| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2023-09-04 | |||
| 07:58:32 | frickler | really necessary? | |
| 08:00:02 | bauzas | frickler: you're doing a point, we did hit the ground hard last cycle due to a very last dependency bump | |
| 08:01:45 | bauzas | https://governance.openstack.org/tc/reference/runtimes/2023.2.html | |
| 08:02:47 | frickler | I know that doc, but it doesn't say that projects must be incompatible to everything else | |
| 08:02:59 | bauzas | so, yeah, basically, those projects were lagging, but maybe the bump is quite unappropriate at this period in time of a release | |
| 08:03:27 | frickler | yes, doing the bump at the beginning of the cycle, I would have had no objections at all | |
| 08:03:31 | bauzas | frickler: I thought we had a long conversation last cycle with the TC about PTIs | |
| 08:06:00 | bauzas | this time, we were clear on the fact that Focal was out of support for a long time | |
| 08:06:20 | bauzas | but yeah, we lagged merging this patch | |
| 08:06:40 | bauzas | at least, projects have 1.5 weeks to remove their focal jobs before RC1... | |
| 08:06:47 | bauzas | (half-kidding) | |
| 09:09:51 | sahid | o/ | |
| 09:10:31 | sahid | quick question guys, is there a way to restrict the usage of a AZ for certain projects or users? | |
| 09:46:07 | admin1 | sahid, you can create an aggregate and filer it by tenant_id and add AggregateMultiTenancyIsolation for nova | |
| 09:51:43 | sean-k-mooney | frickler: it was ment to done last cycle and was intened to be done at milestone one this cycle | |
| 09:52:52 | sean-k-mooney | frickler: so while not ideal im pretty storongly agasint not bumping it again. it has been weel over due and we are just bumping it to what we actully use in ci. | |
| 09:54:42 | sean-k-mooney | i dont think it would have been reasonable to wait to calacal for a bump that was ment to be in antilope when it got punted form antelope because it was beign done late in the cycle. | |
| 09:56:35 | sahid | admin1: interesting point thank you | |
| 09:56:59 | sean-k-mooney | sahid: there is a way to do that wiht placement too instead of a filter | |
| 09:57:01 | sahid | if a situation where I have several tenant | |
| 09:57:14 | sahid | sean-k-mooney: how do you see that? | |
| 09:57:23 | sean-k-mooney | https://docs.openstack.org/nova/latest/admin/aggregates.html#tenant-isolation-with-placement | |
| 09:57:50 | sean-k-mooney | it works more or less thet same but placement enforces it instead of a post filter | |
| 09:58:25 | sean-k-mooney | which is imporant if you return a small set of canidates form placement like cern does (they limit to <20 allocation candiates) | |
| 10:01:04 | sahid | perfect, thanks a lot for your help ! | |
| 10:07:44 | sahid | sean-k-mooney: if I set a AZ for the aggregate that should avoid to manually set all the hosts related, right? | |
| 10:08:30 | admin1 | i think that is the idea from the examples | |
| 10:09:10 | sean-k-mooney | sahid: so often we tell people to not set any addtional metadta on the host aggreate that holds the AZ metadata | |
| 10:09:26 | sean-k-mooney | but you can and in this case that is proably what you want to do | |
| 10:09:43 | sean-k-mooney | if you want to resrict an az to a tenant and not have them ever get out of sync | |
| 10:10:21 | sahid | i'm glad it's exactly what I was needed :-) | |
| 10:10:26 | sahid | thanks you admin1, sean-k-mooney | |
| 12:49:06 | elodilles | bauzas auniyal : hi, i've updated the hash in the yoga release patch, please review it and then we can merge it: https://review.opendev.org/c/openstack/releases/+/889517 | |
| 12:49:17 | bauzas | cool cool | |
| 12:53:28 | elodilles | thanks o/ | |
| 13:25:43 | opendevreview | Merged openstack/nova master: [alembic] Alembic operations require keywords only arguments https://review.opendev.org/c/openstack/nova/+/883344 | |
| 13:49:05 | auniyal | thanks elodilles | |
| 13:52:14 | elodilles | np :) | |
| 14:24:31 | sean-k-mooney | hi folks what should i be setting oc_header too in vars.yaml? | |
| 14:26:19 | sean-k-mooney | oh wrong window... | |
| #openstack-nova - 2023-09-05 | |||
| 03:10:04 | opendevreview | OpenStack Proposal Bot proposed openstack/nova master: Imported Translations from Zanata https://review.opendev.org/c/openstack/nova/+/891648 | |
| 08:38:09 | bauzas | sean-k-mooney: can you review https://review.opendev.org/c/openstack/nova/+/893462 ? | |
| 08:49:55 | sean-k-mooney | i need to leave fo a doctors approintment soon but ill take a look at it today | |
| 08:57:11 | opendevreview | Merged openstack/nova master: Imported Translations from Zanata https://review.opendev.org/c/openstack/nova/+/891648 | |
| 09:58:06 | opendevreview | Danylo Vodopianov proposed openstack/nova master: Packed virtqueue support was added. https://review.opendev.org/c/openstack/nova/+/876075 | |
| 10:10:32 | plibeau | hello guys, I need your help on this change https://review.opendev.org/c/openstack/nova/+/861172 | |
| 12:21:41 | dvo-plv | sean-k-mooney, dansmith Uggla sahid Hello, Mayby you will have a chance to review patchset https://review.opendev.org/c/openstack/nova/+/876075 | |
