Skip to main content

Custom

To install on any other conformant Kubernetes, use the custom provider. Make sure your infrastructure adheres to the following pre-requisites:

Prerequisites​

Kubernetes​

Have a Kubernetes cluster running with one of the following Kubernetes versions:

  • 1.29
  • 1.30
  • 1.31

and a node pool with at least 12 vCPU and 24 GB RAM.

Default storage class​

The default storage class will be used. If your cluster has a storage class, make sure it is set to be the default:

kubectl patch storageclass <your-storage-class> -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

Use the _rawValues to specify another StorageClass per app. This is an example for Harbor:

app:
  harbor:
    _rawValues:
      persistence:
        persistentVolumeClaim:
          registry:
            storageClass: <your-storage-class>
            size: 5Gi

External IP​

During install a Kubernetes LoadBalancer Service is created that obtains an external IP address. This IP address needs to be accessible from within the cluster. Use Metallb for on-prem installations to allow the installer to create a LoadBalancer Service:

# Install Metallb with Helm
kubectl create namespace mlb
helm repo add metallb https://metallb.github.io/metallb
helm repo update
helm install metallb metallb/metallb -n mlb

sleep 60
# Create the IPAddressPool and L2Advertisement
cat <<EOF | kubectl apply -f -
apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
  name: default-pool
  namespace: mlb
spec:
  addresses:
  - <start-ip>-<end-ip>
---
apiVersion: metallb.io/v1beta1
kind: L2Advertisement
metadata:
  name: default-pool
  namespace: mlb
EOF

A Nginx Ingress Controller is installed. If the cloud provider requires specific annotations to be set on the LoadBalancer Service, add the required annotations to the service in the chart values:

ingress:
    platformClass:
        entrypoint: ''
        annotations:
            - key: service.beta.kubernetes.io/do-loadbalancer-enable-proxy-protocol
              value: true

CNI​

To use the network policies feature, make sure to install the Calico CNI or any other CNI that supports Kubernetes network polices.

Install Tigera Operator:

helm repo add projectcalico https://docs.tigera.io/calico/charts
helm repo update
kubectl create namespace tigera-operator
helm install calico projectcalico/tigera-operator --version v3.26.3 --namespace tigera-operator

Or install Calico minimal:

kubectl apply -f https://raw.githubusercontent.com/projectcalico/calico/v3.26.3/manifests/calico.yaml

Metrics Server​

The Metrics Server is installed. If your provider already installed Metrics Server on their managed Kubernetes service, then disable the Metrics Server:

apps:
  metrics-server:
    enabled: false

API certificates​

If your Kubernetes cluster uses untrusted certificates, make sure to set metrics-server for kubelet-insecure-tls:

apps:
  metrics-server:
    extraArgs:
      kubelet-insecure-tls: true
      kubelet-preferred-address-types: InternalIP

Cluster Autoscaler​

A Cluster Autoscaler is NOT installed. If your provider does not install a Cluster Autoscaler, install one yourself if required.

DNS​

Access to a DNS zone. See the DNS section for more information. The custom provider can be used in combination with any DNS provider.

Install using Helm​

  1. Create a values.yaml:
tee values.yaml<<EOF
cluster:
  name: $CLUSTER_NAME
  provider: custom
# optionally configure metrics-server for kubelet-insecure-tls
apps:
  metrics-server:
    extraArgs:
      kubelet-insecure-tls: true
      kubelet-preferred-address-types: InternalIP
# dns is required!
dns:
  domainFilters: 
    - <your-domain>
  provider:
    linode:
      apiToken: $LINODE_TOKEN
EOF
  1. Add the repository:
helm repo add apl https://linode.github.io/apl-core
helm repo update
  1. Install the Helm chart:
helm install -f values.yaml apl apl/apl