Upgrade Confluent for Kubernetes and Confluent Platform

Before you start the upgrade process, make sure your Kubernetes cluster is among the Supported Environments for the target version of Confluent for Kubernetes.

In general, it is best to minimize the number of changes occurring at the same time. For the purposes of upgrading Confluent Platform from one version to another, we recommend that you perform this upgrade separate from making other changes to your clusters, such as scaling up, changing configuration, or rotating credentials.

Upgrade Confluent for Kubernetes 2.0.0 to 2.0.1

When you upgrade Confluent for Kubernetes (CFK) from 2.0.0 to 2.0.1, all Confluent Platform components will rolling restart.

To upgrade CFK:

  1. Get the latest CFK chart:

    helm repo update
  2. Upgrade CFK to 2.0.1:

    helm upgrade --install confluent-operator \
      confluentinc/confluent-for-kubernetes \
      --namespace <namespace>

Upgrade Confluent Platform

Upgrade each Confluent Platform component as below.

  1. Temporarily disable resource reconciliation for each namespace where you have deployed the Confluent Platform component.

    If you do not disable the reconcile operations before you run the upgrade, component will begin rolling restarts and the pods may go into the CrashLoopBackOff status.

    kubectl annotate <component CR type> <cluster name> \
      platform.confluent.io/block-reconcile=true \
      --namespace <namespace>
  2. Upgrade the component:

    kubectl apply -f <component CR file> --name <namespace>
  3. Enable reconciliation:

    kubectl annotate <component CR type> <cluster name> \
      platform.confluent.io/block-reconcile- \
      --namespace <namespace>