Upgrading Trident Operator Deletes Connectivity to Storage when on Dedicated Network

[jsonesRedHat]

Describe the bug When customer has dedicated storage network that is not on the openshift pod network, one must create a network attachment definition and then apply an annotation to the Trident Controller Deployment to connect to this storage network. For example

apiVersion: apps/v1 kind: Deployment spec: template: metadata: annotations: k8s.v1.cni.cncf.io/networks:

We have learned that when the trident operator does an upgrade, it deletes the deployment and creates a new one resulting in annotation being removed, resulting in lost connection to backend storage. There also seems to be no CRD that enables us make Operator aware of this need for the annotation for connectivity.

Environment Provide accurate information about the environment to help us reproduce the issue.

• Trident version: any
• Trident installation flags used: OpenShift operator hub install and upgrade
• Kubernetes orchestrator: OpenShift 4.17
• :

To Reproduce Steps to reproduce the behavior: Run Trident Operator Upgrade after adding an annotation. It will be deleted

Expected behavior There should be a way to keep annotation in place so that controller has connectivity to backend storage

[torirevilla]

Hi @jsonesRedHat we will investigate and reply when we have an update. Thanks

[rpalhares26]

I have the same issue in my cluster. Do you have an estimate to fix it ? Thanks

[fredsilvas]

I went through the same issue... I had to set the operator to manual update so as not to lose the settings... it would be great to have this fix implemented.

Thanks in advance