Friday, 2026-06-12

opendevreviewsean mooney proposed openstack/devstack master: add support for ubuntu 26.04  https://review.opendev.org/c/openstack/devstack/+/98848209:10
opendevreviewsean mooney proposed openstack/devstack master: Remove resolute noVNC eventlet workaround  https://review.opendev.org/c/openstack/devstack/+/99309109:10
opendevreviewBalazs Gibizer proposed openstack/whitebox-tempest-plugin master: Add coverage for iothread support  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99231309:27
opendevreviewsean mooney proposed openstack/devstack master: Add Ubuntu 26.04 platform support  https://review.opendev.org/c/openstack/devstack/+/98848209:44
opendevreviewsean mooney proposed openstack/devstack master: Remove resolute noVNC eventlet workaround  https://review.opendev.org/c/openstack/devstack/+/99309109:44
*** ykarel_ is now known as ykarel11:37
*** jgilaber_ is now known as jgilaber12:53
opendevreviewDouglas Mendizábal proposed openstack/devstack master: Enforce new keystone policies by default  https://review.opendev.org/c/openstack/devstack/+/95621015:30
opendevreviewDouglas Mendizábal proposed openstack/devstack master: Enforce new keystone policies by default  https://review.opendev.org/c/openstack/devstack/+/95621015:31
opendevreviewDouglas Mendizábal proposed openstack/devstack master: Enforce new keystone policies by default  https://review.opendev.org/c/openstack/devstack/+/95621015:32
opendevreviewsean mooney proposed openstack/devstack master: Add Ubuntu 26.04 platform support  https://review.opendev.org/c/openstack/devstack/+/98848216:56
opendevreviewsean mooney proposed openstack/devstack master: Re-enable resolute noVNC proxy  https://review.opendev.org/c/openstack/devstack/+/99309116:56
sean-k-mooneythe scodn patch passed the last time i think they both should pass this time so we will have funcitonal resolute jobs once all the depencies have merged16:57
sean-k-mooneygmaan:^16:58
sean-k-mooneyso maybe we will be able to call that done in a week or two16:58
sean-k-mooneyand start usign it in services16:59
gmaancool17:01
sean-k-mooneyim wondering if we/i should create a tempest-full-py314 job based off that nodeset that project can opt into17:10
sean-k-mooneyor if we should add it to a templeate or something as non voting17:11
sean-k-mooneywhat the normal pattren for this?17:11
gmaanplatform job is enough that still run tempest testts17:12
sean-k-mooneyit does yes so for project that want to opt in they should either add that or inherit form it?17:13
gmaanwhenever we switch from noble to resolute, we do move all tempest jobs (automatically via base job change or some of them manually )17:13
gmaanthey can just use the nodeset and prepare their resolute  jobs, does not need to inherit from platform job17:14
sean-k-mooneythey can but until the novnc issues is fixed they woudl have to duplciat eth workaround17:15
sean-k-mooneybut ok ill leave it like that17:15
sean-k-mooneyand just add a job to cybrog and or nova etcra where i want to do the early testing17:15
gmaanyeah17:17
*** vhari_ is now known as vhari17:21
opendevreviewDan Smith proposed openstack/tempest master: Test nova unpin-AZ behavior  https://review.opendev.org/c/openstack/tempest/+/99225717:29
gmaandansmith: thanks for update, do you think adding a scenario tests verifying that server (created with pin AZ is) can be moved to other host after unpin ? that will make sure end goal of unpin az https://review.opendev.org/c/openstack/tempest/+/992257/comments/4f21cb3c_77891d6217:32
dansmithgmaan: the pinning is not new here, but do you mean verify that we can create a pinned server then unpin and move it?17:40
gmaandansmith: yeah17:40
gmaanafter unpin server is movable17:40
dansmithgmaan: that requires us to have two nodes in two AZs, both of which are valid for our instance17:41
dansmithI thought that was probably too specific for regular tempest17:41
gmaanwe have otehr live migration tests requiring two node and there we can add it https://github.com/openstack/tempest/blob/master/tempest/api/compute/admin/test_live_migration.py17:41
gmaanthose run in live migration job on nova side and multinode job 17:42
gmaanoh two AZ17:42
dansmithokay I can try, but won't be able to test locally.. can that be a separate patch?17:42
gmaanmaybe creating in one AZ and then just unpin 17:42
gmaansure, ok with separate test17:42
gmaani mean separate change17:43
dansmithack17:43
dansmithgmaan: we have to have two AZs in order to test.. one AZ with two nodes still allows migration between them17:45
gmaanyeah, you are right, single AZ env would not be able to test it. let me check if multinode job is setup with two different AZ17:52
gmaandansmith: i cannot find any job having two diff AZ.  checked in cinder side also if they are testing cross az attachment but could not find18:06
dansmithokay18:07
gmaaneither we need to setup one or we can leave the tempest test and cover it in nova functional tests?18:07
opendevreviewGhanshyam Maan proposed openstack/whitebox-tempest-plugin master: Add supported stable branch testing on master gate  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99299818:13
melwittoh, are host aggs and AZ controlled by node provider only or such?18:17
melwittlike we can't use nova API to make host aggregates and AZ ourselves or?18:17
dansmithmelwitt: no, we can of course :)18:18
dansmithwe just don't have such a job I think is the point,18:19
dansmithand it would not be useful for any test bucket that has migration tests18:19
dansmithbecause they would all want to move and couldn't18:19
gmaanyeah, if we set up one it will be more useful for the new tests of moving server after unpin or maybe to test the cinder cross az attachment etc18:21
melwitthm... would it not be scoped per test tenant though? I guess I was thinking the test in question would set it up for itself only and not affect other things. or are AZ scoped per region at the finest grain level?18:21
dansmithAZ is per deployment18:29
dansmithi.e. for all of a nova (region)18:29
melwittahh k18:30
dansmithso it'd have to be like other live migration tests where we run tempest multiple times with reconfigures/restarts in the middle18:31
dansmithwhich is a bit much for a single test18:31
melwittyeah, bummer18:31
dansmithI'm also really not very worried about it because it's just setting a database field to null, and then at that point, the rest of the migration tests are theoretically proving that migration of unpinned instances works :)18:32
melwittyeah, I'm not either but I wondered about the AZ concepts in general. hadn't really thought about it before in this context18:33
opendevreviewGhanshyam Maan proposed openstack/whitebox-tempest-plugin master: Prepare the stable branches in whitebox-tempest-plugin  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99299819:31
opendevreviewGhanshyam Maan proposed openstack/whitebox-tempest-plugin master: Prepare the stable branches in whitebox-tempest-plugin  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99299819:33
opendevreviewGhanshyam Maan proposed openstack/whitebox-tempest-plugin master: Prepare the stable branches in whitebox-tempest-plugin  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99299819:34
opendevreviewGhanshyam Maan proposed openstack/whitebox-tempest-plugin master: Prepare the stable branches in whitebox-tempest-plugin  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99299819:34
opendevreviewMerged openstack/whitebox-tempest-plugin master: Add coverage for iothread support  https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/99231319:53

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!