When the operator starts for the first time, for example, the CR might now exist yet. In that that case, we shouldn't try to delete the CSI driver. We should only delete the operand if the CR has a deletion timestamp in it.
Verified with: # oc get csv -n openshift-aws-ebs-csi-driver-operator NAME DISPLAY VERSION REPLACES PHASE aws-ebs-csi-driver-operator.v4.5.0 AWS EBS CSI Operator 4.5.0 Succeeded ose-aws-ebs-csi-driver-operator-bundle-container-v4.5-6 ose-aws-ebs-csi-driver-operator-container-v4.5-5 # oc describe awsebsdriver cluster|grep "Deletion Timestamp" Deletion Timestamp: 2020-05-09T11:52:36Z # oc get pod -n openshift-aws-ebs-csi-driver-operator NAME READY STATUS RESTARTS AGE aws-ebs-csi-driver-operator-f84975c47-f7p8d 1/1 Running 0 57s # oc get pod -n openshift-aws-ebs-csi-driver NAME READY STATUS RESTARTS AGE aws-ebs-csi-driver-controller-5467c64898-gswbw 5/5 Running 0 7m6s aws-ebs-csi-driver-node-8cdpk 3/3 Running 0 7m6s aws-ebs-csi-driver-node-fjm75 3/3 Running 0 7m6s aws-ebs-csi-driver-node-wx7xd 3/3 Running 0 7m6s # oc get awsebsdriver No resources found in default namespace.
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/RHBA-2020:2409