-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
4.14
-
None
Description of problem:
CSI pods don't deploy when there is no AZ defined in Cinder
Version-Release number of selected component (if applicable):
4.14
How reproducible:
Try a deployment of OCP in OSP18 DP3
Steps to Reproduce:
1. 2. 3.
Actual results:
CSI deployment is blocked and showing following state OpenStackCinderCSIDriverOperatorCRDegraded: ConfigSyncDegraded: couldn't collect info about cloud availability zones: failed to generate OpenStack cloud info: could not find an available volume availability zone OpenStackCinderCSIDriverOperatorCRProgressing: OpenStackCinderDriverControllerServiceControllerProgressing: Waiting for Deployment to deploy pods OpenStackCinderCSIDriverOperatorCRProgressing: OpenStackCinderDriverNodeServiceControllerProgressing: Waiting for DaemonSet to deploy node pods
Expected results:
Additional info:
From the conversation [1] in slack, it appears to have been triggered by a config change in Cinder which is not returning any AZ anymore in OSP18 1. https://redhat-internal.slack.com/archives/CH98TDJUD/p1708616694485879