| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2026-08-26 | |||
| 11:51:39 | stephenfin | gibi: done. dead simple | |
| 11:53:05 | sean-k-mooney | gibi: dont we alrady have a wait_for_migration_status in the integrate helpers | |
| 11:54:02 | stephenfin | sean-k-mooney: read the commit message :) | |
| 11:54:03 | sean-k-mooney | we copy that function in some of the regression tets and we reuse it in others | |
| 11:54:04 | gibi | sean-k-mooney: as I noted in the commit message this regression does not use our helper mixing | |
| 11:54:09 | gibi | s/g// | |
| 11:54:21 | gibi | hence I vendored the helper | |
| 11:54:39 | gibi | I didn't wanted the bigger change to pull in a new base class | |
| 11:55:09 | sean-k-mooney | ya i was just debating the value of that vs the copy | |
| 11:55:11 | sean-k-mooney | +2w | |
| 11:55:34 | gibi | thanks | |
| 11:58:09 | sean-k-mooney | im alway 50:50 on that, i perfer when our regression tests follow our docs and inherit form our base test case with minimal depency on those deper classes, on the other hand the integrate heleprs mixin is very useful and keeps them small | |
| 11:58:47 | sean-k-mooney | so i prefer reading them when it uses it :) and writeing them i guess but it creats more depencnece on the internals | |
| 11:59:39 | sean-k-mooney | stephenfin: so just to confirm with the cinder change i should now be able to use valid_interface internal,public for cidner right | |
| 11:59:49 | sean-k-mooney | we get that for free form ksa | |
| 12:00:38 | sean-k-mooney | i think in our installer we woudl still jut have it defautl to only internal | |
| 12:00:53 | sean-k-mooney | but its nice to have the option to have it be a list | |
| 12:03:27 | stephenfin | without the change you can use interface already. I'm not sure about valid_interface | |
| 12:03:34 | stephenfin | but with the change, you can use either | |
| 12:03:39 | sean-k-mooney | interface is not a list | |
| 12:03:53 | sean-k-mooney | so withoug the change you can have a list of endpoint to fall back on | |
| 12:04:13 | sean-k-mooney | valid_interfaces allwos you to do internal,public,admin | |
| 12:04:23 | sean-k-mooney | and it will try them in that order | |
| 12:04:39 | sean-k-mooney | well try to resovle them in that order | |
| 12:04:56 | sean-k-mooney | i.e. it check if there is an internal endpoint in the catalong and ifn not i twill check for public | |
| 12:05:04 | sean-k-mooney | it wont change the number of requests | |
| 12:05:27 | stephenfin | oh yeah I know there's a difference. What I don't know is whether cinderclient supported it | |
| 12:05:33 | sean-k-mooney | but it allwos you to write yoru config a bit more geneircally | |
| 12:06:35 | sean-k-mooney | i think that is resoveld before the cindercient but not sure, in anycasye the cidner change looks good to me | |
| 12:08:22 | stephenfin | looks like if you pass a list of strings to cinderclient's Client, it will pass it straight through to ksa's LegacyJsonAdapter | |
| 12:08:58 | sean-k-mooney | i tought it was ksa that handeedl this to be honest | |
| 12:09:14 | sean-k-mooney | i.e. that it was chooing the endpoint to use rather then the client | |
| 12:09:25 | stephenfin | but I guess you'd need a ListOpt to ensure the nova.conf value is transformed to a list instead of being given as a string | |
| 12:09:30 | sean-k-mooney | since its realy just the url for the request form the catalog that this is selecting | |
| 12:09:32 | stephenfin | yeah, that's been ksa for years | |
| 12:09:43 | stephenfin | and keystoneclient before that, iirc | |
| 12:10:13 | sean-k-mooney | ya so it should "just work" vai the common shared code path | |
| 12:10:27 | sean-k-mooney | cinder client is really just dealing with formating the payload | |
| 12:11:43 | sean-k-mooney | so my meta hope is to move placement and cybrog to using the sdk and proxy apis in 2027.1 | |
| 12:12:02 | sean-k-mooney | and then all that is left i think is neutron and cidner to have nova fully using the sdk | |
| 12:12:22 | sean-k-mooney | bar barbican with goes via castalan | |
| 12:13:01 | stephenfin | and glance? | |
| 12:13:13 | sean-k-mooney | oh i tought glance was already done | |
| 12:13:19 | sean-k-mooney | maybe not | |
| 12:13:42 | stephenfin | I think I tried glance before but it's weird because it can talk to multiple glance deployments or something like that (?) | |
| 12:13:44 | sean-k-mooney | oh your right | |
| 12:14:09 | sean-k-mooney | so glancce is the only thing that cybrog talks to not via the sdk today | |
| 12:14:33 | sean-k-mooney | so maybe i can find some time to do that for both nova and cybrog | |
| 12:15:09 | sean-k-mooney | i know the glance client has some extra logic aroudn how ti does image streaming that a bit more complx then a standard rest call | |
| 12:15:17 | sean-k-mooney | all the multipatrt mime stuff | |
| 12:15:46 | sean-k-mooney | so ya the fact it supprot multi store | |
| 12:15:52 | sean-k-mooney | and that make galnce a bit tricker | |
| 12:16:19 | sean-k-mooney | im 99% sure if you try to use any of the fancier parts of the glance api cyborg will just break | |
| 12:18:32 | sean-k-mooney | there has never been any testign of multi store there and beyond that while the client was litrally copy pasted form nova it didnt include all its depencies form nova so i suspect adopting the sdk there and actully testing it will be a large improvment | |
