Thursday, 2025-09-11

*** mhen_ is now known as mhen01:57
*** ykarel_ is now known as ykarel07:00
opendevreviewMerged openstack/nova master: Add Flamingo prelude section  https://review.opendev.org/c/openstack/nova/+/95918809:03
opendevreviewStephen Finucane proposed openstack/nova master: DNM: Test patch removing glance standalone  https://review.opendev.org/c/openstack/nova/+/96044609:32
opendevreviewOpenStack Release Bot proposed openstack/placement stable/2025.2: Update .gitreview for stable/2025.2  https://review.opendev.org/c/openstack/placement/+/96050712:32
opendevreviewOpenStack Release Bot proposed openstack/placement stable/2025.2: Update TOX_CONSTRAINTS_FILE for stable/2025.2  https://review.opendev.org/c/openstack/placement/+/96050812:32
opendevreviewOpenStack Release Bot proposed openstack/placement master: Update master for stable/2025.2  https://review.opendev.org/c/openstack/placement/+/96050912:32
opendevreviewThomas Goirand proposed openstack/nova-specs master: Restrict booting from volume with disk  https://review.opendev.org/c/openstack/nova-specs/+/96051312:34
zigosean-k-mooney: It's the first time I'm writing a spec, I hope it's ok...12:34
jlejeunesean-k-mooney: gibi: Hello, I recently resumed my work on bug #2044235. You asked me to create a regression test to confirm the bug, but I'm having trouble reproducing it. When I raise the messagingTimeout exception from the ‘check_can_live_migrate_source’ method, it seems like this exception is caught by the fake transport driver, causing my test to fail. When I try with any other exception, 13:09
jlejeunemy test works fine. Have you ever seen this behavior before?13:09
*** iurygregory_ is now known as iurygregory13:10
sean-k-mooneyzigo: you removed section instead of jsut say None so that is why pep8 is failing13:11
zigoOk, will re-add it.13:12
sean-k-mooneyzigo: ill take a look later but thansk for creating one. we can discuss in the spec what the default for the policy shoudl be13:12
sean-k-mooneymy guess is member but we may decied to change it in the cycle after to admin13:12
sean-k-mooneyzigo: the approch we normally take is add a new policy maintianing the same beahvior then update it after, however that in sope to dicuss in the spec13:15
sean-k-mooneybasiclly do we have an upgrade impact or not13:15
opendevreviewThomas Goirand proposed openstack/nova-specs master: Restrict booting from volume with disk  https://review.opendev.org/c/openstack/nova-specs/+/96051313:16
sean-k-mooneymy expection is if we changed it to admin it would break tempest tests as we dont use speical falvors with 0 disk to test BFV in general13:16
sean-k-mooneyor at least i dont think we do that exclusively13:16
opendevreviewJulien LE JEUNE proposed openstack/nova master: Adds regression test for bug LP#2044235  https://review.opendev.org/c/openstack/nova/+/96034914:14
sean-k-mooneyjlejeune: sorry i am in a few converstaion at once. i see you pushed the tst so i can try and reveiw that patch and play with it loccally 14:31
jlejeunesean-k-mooney: no worry, I saw that you're in discussion with zigo 14:33
sean-k-mooneyand on slack and oslo and watcher channel14:33
jlejeune:)14:33
jlejeuneyou're such a busy man :)14:34
sean-k-mooneythat or i have the attention span fo a magpie14:34
jlejeune^^14:35
sean-k-mooneyso to your question atbout messagingTimeout it might be because of how we mock our messaage bus using the olso in memory oslo.messaging driver14:35
jlejeuneexactly14:35
sean-k-mooneywe sometimes need to disable case as call to get this to work as well14:36
jlejeuneI didn't find any example of that, do you have some in mind ?14:37
sean-k-mooneyno but i think we have or had example of causing live migration to fail in other tests14:37
sean-k-mooneyim not sure if your familar with the CAST_AS_CALL fixture as the name suggest it turn rpc casts which are fire and forget into a sync call that run in place and returns only when its complete14:38
sean-k-mooneyi dont think that is the problem here14:39
sean-k-mooneybut it can cause slightly diffent rpc behavior in teh regress test and sometime we have to turn it off14:39
jlejeuneyes I saw that flag and I tried to turn it off, but without any success14:40
sean-k-mooneyso we do have some regression tests for  live migration roolback https://github.com/openstack/nova/blob/759e03c35d91dac4a04237d58493665c01952c9f/nova/tests/functional/regressions/test_bug_1848343.py#L10514:52
opendevreviewJulien LE JEUNE proposed openstack/nova master: Adds regression test for bug LP#2044235  https://review.opendev.org/c/openstack/nova/+/96034914:53
sean-k-mooneyim wonderign if there is a bette way to similate the failrue14:55
sean-k-mooneyah you now just raising a generic excption14:58
jlejeuneyes because I can't raise the MessagingTimeout exception, if I do so, it's caught by the secheduler when it tries to find a destination host15:01
jlejeunebut after thinking about the issue, for my point of view, it's not very a big deal to catch all exceptions coming from that rpc call during the pre live migration check, what do you think about it ?15:02
opendevreviewJulien LE JEUNE proposed openstack/nova master: Adds regression test for bug LP#2044235  https://review.opendev.org/c/openstack/nova/+/96034915:09
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013016:17
sean-k-mooneyjlejeune: am it depend we do have some logic that may care but lets look at it in the fix patch 16:32
sean-k-mooneyjlejeune: we try to avoid `except Exception` in gernal16:33
sean-k-mooneybecause it can hide effort16:33
sean-k-mooney*errors16:33
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013016:52
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013018:27
opendevreviewLajos Katona proposed openstack/nova master: Use SDK for Neutron networks  https://review.opendev.org/c/openstack/nova/+/92802218:47
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013019:00
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013019:37
opendevreviewBalazs Gibizer proposed openstack/nova master: Rally job for eventlet-removal  https://review.opendev.org/c/openstack/nova/+/96013020:02

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!