Rancher pod restarts with feature flag value has changed, rancher must be restarted

Rancher 2.x
1

Rancher Server Setup

  • Rancher version: 2.6.5
  • Installation option : Helm chart
    • EKS (v1.21)
  • Cert Details: ACM

I have deployed with latest stable helm chart (2.6.5), and rancher pod is restarting with the below fatal error. At this time, we are only having one replica to test

feature flag [fleet] value has changed, rancher must be restarted

But later I had updated some features to be disabled -

- name: CATTLE_FEATURES
  value: fleet=true,continous-delivery=false,multi-cluster-management=false,rke2=false,harvester=false

But again, it restarted with the similar error
feature flag [multi-cluster-management] value has changed, rancher must be restarted

What value is being changed here which is causing this restarts, and how can I get the rancher pod running when dealing with the features? Because these are not being passed in 2.6.0 which we were using earlier.