Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2023-07-11
16:33:50 bauzas that was it for me
16:34:12 bauzas as every week, anything anyone ?
16:34:17 gibi -
16:34:35 elodilles -
16:35:13 bauzas cool
16:35:20 bauzas I give you all 25 mins back of your time !
16:35:24 gibi thanks!
16:35:25 bauzas thanks all
16:35:28 opendevmeet Log: https://meetings.opendev.org/meetings/nova/2023/nova.2023-07-11-16.00.log.html
16:35:28 opendevmeet Minutes (text): https://meetings.opendev.org/meetings/nova/2023/nova.2023-07-11-16.00.txt
16:35:28 opendevmeet Minutes: https://meetings.opendev.org/meetings/nova/2023/nova.2023-07-11-16.00.html
16:35:28 opendevmeet Meeting ended Tue Jul 11 16:35:28 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
16:35:28 bauzas #endmeeting
16:35:44 elodilles thanks bauzas & all o/
16:37:42 sean-k-mooney elodilles: back to os-vif i personally dont have an issue with signel core approving yoru gate fix for train given noone else has objected
16:38:01 sean-k-mooney in almost 2 years
16:38:25 elodilles sean-k-mooney: note that since it's 2 yrs old, it probably won't merge
16:38:53 sean-k-mooney probably but lookign at https://review.opendev.org/q/project:openstack%252Fos-vif+branch:stable%252Ftrain+status:open
16:39:18 sean-k-mooney thre are some patches that might be worht landing but i dont rellly want the risk fo breakign the eol tag by rushing
16:39:40 sean-k-mooney so really im wondiering if its worth fixign those issue vs just taging what we have
16:39:46 elodilles yes, don't break things before the EOL :S
16:40:12 elodilles sean-k-mooney: i'm OK with leaving the branch as it is
16:40:48 sean-k-mooney ok that also works for me. the patchs are aviable and people could apply them to the eol tag if they wanted too
16:40:55 sean-k-mooney so im fine with leave it as is
16:41:51 elodilles the patches will be abandoned, yes, but who else needs it can cherry pick them downstream later
16:43:35 sean-k-mooney actully we have the switchdev ones backported downstream already and i was updating the other patches because i was pulling it downstream too
16:45:17 sean-k-mooney so im not actully that worreied that they will break thigns but just dont want to force in 5 patches right before eol
16:47:40 sean-k-mooney we dont patch things in our downstream without it landing in master first but we do backport thing that gerenally cant be backported upstream due to policy so just because it happend to work in our downstream buidl does not mean it wont break somethign upstream
16:53:05 elodilles sean-k-mooney: yeah, i guess upstream testing can be a bit different...
22:51:44 opendevreview melanie witt proposed openstack/nova stable/2023.1: Add debug logging when Instance raises OrphanedObjectError https://review.opendev.org/c/openstack/nova/+/888160
22:54:54 opendevreview melanie witt proposed openstack/nova stable/zed: Add debug logging when Instance raises OrphanedObjectError https://review.opendev.org/c/openstack/nova/+/888161
22:55:35 opendevreview melanie witt proposed openstack/nova stable/yoga: Add debug logging when Instance raises OrphanedObjectError https://review.opendev.org/c/openstack/nova/+/888162
22:56:09 opendevreview melanie witt proposed openstack/nova stable/xena: Add debug logging when Instance raises OrphanedObjectError https://review.opendev.org/c/openstack/nova/+/888163
22:56:44 opendevreview melanie witt proposed openstack/nova stable/wallaby: Add debug logging when Instance raises OrphanedObjectError https://review.opendev.org/c/openstack/nova/+/888164
#openstack-nova - 2023-07-12
02:38:50 opendevreview melanie witt proposed openstack/nova stable/victoria: Reproducer for bug 2003991 unshelving offloaded instance https://review.opendev.org/c/openstack/nova/+/888203
02:38:51 opendevreview melanie witt proposed openstack/nova stable/victoria: Enforce quota usage from placement when unshelving https://review.opendev.org/c/openstack/nova/+/888204
