docs: Manually create Equinix cluster

docs/create_equinix_cluster.md

@@ -0,0 +1,122 @@
+# Create Equinix Metal cluster using CAPI / CAPEM
+
+## Overview
+
+- We are using machines contributed to the CNCF by Equinix Metal.
+- These steps manually create a cluster using CAPI / CAPEM using a local
+Kind cluster for the management cluster.
+- These docs are based on the Equinix CAPI [guide](https://deploy.equinix.com/developers/guides/kubernetes-cluster-api/).
+- Using [Podman](https://github.com/kubernetes-sigs/cluster-api-provider-packet/issues/496)
+is recommended by Equinix for MacOS.
+- We plan to automate these steps later using GitHub Actions and an IaC tool.
+
+## Create Cluster
+
+- Log in to the Equinix Metal [console](https://console.equinix.com/) and get
+the project API key from the project settings. If it doesn't exist create it.
+
+```sh
+export PACKET_API_KEY="<PROJECT_API_KEY>"
+```
+
+- Install the CAPI controllers using [clusterctl](https://cluster-api.sigs.k8s.io/user/quick-start.html#install-clusterctl).
+
+```sh
+clusterctl init --infrastructure packet
+```
+
+- Set env vars with cluster config.
+
+```sh
+# Get the project ID from the project settings in the console
+export PROJECT_ID="<PROJECT_ID>"
+
+# Use Paris metro (Equinix region)
+export METRO="pa"
+
+# Use Ubuntu 22.04 with cGroup v2 for Kepler
+export NODE_OS="ubuntu_22_04"
+
+# The pod and service CIDRs for the new cluster
+export POD_CIDR="192.168.0.0/16"
+export SERVICE_CIDR="172.26.0.0/16"
+
+# Use node type with Intel CPUs for RAPL support
+export CONTROLPLANE_NODE_TYPE="m3.small.x86"
+export WORKER_NODE_TYPE="m3.small.x86"
+
+# SSH key to use for access to nodes
+export SSH_KEY="<YOUR_SSH_KEY>"
+
+# Kubernetes version to install
+export KUBERNETES_VERSION="v1.28.2"
+```
+
+- Generate cluster manifests.
+
+```sh
+clusterctl generate cluster wg-green-reviews \
+  --kubernetes-version $KUBERNETES_VERSION \
+  --control-plane-machine-count=1 \
+  --worker-machine-count=1 \
+  > wg-green-reviews.yaml
+```
+
+- Apply cluster manifests.
+
+```sh
+kubectl apply -f wg-green-reviews.yaml
+```
+
+- Get kubeconfig and store it securely.
+
+```sh
+clusterctl get kubeconfig wg-green-reviews > wg-green-reviews.kubeconfig
+```
+
+- Set `KUBECONFIG` env var so following commands are run on the cluster.
+
+```sh
+export KUBECONFIG=wg-green-reviews.kubeconfig
+```
+
+- Install Cilium as CNI.
+
+```sh
+helm repo add cilium https://helm.cilium.io/
+helm install cilium cilium/cilium  --version 1.14.2 --namespace kube-system
+```
+
+- SSH to each cluster node and ensure Kepler dependencies are installed
+(user is named `root`).
+
+```sh
+apt update
+apt install linux-headers-$(uname -r)
+apt install linux-modules-extra-$(uname -r) 
+modprobe intel_rapl_common
+```
+
+- Install Kepler.
+
+```sh
+helm repo add kepler https://sustainable-computing-io.github.io/kepler-helm-chart
+helm install kepler kepler/kepler --namespace kepler --create-namespace
+```
+
+- Check Kepler container metrics are non-zero. 
+
+```sh
+kubectl exec -ti -n kepler daemonset/kepler \
+    -- bash -c "curl localhost:9102/metrics" | grep 'kepler_container_package_joules_total'
+```
+
+## Delete Cluster
+
+- If the Kind cluster still exists it can be used to delete the Equinix cluster.
+
+```sh
+kubectl delete cluster wg-green-reviews
+```
+
+- Otherwise delete both servers and the elastic IP via the Equinix console.