Instructions for using external etcd cluster are incomplete

[scottslowe]

What steps did you take and what happened:

I followed the instructions found at https://github.com/kubernetes-sigs/cluster-api/blob/master/bootstrap/kubeadm/docs/external-etcd.md and noted several errors in the document:

1. One of the kubectl create secret commands references the etcd server certificate and key, but it should reference the etcd CA.
2. The kubectl create secret tls command requires both certificate and key, but one of the commands references only the certificate.
3. The instructions don't indicate that users will need to manually update security groups when using this procedure on AWS. (I would imagine that similar steps will be necessary for Azure, GCP, etc., although I haven't tested this process on those platforms yet.)

What did you expect to happen:

I expected the instructions to provide correct information on how to use an external etcd cluster with Cluster API.

Anything else you would like to add:

I have only tested this process using CAPA (so far).

Environment:

• Cluster-api version: 0.3.3
• Minikube/KIND version: N/A (tested on AWS)
• Kubernetes version: (use kubectl version): v1.16.4 (management cluster), v1.17.3 (workload cluster)
• OS (e.g. from /etc/os-release): Ubuntu 18.04

/kind bug

[fabriziopandini]

/kind documentation

[vincepri]

/milestone v0.3.x

[CecileRobertMichon]

/priority important-soon

[CecileRobertMichon]

/help

[scottslowe]

As this is something I'm actively testing right now, I'd like to take this issue.

/assign

[detiber]

/lifecycle active
/remove-help

[vincepri]

/milestone v0.3.6