02:41:04 opendevreview melanie witt proposed openstack/nova stable/ussuri: Reproducer for bug 2003991 unshelving offloaded instance https://review.opendev.org/c/openstack/nova/+/888205
02:41:05 opendevreview melanie witt proposed openstack/nova stable/ussuri: Enforce quota usage from placement when unshelving https://review.opendev.org/c/openstack/nova/+/888206
09:01:49 kashyap gibi (or anyone): When you get a min, I feel _something_ is off with the code removal of MIN_LIBVIRT_AARCH64_CPU_COMPARE: https://review.opendev.org/c/openstack/nova/+/887255/2/nova/virt/libvirt/driver.py#b9978
09:01:55 kashyap I can't quite tell what, though
09:42:02 gibi kashyap: left feedback in the patch
09:42:02 opendevreview alecorps proposed openstack/nova master: Workaround for issues with ephemeral disk named disk.local during resize https://review.opendev.org/c/openstack/nova/+/888220
09:42:03 opendevreview alecorps proposed openstack/nova master: Workaround to fix the issue in resize with ephemeral disk named disk.local. Add a condition to ensure the backing file is present on target host. https://review.opendev.org/c/openstack/nova/+/888221
09:49:51 kashyap gibi: Thank you; looking
10:45:05 gibi kashyap: added one more comment to that review now
10:49:24 frickler sean-k-mooney: gibi: may I turn your attention to https://review.opendev.org/c/openstack/nova/+/871968 once more? there's a comment about a missing backport and I'm not sure whether the CI failure is maybe related or not
10:51:07 sean-k-mooney i belive that is blocked by my fixes for the gate blocker
10:51:19 gibi so one of the error:
10:51:20 gibi tempest.lib.exceptions.InvalidParam: Invalid Parameter passed: When validation.connect_method equals floating, validation_resources cannot be None
10:51:25 sean-k-mooney frickler: basically a test was merged in tempest that required validation to be enabled
10:51:26 gibi this is a known issue
10:51:44 sean-k-mooney i have not backported that fix for nova-lvm to zed yet
10:51:47 gibi ahh both job hit the same
10:52:13 sean-k-mooney only master and the 2023.1 patch is pending a second backport in the devstack ceph plugin
10:53:54 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/887674 is the current backport on the nova side
10:54:25 sean-k-mooney but we might need https://review.opendev.org/c/openstack/tempest/+/886991 https://review.opendev.org/c/openstack/devstack/+/887660 or https://review.opendev.org/c/openstack/devstack-plugin-ceph/+/887879 as well
10:55:42 opendevreview sean mooney proposed openstack/nova stable/2023.1: enable validations in nova-lvm https://review.opendev.org/c/openstack/nova/+/887674
10:56:24 sean-k-mooney i have also added the depends-on for the change elodilles suggested in devstack and we can see if that helps
10:56:59 sean-k-mooney if it fails again ill try also pulling in the second devstack-plugin-ceph patch but im not actully sure that is requried
11:00:26 frickler so what about backporting https://review.opendev.org/c/openstack/nova/+/869950 , should that still also happen? at least to zed?
11:00:46 sean-k-mooney it needs to go to yoga im currently creatign one for zed
11:01:14 sean-k-mooney i was waiting for the antelope patch to actully be working before creating it but i can push one up qickly
11:04:33 sean-k-mooney i need to cherry pick 3 patchs total across 3 repos i think so im goign to do the other two first and then the nova patch which depend on them
11:11:32 opendevreview sean mooney proposed openstack/nova stable/zed: enable validations in nova-lvm https://review.opendev.org/c/openstack/nova/+/888230
11:11:55 sean-k-mooney frickler: so i think ^ will work if the antelope one is now correct
11:49:25 kashyap gibi: Thanks! That return was the thing I was not quite sure of :)
11:49:34 kashyap But your reason makes sense
11:52:27 kashyap gibi: Just nuking that return should be fine there, I think.
