-
Bug
-
Resolution: Done-Errata
-
Undefined
-
None
-
4.15
-
?
-
Important
-
Yes
-
False
-
-
-
Bug Fix
-
In Progress
-
-
-
-
Description of problem:
After upgrading the OCP cluster from 4.14.22 to 4.15.11 customer started to see the aws-efs-csi-driver-controller pods in crash look back off status in the openshift-cluster-csi-drivers namespace
Version-Release number of selected component (if applicable):
OCP 4.15.11 AWS EFS CSI Driver v1.7.0
How reproducible:
It happens all the time in customer environment
Steps to Reproduce:
1. Upgrade OCP from 4.14 to 4.15, AWS driver version that comes with 4.15 is v1.7.0, while with 4.14 was v1.5.3 2. Try to provision a new volume with AWS EFS CSI driver
Actual results:
Provisioning fails with an "invalid memory address or nil pointer dereference"
Expected results:
Provisioning should work without issues
Additional info:
Issue described seems similar to what was reported in https://github.com/kubernetes-sigs/aws-efs-csi-driver/issues/1164 and https://github.com/kubernetes-sigs/aws-efs-csi-driver/issues/1173 PR with the fix was https://github.com/kubernetes-sigs/aws-efs-csi-driver/pull/1167 but it wasn't merged According to https://github.com/kubernetes-sigs/aws-efs-csi-driver/issues/1164#issuecomment-1806237765, this was already fixed in v.1.7.1 but this version of the driver isn't available in OCP 4.15.
- is blocked by
-
OCPBUGS-35014 AWS EFS CSI driver: panic: runtime error: invalid memory address or nil pointer dereference
- Closed
- is cloned by
-
OCPBUGS-35014 AWS EFS CSI driver: panic: runtime error: invalid memory address or nil pointer dereference
- Closed
- links to
-
RHBA-2024:3889 OpenShift Container Platform 4.15.z bug fix update