| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-07-23 | |||
| 14:55:20 | sean-k-mooney | i might submit a 2 liner patch to adress that | |
| 14:55:56 | gibi | :) | |
| 14:55:58 | sean-k-mooney | the code is form 6 years ago so its possibel that the function signiture has changeed and it was possibel before or something like that and the message just never got fixed | |
| 14:57:04 | sean-k-mooney | re reading it i can see the interpertation you are suggesting just looking at it in the nova logs it looks odd | |
| 14:57:15 | gibi | I looked at the blame for that reason but I don't see signature changes | |
| 15:04:55 | stephenfin | bauzas: RE: your comment here https://review.opendev.org/c/openstack/nova/+/754116/1/nova/tests/functional/test_scheduler.py#150, is that a thing? | |
| 15:05:02 | stephenfin | the code says no | |
| 15:05:19 | stephenfin | I don't see us stripping the prefix from the metadata properties | |
| 15:05:25 | opendevreview | Samuel proposed openstack/nova-specs master: Migrate Instance Between Projects https://review.opendev.org/c/openstack/nova-specs/+/802034 | |
| 15:05:28 | bauzas | did I miss to provide the URL ? my bad | |
| 15:05:58 | bauzas | https://github.com/openstack/nova/blob/3508263f236ea6003a76661b9e657ded4d46b413/nova/scheduler/filters/aggregate_instance_extra_specs.py#L55-L58 | |
| 15:06:32 | bauzas | stephenfin: this both doesn't verify keys with not the prefix, but also strips the key if the prefix is right | |
| 15:06:47 | stephenfin | it does that for the flavor though | |
| 15:06:57 | stephenfin | not the aggregate | |
| 15:07:06 | bauzas | oh, whoops, you're absolutely right | |
| 15:07:31 | stephenfin | I don't want to test the unprefixed variant since that behavior is deprecated and would be rejected with recent microversions | |
| 15:07:32 | bauzas | it's a flavor key, not a metadata extra spec | |
| 15:07:35 | stephenfin | yeah | |
| 15:07:55 | bauzas | stephenfin: ah | |
| 15:08:16 | bauzas | did we deprecated non-prefixed keys ? | |
| 15:08:30 | sean-k-mooney | bauzas: we did i think | |
| 15:08:39 | sean-k-mooney | bauzas: ut the filter still check them | |
| 15:08:40 | opendevreview | Samuel proposed openstack/nova-specs master: Migrate Instance Between Projects https://review.opendev.org/c/openstack/nova-specs/+/802034 | |
| 15:08:46 | stephenfin | sort of | |
| 15:09:02 | stephenfin | the docs tell you not to use them, but there's no in-code warning | |
| 15:09:05 | bauzas | my brain fucked somehow | |
| 15:09:07 | sean-k-mooney | we deprecated it because you cant use the aggret extra specs filter and compute capablities filter at the same time if you use unprefixed keys | |
| 15:09:20 | bauzas | sean-k-mooney: sure, I remember why we have this prefix | |
| 15:09:36 | bauzas | but I thought we only *documented* to use it if you were using both filters | |
| 15:09:45 | bauzas | not if you were only using one | |
| 15:09:56 | stephenfin | Multiple values can be given, as a comma-separated list. For backward compatibility, also works with non-scoped specifications; this action is highly discouraged because it conflicts with ComputeCapabilitiesFilter filter when you enable both filters. | |
| 15:10:04 | stephenfin | from way back in Mitaka https://docs.openstack.org/mitaka/config-reference/compute/scheduler.html | |
| 15:10:14 | stephenfin | "highly discouraged" is pretty strong :) | |
| 15:10:26 | bauzas | okay, so this is docs | |
| 15:10:35 | stephenfin | I can add the tests if you really want them though? | |
| 15:10:42 | sean-k-mooney | yes but with the intorctution of the flavor vlaidation it should be even more discuraged | |
| 15:10:42 | bauzas | the problem is that I know more ops that use this filter than the computecaps one | |
| 15:11:04 | bauzas | as the computecapabilities one is pretty useless now we have traits | |
| 15:11:20 | sean-k-mooney | bauzas: it can do thing tratis cant | |
| 15:11:36 | bauzas | stephenfin: as you wish, I won't both | |
| 15:11:38 | bauzas | bother | |
| 15:11:47 | bauzas | stephenfin: you added code, you proved me I was wrong | |
| 15:11:54 | bauzas | so that's a good start | |
| 15:12:02 | stephenfin | well I've done it now | |
| 15:12:07 | stephenfin | so you're getting the tests | |
| 15:12:11 | stephenfin | and you'll be happy with them | |
| 15:12:14 | sean-k-mooney | for what its worth i suggest we require custom: as a prefix when we were adding the flavor extra spec validation | |
| 15:12:38 | opendevreview | Stephen Finucane proposed openstack/nova master: Add tests for 'AggregateImagePropertiesIsolation' https://review.opendev.org/c/openstack/nova/+/754115 | |
| 15:12:38 | opendevreview | Stephen Finucane proposed openstack/nova master: Add tests for 'AggregateInstanceExtraSpecsFilter' https://review.opendev.org/c/openstack/nova/+/754116 | |
