loadbalancer

Kubernetes load balancer

Before starting with installation, make sure you meet all the requirements. In particular, you should pay attention to network addon compatibility.

If you’re trying to run MetalLB on a cloud platform, you should also look at the cloud compatibility page and make sure your cloud platform can work with MetalLB (most cannot).

There are three supported ways to install MetalLB: using plain Kubernetes manifests, using Kustomize, or using Helm.

Run

Switch to the first cluster:

export KUBECONFIG=$KUBECONFIG1

Apply metallb for the first cluster:

if [[ ! -z $CLUSTER1_CIDR ]]; then
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
  kubectl create secret generic -n metallb-system memberlist --from-literal=secretkey="$(openssl rand -base64 128)" 
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml
  cat > metallb-config.yaml <<EOF
  apiVersion: v1
  kind: ConfigMap
  metadata:
    namespace: metallb-system
    name: config
  data:
    config: |
      address-pools:
      - name: default
        protocol: layer2
        addresses:
        - $CLUSTER1_CIDR
EOF
  kubectl apply -f metallb-config.yaml
  kubectl wait --for=condition=ready --timeout=5m pod -l app=metallb -n metallb-system
fi

Switch to the second cluster:

export KUBECONFIG=$KUBECONFIG2

Apply metallb for the second cluster:

if [[ ! -z $CLUSTER2_CIDR ]]; then
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
  kubectl create secret generic -n metallb-system memberlist --from-literal=secretkey="$(openssl rand -base64 128)" 
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml
  cat > metallb-config.yaml <<EOF
  apiVersion: v1
  kind: ConfigMap
  metadata:
    namespace: metallb-system
    name: config
  data:
    config: |
      address-pools:
      - name: default
        protocol: layer2
        addresses:
        - $CLUSTER2_CIDR
EOF
  kubectl apply -f metallb-config.yaml
  kubectl wait --for=condition=ready --timeout=5m pod -l app=metallb -n metallb-system
fi

Switch to the third cluster:

export KUBECONFIG=$KUBECONFIG3

Apply metallb for the second cluster:

if [[ ! -z $CLUSTER3_CIDR ]]; then
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
  kubectl create secret generic -n metallb-system memberlist --from-literal=secretkey="$(openssl rand -base64 128)" 
  kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml
  cat > metallb-config.yaml <<EOF
  apiVersion: v1
  kind: ConfigMap
  metadata:
    namespace: metallb-system
    name: config
  data:
    config: |
      address-pools:
      - name: default
        protocol: layer2
        addresses:
        - $CLUSTER3_CIDR
EOF
  kubectl apply -f metallb-config.yaml
  kubectl wait --for=condition=ready --timeout=5m pod -l app=metallb -n metallb-system
fi

Cleanup

Delete metallb-system namespace from all clusters:

export KUBECONFIG=$KUBECONFIG1 
if [[ ! -z $CLUSTER1_CIDR ]]; then
  kubectl delete ns metallb-system  
fi

export KUBECONFIG=$KUBECONFIG2
if [[ ! -z $CLUSTER2_CIDR ]]; then
  kubectl delete ns metallb-system  
fi

export KUBECONFIG=$KUBECONFIG3
if [[ ! -z $CLUSTER3_CIDR ]]; then
  kubectl delete ns metallb-system  
fi