| 12:27:26 | sean-k-mooney | dvo-plv: unfortunetlly we are past feature freeze fo this cycle | |
| 12:28:26 | sean-k-mooney | i could take a look but it cant merge until next cycle after RC1 is cut so thats 2-3 weeks form now | |
| 12:29:18 | sean-k-mooney | bauzas: unless we want to consider a FFE for ^ | |
| 12:29:28 | sean-k-mooney | i was +2 on it a month ago | |
| 12:29:42 | sean-k-mooney | it would have been nice to land that but we are now out of time | |
| 12:30:25 | bauzas | I'm not super happy to create a FFE for this due to https://review.opendev.org/c/openstack/nova/+/876075/23/nova/objects/image_meta.py | |
| 12:30:49 | dvo-plv | I see,maybe it will be good to take a look, if you have new comments, we will have a time to fix until next cycle | |
| 12:31:55 | bauzas | I'd prefer to look at it in two weeks, after RC1 while the 2023.2 branch is created and then master is for 2024.1 | |
| 12:39:19 | opendevreview | Danylo Vodopianov proposed openstack/nova master: Packed virtqueue support was added. https://review.opendev.org/c/openstack/nova/+/876075 | |
| 13:05:39 | opendevreview | Danylo Vodopianov proposed openstack/nova master: Packed virtqueue support was added. https://review.opendev.org/c/openstack/nova/+/876075 | |
| 14:20:22 | opendevreview | Sylvain Bauza proposed openstack/nova master: doc: mark the maximum microversion for 2023.2 Bobcat https://review.opendev.org/c/openstack/nova/+/893742 | |
| 14:26:15 | opendevreview | Sylvain Bauza proposed openstack/nova master: Update compute rpc alias for bobcat https://review.opendev.org/c/openstack/nova/+/893744 | |
| 14:26:25 | ygk_12345 | some of my nodes in an aggregate dont have the aggregate mapping in the placement db for those nodes | |
| 14:26:34 | ygk_12345 | is this a bug in nova ? | |
| 14:27:18 | ygk_12345 | i am trying to use the command "openstack resource provider aggregate list <resource-id> | |
| 14:27:42 | ygk_12345 | that is returning empty result for those nodes in that aggregate | |
| 14:28:16 | ygk_12345 | i tried placement syc aggregates also | |
| 14:28:20 | ygk_12345 | but no result | |
| 14:28:37 | ygk_12345 | *sync | |
| 14:36:10 | opendevreview | Merged openstack/nova master: Follow up for unified limits: PCPU and documentation https://review.opendev.org/c/openstack/nova/+/893462 | |
| 14:41:14 | opendevreview | Sylvain Bauza proposed openstack/nova master: Add service version for Bobcat https://review.opendev.org/c/openstack/nova/+/893749 | |
| 14:49:43 | ygk_12345 | any idea anyone ? | |
| 14:50:37 | dansmith | ygk_12345: not really enough information to say (like which release are you running?) but in general the placement aggregates should match the nova ones, yes | |
| 14:51:04 | ygk_12345 | dansmith: it is yoga | |
| 14:52:24 | ygk_12345 | the aggregate matching ismissing for one set of nodes in the az | |
| 14:53:26 | dansmith | ygk_12345: and that set of nodes is different how? | |
| 14:54:01 | ygk_12345 | dansmith: they are just added to the az along with others | |
| 14:56:02 | ygk_12345 | do I need to manually add all these nodes from that aggregate to match that aggregate using the command openstack resource provider aggregate set ? | |
| 14:56:40 | ygk_12345 | it succeeded for one node now | |
| 14:56:44 | dansmith | ygk_12345: when you add a node to an aggregate in nova, it should mirror that to placement for you | |
| 14:56:56 | ygk_12345 | but why did not it add those nodes automatically ? | |
| 14:57:41 | ygk_12345 | dansmith: but it did not do so for an aggregate | |
| 14:58:46 | dansmith | ygk_12345: not sure what to say without logs and things... try removing it from the aggregate in nova, re-add it and look at the nova-api logs to see if there's a failure talking to placement or something | |
| 14:59:05 | ygk_12345 | dansmith: ok | |
| 15:00:19 | bauzas | reminder: nova meeting in 1 hour | |
| 15:00:53 | bauzas | ygk_12345: we have some method that calls placement for creating the same aggregates, sec | |
| 15:01:35 | ygk_12345 | bauzas: ok | |
| 15:01:45 | ygk_12345 | bauzas: please let me know | |
| 15:04:33 | bauzas | eventually found the change https://review.opendev.org/c/openstack/nova/+/553597 | |
| 15:07:31 | ygk_12345 | bauzas: I have removed and re added and now it is working. Thank you | |
| 15:07:44 | bauzas | ok good to know | |
| 15:09:28 | dansmith | ygk_12345: that likely means nova-api wasn't able to talk to placement at the time you added those nodes originally, for whatever reason | |
| 15:09:57 | ygk_12345 | dansmith: seems so. Thanks for your time and patience | |
| 16:00:02 | bauzas | #startmeeting nova | |
| 16:00:02 | opendevmeet | Meeting started Tue Sep 5 16:00:02 2023 UTC and is due to finish in 60 minutes. The chair is bauzas. Information about MeetBot at http://wiki.debian.org/MeetBot. | |
| 16:00:02 | opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | |
| 16:00:02 | opendevmeet | The meeting name has been set to 'nova' | |
| 16:00:07 | bauzas | welcome folks | |
| 16:00:08 | gibi | o/ | |
| 16:00:11 | sean-k-mooney | o/ | |
| 16:00:11 | Uggla | o/ | |
| 16:00:14 | elodilles | o/ | |
| 16:00:15 | auniyal | o/ | |