Page https://redis.io/docs/latest/operate/kubernetes/re-clusters/cluster-recovery/#recover-a-cluster
Some less experienced K8s customers may get tripped up by this command because it doesn't specify the namespace ("-n" flag). If their kubectl session context is not already set to the appropriate namespace of their REC, they will get an error from the K8s API about a missing resource that they might not understand.
I suggest we either include a templated namespace flag in the command itself...
kubectl patch rec <cluster-name> -n <rec_namespace> --type merge --patch '{"spec":{"clusterRecovery":true}}'
... or we could include a note that mentions their kubectl context must be set to the namespace of the REC beforehand.
kubectl config set-context --current --namespace=<rec_namespace>
If (and only if) their context is already set correctly, the -n <rec_namespace> flag is optional. If not set correctly, the command will not work without it.
More experienced K8s users will know that this flag is optional when they've already set their kubectl context correctly, but less experienced ones may not know it's required when they've also forgotten to set the context.