11:58:18 frickler sean-k-mooney: that's not the change I was mentioning above, but ok let's wait how your patches fare in CI first
12:02:00 sean-k-mooney frickler: i know i did not want to update the change you asked about
12:02:28 sean-k-mooney frickler: its blocked by the know gate issue and the patch i linked should solve the know gate issue
12:02:53 sean-k-mooney right now all nova stable branches are blocked between antelope and yoga
12:03:12 sean-k-mooney older ones em branches have pinned tempest
12:04:14 gibi kashyap: I did not notice that return first either. Needed to run the unit test to catch it
12:04:49 kashyap gibi: Thanks for taking the time to do it!
13:04:58 opendevreview Amit Uniyal proposed openstack/nova master: Refactor CinderFixture https://review.opendev.org/c/openstack/nova/+/885756
13:17:47 astupnik Hello. Looking for +2/WF+1 for stable/wallaby backport https://review.opendev.org/c/openstack/nova/+/879405 (improves logging for nova-scheduler) Thanks in advance!
13:19:31 dansmith got it
13:21:00 astupnik thank you very much dansmith
13:21:03 opendevreview Amit Uniyal proposed openstack/nova master: Refactor CinderFixture https://review.opendev.org/c/openstack/nova/+/885756
13:21:04 opendevreview Amit Uniyal proposed openstack/nova master: Delete dangling bdms https://review.opendev.org/c/openstack/nova/+/882284
13:21:04 opendevreview Amit Uniyal proposed openstack/nova master: Reproducer for dangling bdms https://review.opendev.org/c/openstack/nova/+/881457
14:09:15 opendevreview Stephen Finucane proposed openstack/placement master: db: Workaround SQLAlchemy 2.x bug https://review.opendev.org/c/openstack/placement/+/888281
14:20:03 opendevreview Amit Uniyal proposed openstack/nova master: Refactor CinderFixture https://review.opendev.org/c/openstack/nova/+/885756
14:20:04 opendevreview Amit Uniyal proposed openstack/nova master: Delete dangling bdms https://review.opendev.org/c/openstack/nova/+/882284
14:20:04 opendevreview Amit Uniyal proposed openstack/nova master: Reproducer for dangling bdms https://review.opendev.org/c/openstack/nova/+/881457
15:00:08 auniyal will it be okay to use RuntimeError directly inside compute/manager ? or new Exception InvalidConnectorError(RuntimeError) should be in exception.py ?
15:00:08 auniyal I think if new Exception shouldn't be from NovaException then as you suggested RuntimeError makes more sense;
15:00:08 auniyal dansmith, regarding which exception to raise if provided connector is not correct; https://review.opendev.org/c/openstack/nova/+/877446/comment/16f7cd47_d2c8aef9/;
15:12:07 opendevreview Merged openstack/nova stable/wallaby: Add debug log for scheduler weight calculation https://review.opendev.org/c/openstack/nova/+/879405
15:22:01 dansmith auniyal: this is nova-manage right?
15:22:51 dansmith I think it's fine to have "class HostConflict(Exception): pass" in manage.py and then use that internally to manage itself
15:22:58 dansmith I would not do that in any service code probably
15:24:36 auniyal yes, nova-manage
15:24:37 auniyal ack
15:27:22 dansmith sean-k-mooney: can you rubber-stamp this cinder fixture refactor? https://review.opendev.org/c/openstack/nova/+/885756
15:27:44 dansmith converts stub_out to MockPatch so we can get that out from under auniyal's other stack
15:28:05 sean-k-mooney ill take a look now
15:28:17 dansmith don't look, just +W (j/k)
15:28:19 sean-k-mooney oh this hit the complexity limit right

Earlier   Later