Monday, 2025-06-16

opendevreviewRajesh Tailor proposed openstack/nova-specs master: Show finish_time field in instance action show  https://review.opendev.org/c/openstack/nova-specs/+/92978006:10
opendevreviewLajos Katona proposed openstack/os-vif stable/2025.1: OVS Trunk: Add bridge_name to external_ids  https://review.opendev.org/c/openstack/os-vif/+/95256708:57
*** tosky_ is now known as tosky09:49
opendevreviewMerged openstack/nova master: Remove Unicode characters  https://review.opendev.org/c/openstack/nova/+/95262111:11
opendevreviewBalazs Gibizer proposed openstack/nova master: Warn on long task wait time for executor  https://review.opendev.org/c/openstack/nova/+/95266611:23
*** elodilles is now known as elodilles_pto12:20
opendevreviewMerged openstack/nova master: api: Address issues with instance actions API  https://review.opendev.org/c/openstack/nova/+/95194115:05
UgglaReminder that tomorrow is spec review day.16:12
bauzasUggla: good point, thanks for the reminder !16:39
Ugglabauzas, you are welcome.16:39
sean-k-mooneyoh ya... i had forgotten about that16:39
UgglaAnd we have a lot of spec to review.16:42
gibidansmith: sean-k-mooney: do I see it correctly that the build_and_run_instance does not ever report error and revert task state at failure via these decorators as the implementation immediately calls spawn but never waits for the result so the error is not raised back? https://github.com/openstack/nova/blob/64ca204c9cf497b0dcfff2d3a24b0dd795a57d1d/nova/compute/manager.py#L2341-L234417:33
gibiI need to drop but I can read back tomorrow and try to reproduce it as a bug17:34
dansmithlooks like it17:35
dansmithmaybe those decorators need to be moved to the inner _locked funciton17:35
dansmiththe finally in there will mark the instance is error I think, so it might only be the task_state (not important on a new instance you are going to delete) and the event (which might not be noticed either)17:36
sean-k-mooneyya that looks like what is happening dan is likely correct about moving them although we woudl have to look at git blame to see when/how they were orgianlly added18:29
sean-k-mooneyit looks like we used to do some network validation https://opendev.org/openstack/nova/commit/1d8eddb2614de8daaddddd64ad1a8de4c215fe7a18:32
sean-k-mooneywell no not validation exactly18:33
sean-k-mooneyits possibel althogh unlikely that the objects.NetworkRequest.from_tuple(t) calls coudl have raised.18:34
sean-k-mooney_do_build_and_run_instance seams to have the real logic that we need those decorator for.18:35
sean-k-mooneyah https://opendev.org/openstack/nova/commit/437a1e92c11d51fb828d355091cdcdd4694dfed7 that when they were actully added18:37
sean-k-mooneyand orgianlly we did not use spawn_n. it is technically possibel for  @utils.synchronized(instance.uuid) to raise as well18:38
opendevreviewmelanie witt proposed openstack/nova master: Add vtpm_secret_(uuid|value) to LiveMigrateData  https://review.opendev.org/c/openstack/nova/+/95262821:32
opendevreviewmelanie witt proposed openstack/nova master: TPM: support live migration of `host` secret security  https://review.opendev.org/c/openstack/nova/+/94148321:32
opendevreviewmelanie witt proposed openstack/nova master: TPM: support instances with `deployment` secret security  https://review.opendev.org/c/openstack/nova/+/94202121:32
opendevreviewmelanie witt proposed openstack/nova master: TPM: support live migration of `deployment` secret security  https://review.opendev.org/c/openstack/nova/+/92577121:32
opendevreviewmelanie witt proposed openstack/nova master: TPM: test live migration between hosts with different security  https://review.opendev.org/c/openstack/nova/+/95262921:32
opendevreviewmelanie witt proposed openstack/nova master: TPM: update instance request_spec with secret security  https://review.opendev.org/c/openstack/nova/+/95263021:32

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