thanks but no thanks. | 18:48 |
| zigo | Anways, got to go, or my wife will get angry. :) | 18:49 |
| sean-k-mooney | :) | 18:49 |
| sean-k-mooney | gmaan: can you fix the bug superviro on placment https://bugs.launchpad.net/placement its currently set to you but it shoudl be an open lauchpad team like nova-bugs https://launchpad.net/~nova-bugs | 19:38 |
| sean-k-mooney | i want to set https://bugs.launchpad.net/placement/+bug/2171047 to wishlist | 19:38 |
| sean-k-mooney | gmaan: we coudl also jsut use nova-bugs for both | 19:39 |
| gmaan | sean-k-mooney: checking | 19:39 |
| sean-k-mooney | since we are usign nova-drivers for both | 19:39 |
| gmaan | humm, I am not able to change it. no edit option | 19:45 |
| gmaan | maybe melwitt might as admin in nova-drivers ? | 19:46 |
| gmaan | or bauzas | 19:47 |
| gmaan | no, bauzas is not admin there | 19:48 |
| sean-k-mooney | you do not need to be an admin | 19:48 |
| sean-k-mooney | just a member of the group | 19:48 |
| gmaan | k, it seems i am not member of that. that is why. | 19:48 |
| sean-k-mooney | if your in nova-drivers you shoudl hae an edit but beside bug supervisor i belive | 19:49 |
| sean-k-mooney | same | 19:49 |
| sean-k-mooney | i woudl fix it if i was but currently i think you are the only person who can triage placement bugs | 19:49 |
| gmaan | humm | 19:49 |
| sean-k-mooney | thats what "Bug supervisor" in launchpad contols as far as im aware and why its normally a team rather then a person | 19:50 |
| gmaan | we should change it to nova-drivers, let's wait if either dansmith melwitt gibi bauzas or Uggla can change it | 19:50 |
| sean-k-mooney | to be fair all the drivers teams im in im also an admin of so i cant fully say i does not requrie admin but i dont think it does | 19:50 |
| sean-k-mooney | nova-drivers is not correct | 19:51 |
| sean-k-mooney | the bug team shoudl be an open team | 19:51 |
| sean-k-mooney | at least hat what most project doo | 19:51 |
| sean-k-mooney | this is spereate form teh sercurity bug configuration | 19:51 |
| sean-k-mooney | you restirct secruity bugs though https://launchpad.net/<project>/+sharing | 19:52 |
| sean-k-mooney | anyway ya no rush | 19:53 |
| sean-k-mooney | i just noticed i could not triage it when reviewing zigo's patch to placement | 19:53 |
| sean-k-mooney | gmaan: on a sperate note the tox change i did fixed most of the jobs but the devstack functionl jobs in the sdk/client are still broken | 19:59 |
| sean-k-mooney | i.e. openstacksdk-functional-devstack | 20:00 |
| gmaan | Right, not nova-drivers. I think instead of creating new group, we can use nova-bug group only | 20:14 |
| opendevreview | Merged openstack/nova stable/2026.2: Fix volume attachment pagination https://review.opendev.org/c/openstack/nova/+/1012924 | 22:12 |
| gmaan | Uggla: once you are online, reminder about it https://review.opendev.org/c/openstack/nova/+/1007750 | 22:55 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!