| 15:12:46 | stephenfin | gibi, bauzas: comments addressed ^ | |
| 15:13:10 | bauzas | stephenfin: looking again, then | |
| 15:14:20 | stephenfin | I'm pretty sure I had a fairly thorough reworking of the scheduler filter docs at some point. I wonder if that merged? | |
| 15:14:58 | stephenfin | nope https://review.opendev.org/c/openstack/nova/+/773645 | |
| 15:16:22 | gibi | I'm +2 on both test patch. thanks stephenfin | |
| 15:16:29 | stephenfin | thank you | |
| 15:17:16 | bauzas | sent to the gate, then | |
| 15:17:29 | bauzas | .... and, given this, folks, \o | |
| 15:17:47 | gibi | bauzas: o/, please remember I'm off next week, and the meeting chair is your :D | |
| 15:19:14 | sean-k-mooney | stephenfin: there are a few patches before that can it be pulled forward or should we try and merge the others first | |
| 15:19:32 | sean-k-mooney | the docs patch that is https://review.opendev.org/c/openstack/nova/+/773645 | |
| 15:19:46 | stephenfin | I'd say the latter. The docs changes assume the filter scheduler is the only one left, which requires the preceding changes | |
| 15:19:59 | sean-k-mooney | ack ok ill start form the bottom so | |
| 15:20:48 | bauzas | gibi: oh, I remember this, but I haven't said "good vacations" to you :p | |
| 15:21:14 | bauzas | gibi: fwiw, I'll be on PTO between Aug 9 and Aug 30 | |
| 15:21:27 | bauzas | ... just before end of Xena-3 :) | |
| 15:22:50 | gibi | bauzas: thanks | |
| 16:06:19 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 16:06:19 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 16:07:36 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 16:10:11 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 16:41:46 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 16:41:47 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 16:43:02 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 16:45:33 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 16:46:50 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 16:48:05 | opendevreview | Balazs Gibizer proposed openstack/nova master: WIP support extended res req in heal port allocation https://review.opendev.org/c/openstack/nova/+/802060 | |
| 17:27:46 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 17:28:53 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 17:30:10 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 17:30:11 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 17:32:42 | opendevreview | Balazs Gibizer proposed openstack/nova master: WIP support extended res req in heal port allocation https://review.opendev.org/c/openstack/nova/+/802060 | |
| 17:45:19 | gmann | gibi: ah I thought placement is in cycle-with-intermediary model but it is cycle-with-rc so need to wait for rc release. | |
| 17:47:13 | opendevreview | Samuel proposed openstack/nova-specs master: Migrate Instance Between Projects https://review.opendev.org/c/openstack/nova-specs/+/802034 | |
| 17:48:32 | opendevreview | Samuel proposed openstack/nova-specs master: Migrate Instance Between Projects https://review.opendev.org/c/openstack/nova-specs/+/802034 | |
| 17:50:28 | opendevreview | Samuel proposed openstack/nova-specs master: Migrate Instance Between Projects https://review.opendev.org/c/openstack/nova-specs/+/802034 | |
| 17:57:11 | gmann | gibi: stephenfin in that case we will release placement in rc1 right? which is after oslo feature freeze. with that we have to wait for one more cycle to get https://review.opendev.org/c/openstack/oslo.policy/+/801929 in oslo.policy ? | |
| 17:57:32 | gmann | I think that is how we handle such situation ? | |
| 18:00:05 | sean-k-mooney | to do a release of placement yes we wait until m3 and then can release a release candiate for placmeent | |
| 18:01:19 | sean-k-mooney | even if placment was cycle with intermediary it woudl not affect the oslo patch really | |
| 18:01:42 | sean-k-mooney | gmann: why do you think it would? | |
| 18:02:30 | gmann | sean-k-mooney: we have to release placement first in pypi for fix done for proposed interface changes in oslo | |
| 18:02:53 | gmann | otherwise placement in pypi fail with the latest oslo release which include the interface change | |
| 18:03:07 | sean-k-mooney | wotn we have the same issue with nova | |
| 18:03:11 | sean-k-mooney | or keystone | |
| 18:03:42 | sean-k-mooney | or any other project that are part of the integrated release | |
| 18:03:51 | gmann | technically yes but placement is used from pypi in nova tests so it block gate itself | |
| 18:03:54 | sean-k-mooney | we wont be able to increase upper constirting on master | |
| 18:04:21 | sean-k-mooney | you mena placment lib is used | |
| 18:04:30 | sean-k-mooney | i dont think we use placement directly | |
| 18:04:41 | gmann | if any using repo we use form pypi in gate testing then it would let oslo change merge itself, like placement | |