Description of problem: Problem description https://github.com/kubernetes/kubernetes/issues/44735 If I understood correctly fixing this is one of the goals in 3.6, 3.7 and 3.8 openshift. One solution possibilities for this is available in https://github.com/kubernetes/kubernetes/pull/44798 I hope that you could include this pr already to 3.6 Version-Release number of selected component (if applicable): Openshift Container Platform 3.5 How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: Description of problem: Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Master Log: Node Log (of failed PODs): PV Dump: PVC Dump: StorageClass Dump (if StorageClass used by PV/PVC): Additional info: Maybe related to https://bugzilla.redhat.com/show_bug.cgi?id=1444828
Tested on v3.6.131. Currently we do not have an OpenStack that supports creating cinder volumes in multiple AZs. This bug is verified in this way: 1. Create an OCP cluster with 3 nodes. 2. For each node, label that node with a fake availability zone name, eg: oc label node $nodename1 failure-domain.beta.kubernetes.io/zone=zone1 --overwrite=true oc label node $nodename2 failure-domain.beta.kubernetes.io/zone=zone2 --overwrite=true oc label node $nodename3 failure-domain.beta.kubernetes.io/zone=zone3 --overwrite=true 3. Create StorageClass that has no parameters: apiVersion: storage.k8s.io/v1 kind: StorageClass metadata: name: cinder provisioner: kubernetes.io/cinder 4. Use PVC to provision PV, verify that master tries to select zone from the 3 zones instead of the default 'nova': Jul 4 22:43:44 host-8-172-177 atomic-openshift-master: I0704 22:43:44.978819 17138 util.go:360] Creating volume for PVC "cinderc"; chose zone="zone1" from zones=["zone1" "zone2" "zone3"] Jul 4 22:43:44 host-8-172-177 atomic-openshift-master: I0704 22:43:44.977990 17138 util.go:360] Creating volume for PVC "cinderc1"; chose zone="zone3" from zones=["zone1" "zone2" "zone3"] At this point the availability zone is tested and I think this bug is verified.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2017:1716