Merge pull request #3386 from superseb/cri_dockerd

Add docs on enabling cri-dockerd

content/rke/latest/en/config-options/_index.md

@@ -110,3 +110,11 @@ $ echo $SSH_AUTH_SOCK
 ### Add-ons Job Timeout
 
 You can define [add-ons]({{<baseurl>}}/rke/latest/en/config-options/add-ons/) to be deployed after the Kubernetes cluster comes up, which uses Kubernetes [jobs](https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/). RKE will stop attempting to retrieve the job status after the timeout, which is in seconds. The default timeout value is `30` seconds.
+
+### cri-dockerd
+
+Kubernetes will remove code in the kubelet that interacts with Docker (dockershim) in Kubernetes 1.23. For more information, see [Dockershim Deprecation FAQ](https://kubernetes.io/blog/2020/12/02/dockershim-faq/). The component that replaces this code is called `cri-dockerd` and can be enabled using the following configuration:
+
+```
+enable_cri_dockerd: true
+```

content/rke/latest/en/example-yamls/_index.md

@@ -61,6 +61,14 @@ nodes:
 # are found
 ignore_docker_version: false
 
+# Enable running cri-dockerd
+# Up to Kubernetes 1.23, kubelet contained code called dockershim 
+# to support Docker runtime. The replacement is called cri-dockerd 
+# and should be enabled if you want to keep using Docker as your
+# container runtime
+# Only available to enable in Kubernetes 1.21 and higher
+enable_cri_dockerd: true
+
 # Cluster level SSH private key
 # Used if no ssh information is set for the node
 ssh_key_path: ~/.ssh/test