| 12:51:03 | opendevreview | Merged openstack/nova master: Improve the graceful shutdown job ansible tasks name https://review.opendev.org/c/openstack/nova/+/1001212 | |
| 12:51:16 | opendevreview | Merged openstack/nova master: libvirt: Generalize memory encryption config https://review.opendev.org/c/openstack/nova/+/998466 | |
| 12:56:14 | opendevreview | Merged openstack/nova master: libvirt: Add memory encryption config for TDX https://review.opendev.org/c/openstack/nova/+/998608 | |
| 12:59:25 | opendevreview | Merged openstack/nova master: libvirt: Add os_firmware_secure https://review.opendev.org/c/openstack/nova/+/1000472 | |
| 13:02:48 | sean-k-mooney | gibi: i kind of hate how effective adding random `import time; time.sleep(2)` into the code seams to be at finding these intermitent races | |
| 13:03:34 | sean-k-mooney | it reminds me of time wehre adding a print statement to debug code stop the bug form happening :) | |
| 13:13:24 | sean-k-mooney | am i being stupid "nova.exception.ServiceTooOld: This service is older (v72) than the minimum (v73) version of the rest of the deployment. Unable to continue." | |
| 13:13:36 | sean-k-mooney | so i deployed master yesterday | |
| 13:14:03 | sean-k-mooney | and i just checked out mel's cybrog change now | |
| 13:14:08 | sean-k-mooney | oh | |
| 13:14:14 | sean-k-mooney | i ownder if it need a rebase | |
| 13:14:26 | sean-k-mooney | we might have merged a new version in between? | |
| 13:17:09 | sean-k-mooney | ya ok it works after a rebase | |
| 13:18:34 | gibi | sean-k-mooney: with native threading, task can be switched any time. with eventlet they switch at certain places only (sleep and io) | |
| 13:18:47 | gibi | so our test are relied on those eventlet run to completion guarantees | |
| 13:18:58 | gibi | but it cannot rely on it any more | |
| 13:19:18 | gibi | I'm pretty happy that these are only test instabilities and we haven't found any new code races yet | |
| 13:20:09 | sean-k-mooney | ya, in the real code we were better about treating shared memory access as if it was real thread and using locks | |
| 13:20:27 | sean-k-mooney | i for one alwasy preetened it was a real thread for the most part hwen reviewing | |
| 13:20:52 | sean-k-mooney | but i woudl be surppriesed if we never mised a case where we relised on the eventlet behavior | |
| 13:21:05 | gibi | :) | |
| 13:39:03 | opendevreview | Balazs Gibizer proposed openstack/nova master: Stabilize live migration tests https://review.opendev.org/c/openstack/nova/+/1002493 | |
| 13:39:09 | gibi | the follow up :) ^^ | |
| 14:08:36 | melwitt | sean-k-mooney, stephenfin: from what I understand, the ServiceTokenAuthWrapper is required in order to send user+service token https://github.com/openstack/keystoneauth/blob/9e083fd5ae63aef66349ca980cdd315f5df4c767/keystoneauth1/service_token.py#L26 so that has to be passed in somehow | |
| 14:09:03 | opendevreview | Daniel Schulz proposed openstack/nova master: Prevent orphaned attachments by preemptive delete https://review.opendev.org/c/openstack/nova/+/1002439 | |
| 14:10:12 | melwitt | there's already a helper in nova to create one https://github.com/openstack/nova/blob/bbe520dd911be598273dd3d6d5a09fd35a56ddca/nova/service_auth.py#L69 so you get an auth plugin | |
| 14:11:41 | melwitt | and have to pass that to sdk somehow | |
| 14:17:22 | melwitt | ok I am seeing the rest of the review comments now | |
| 14:40:56 | gibi | gmaan: filed one more instabilities https://bugs.launchpad.net/nova/+bug/2165149 and assigned it to myself | |
| 14:46:36 | gibi | aaand file one more https://bugs.launchpad.net/nova/+bug/2165150 | |
| 14:50:26 | gibi | Uggla: I confirmed that your shelve offload / unshelve race is the same as we see on the gate so I reviewed your repor and fix and left comments | |
| 14:51:00 | Uggla | gibi cool ! I will try to jump on it ASAP | |
| 14:51:03 | sean-k-mooney | melwitt: so i think we can proceed with your change as is | |
| 14:51:34 | sean-k-mooney | melwitt: i think we can tidy things up in a followup and make manilla interaction use the service user tokes ectra based on this infra | |
| 14:51:35 | Uggla | Right now I'm testing cold migration of shares and found couple of issues. | |
| 14:51:36 | gibi | and marked https://bugs.launchpad.net/nova/+bug/2163552 as duplicate of https://bugs.launchpad.net/nova/+bug/2120927 (cc gmaan ) | |
| 14:52:01 | melwitt | sean-k-mooney: ok that is good news! | |
| 14:52:50 | sean-k-mooney | i held +w in case you wanted to look at refactoring the ksa plugin creation in this patch but i think its better to defer that | |
| 14:53:07 | sean-k-mooney | so i can add +w now if you dont intend to make any other changes | |
| 14:53:28 | sean-k-mooney | melwitt: we are expcting to update this code again after the sdk release too right | |
| 14:53:44 | sean-k-mooney | so we can defer those topic to that patch or a followup | |
| 14:54:23 | melwitt | sean-k-mooney: you mean the keystoneauth release? yeah although I think I see that got released this morning | |
| 14:55:24 | melwitt | it's not in upper-constraints yet though so I think we can't use yet | |
| 14:57:20 | melwitt | sean-k-mooney: yeah I mean AFAICT we do need 3 options in get_sdk_adapter (I replied on the review) so I'm not sure there is a cleaner way to do it | |
| 14:59:27 | melwitt | for manila, do we know we want to send service tokens along with it too? I wondered if it is intentional that it does not today | |
| 14:59:37 | sean-k-mooney | yes | |