superseb

#!/usr/bin/env bash

if [ $# -ne 2 ]; then
  echo "Usage: $0 [filename] [etcd_version]"
  exit 1
fi

FILENAME=$1
ETCD_VERSION=$2
TIMESTAMP="$(date +%s)"

superseb

k3s etcd commands

etcd

Setup etcdctl using the instructions at https://github.com/etcd-io/etcd/releases/tag/v3.4.13 (changed path to /usr/local/bin):

ETCD_VER=v3.4.13

# choose either URL

superseb

Generate ECDSA CA, intermediate CA and server certificate with DNS alt names using Terraform in Docker and launch Rancher

Generate ECDSA CA, intermediate CA and server certificate

docker run --rm -v $PWD/testcerts:/tmp/certs/files -e TF_VAR_ip_addresses='["127.0.0.1"]' -e TF_VAR_dns_names='["yolo.seb.local"]' superseb/intermediate-ecdsa

Run Rancher

superseb

Get kube-admin kubeconfig and certificates from cluster.rkestate

See how to retrieve cluster.rkestate from controlplane node here: https://gist.github.com/superseb/e9f2628d1033cb20e54f6ee268683a7a

Get kube-admin kubeconfig from cluster.rkestate

cat cluster.rkestate | jq -r '.currentState.certificatesBundle."kube-admin".config' > kube-admin-kubeconfig.yml

superseb

#!/bin/bash
if [ "$#" -eq 0 ]; then
    echo "No FQDN provided as first parameter, generating xip.io based on found external IP"
    FOUNDIP=$(docker run --rm --net=host appropriate/curl https://api.ipify.org)
    FQDN="minio.${FOUNDIP}.xip.io"
else
    FQDN=$1
fi

echo "Using FQDN: ${FQDN}"

superseb

RKE2 commands

Install

curl -sL https://get.rke2.io | sh
systemctl daemon-reload
systemctl start rke2-server

superseb

TMPDIR=$(mktemp -d $MKTEMP_BASEDIR)
# k3s
if $(command -v k3s >/dev/null 2>&1); then
        mkdir -p $TMPDIR/k3s/crictl
        mkdir -p $TMPDIR/k3s/logs
        mkdir -p $TMPDIR/k3s/podlogs
        mkdir -p $TMPDIR/k3s/kubectl
        k3s check-config > $TMPDIR/k3s/check-config 2>&1
        k3s kubectl get nodes -o json > $TMPDIR/k3s/kubectl/nodes 2>&1
        k3s kubectl version > $TMPDIR/k3s/kubectl/version 2>&1

superseb

Identify RKE / Rancher Launched Kubernetes / Imported clusters

This will describe how to identify clusters that are created by RKE, created by Rancher or managed by Rancher

RKE

Clusters created by RKE CLI have:

• Cluster state stored as Configmap called full-cluster-state in namespace kube-system (cluster-state before RKE v0.2.0)
  • kubectl -n kube-system get configmap full-cluster-state

superseb

{
 "K8sVersionServiceOptions": {
  "v1.10": {
   "etcd": null,
   "kubeapi": {
    "allow-privileged": "true",
    "anonymous-auth": "false",
    "bind-address": "0.0.0.0",
    "enable-admission-plugins": "NamespaceLifecycle,LimitRanger,ServiceAccount,DefaultStorageClass,DefaultTolerationSeconds,MutatingAdmissionWebhook,ValidatingAdmissionWebhook,ResourceQuota,NodeRestriction",
    "insecure-port": "0",

superseb

Rancher 2 HA using Helm and self signed certificate (certificate from files)

This will only cover the part of installing Rancher on a RKE built cluster, see https://rancher.com/docs/rancher/v2.x/en/installation/ha/ how to get there.

Note: make sure kubeconfig is configured correctly

The commands are for Linux, if you are using Mac then you can use md5 instead of md5sum and base64 -D instead of base64 -d.

Generate certificates