Using Multi-Control Plane - HA Clusters

Overview

minikube implements Kubernetes highly available cluster topology using stacked control plane nodes with colocated etcd nodes using kube-vip in ARP mode.

This tutorial will show you how to start and explore a multi-control plane - HA cluster on minikube.

Prerequisites

minikube > v1.32.0
kubectl

Optional

ip_vs kernel modules: ip_vs, ip_vs_rr and nf_conntrack

rationale:

kube-vip supports the control-plane load-balancing by distributing API requests across control-plane nodes using IPVS (IP Virtual Server) and Layer 4 (TCP-based) round-robin.

minikube will try to automatically enable control-plane load-balancing if these ip_vs kernel modules are available, whereas (see the drivers section for more details):

for VM-based drivers (eg, kvm2 or qemu): minikube will automatically try to load ip_vs kernel modules
for container-based or bare-metal-based drivers (eg, docker or “none”): minikube will only check if ip_vs kernel modules are already loaded, but will not try to load them automatically (to avoid unintentional modification of the underlying host’s os/kernel), so it’s up to the user to make them available, if applicable and desired

Caveat

While a minikube HA cluster will continue to operate (although in degraded mode) after loosing any one control-plane node, keep in mind that there might be some components that are attached only to the primary control-plane node, like the storage-provisioner.

Tutorial

optional: if you plan on using a container-based or bare-metal-based driver on top of a Linux OS, check if the ip_vs kernel modules are already loaded

lsmod | grep ip_vs

ip_vs_rr               12288  1
ip_vs                 233472  3 ip_vs_rr
nf_conntrack          217088  9 xt_conntrack,nf_nat,nf_conntrack_tftp,nft_ct,xt_nat,nf_nat_tftp,nf_conntrack_netlink,xt_MASQUERADE,ip_vs
nf_defrag_ipv6         24576  2 nf_conntrack,ip_vs
libcrc32c              12288  5 nf_conntrack,nf_nat,btrfs,nf_tables,ip_vs

Start a HA cluster with the driver and container runtime of your choice (here we chose docker driver and containerd container runtime):

minikube start --ha --driver=docker --container-runtime=containerd --profile ha-demo

😄  [ha-demo] minikube v1.32.0 on Opensuse-Tumbleweed 20240311
✨  Using the docker driver based on user configuration
📌  Using Docker driver with root privileges
👍  Starting "ha-demo" primary control-plane node in "ha-demo" cluster
🚜  Pulling base image v0.0.42-1710284843-18375 ...
🔥  Creating docker container (CPUs=2, Memory=5266MB) ...
📦  Preparing Kubernetes v1.28.4 on containerd 1.6.28 ...
    ▪ Generating certificates and keys ...
    ▪ Booting up control plane ...
    ▪ Configuring RBAC rules ...
🔗  Configuring CNI (Container Networking Interface) ...
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v5
🌟  Enabled addons: storage-provisioner, default-storageclass

👍  Starting "ha-demo-m02" control-plane node in "ha-demo" cluster
🚜  Pulling base image v0.0.42-1710284843-18375 ...
🔥  Creating docker container (CPUs=2, Memory=5266MB) ...
🌐  Found network options:
    ▪ NO_PROXY=192.168.49.2
📦  Preparing Kubernetes v1.28.4 on containerd 1.6.28 ...
    ▪ env NO_PROXY=192.168.49.2
🔎  Verifying Kubernetes components...

👍  Starting "ha-demo-m03" control-plane node in "ha-demo" cluster
🚜  Pulling base image v0.0.42-1710284843-18375 ...
🔥  Creating docker container (CPUs=2, Memory=5266MB) ...
🌐  Found network options:
    ▪ NO_PROXY=192.168.49.2,192.168.49.3
📦  Preparing Kubernetes v1.28.4 on containerd 1.6.28 ...
    ▪ env NO_PROXY=192.168.49.2
    ▪ env NO_PROXY=192.168.49.2,192.168.49.3
🔎  Verifying Kubernetes components...
🏄  Done! kubectl is now configured to use "ha-demo" cluster and "default" namespace by default

List your HA cluster nodes:

kubectl get nodes -owide

NAME          STATUS   ROLES           AGE     VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION    CONTAINER-RUNTIME
ha-demo       Ready    control-plane   4m21s   v1.28.4   192.168.49.2   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m02   Ready    control-plane   4m      v1.28.4   192.168.49.3   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m03   Ready    control-plane   3m37s   v1.28.4   192.168.49.4   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28

Check the status of your HA cluster:

minikube profile list

|---------|-----------|------------|----------------|------|---------|--------|-------|--------|
| Profile | VM Driver |  Runtime   |       IP       | Port | Version | Status | Nodes | Active |
|---------|-----------|------------|----------------|------|---------|--------|-------|--------|
| ha-demo | docker    | containerd | 192.168.49.254 | 8443 | v1.28.4 | HAppy  |     3 |        |
|---------|-----------|------------|----------------|------|---------|--------|-------|--------|

Check the status of your HA cluster nodes:

minikube status -p ha-demo

ha-demo
type: Control Plane
host: Running
kubelet: Running
apiserver: Running
kubeconfig: Configured

ha-demo-m02
type: Control Plane
host: Running
kubelet: Running
apiserver: Running
kubeconfig: Configured

ha-demo-m03
type: Control Plane
host: Running
kubelet: Running
apiserver: Running
kubeconfig: Configured

For a HA cluster, kubeconfig points to the Virual Kubernetes API Server IP

kubectl config view --context ha-demo

apiVersion: v1
clusters:
- cluster:
    certificate-authority: /home/prezha/.minikube/ca.crt
    extensions:
    - extension:
        last-update: Thu, 14 Mar 2024 21:38:07 GMT
        provider: minikube.sigs.k8s.io
        version: v1.32.0
      name: cluster_info
    server: https://192.168.49.254:8443
  name: ha-demo
contexts:
- context:
    cluster: ha-demo
    extensions:
    - extension:
        last-update: Thu, 14 Mar 2024 21:38:07 GMT
        provider: minikube.sigs.k8s.io
        version: v1.32.0
      name: context_info
    namespace: default
    user: ha-demo
  name: ha-demo
current-context: ha-demo
kind: Config
preferences: {}
users:
- name: ha-demo
  user:
    client-certificate: /home/prezha/.minikube/profiles/ha-demo/client.crt
    client-key: /home/prezha/.minikube/profiles/ha-demo/client.key

Overview of the current leader and follower API servers

minikube ssh -p ha-demo -- 'find /var/lib/minikube/binaries -iname kubectl -exec sudo {} --kubeconfig=/var/lib/minikube/kubeconfig logs -n kube-system pod/kube-vip-ha-demo \; -quit'

time="2024-03-14T21:38:34Z" level=info msg="Starting kube-vip.io [v0.7.1]"
time="2024-03-14T21:38:34Z" level=info msg="namespace [kube-system], Mode: [ARP], Features(s): Control Plane:[true], Services:[false]"
time="2024-03-14T21:38:34Z" level=info msg="prometheus HTTP server started"
time="2024-03-14T21:38:34Z" level=info msg="Starting Kube-vip Manager with the ARP engine"
time="2024-03-14T21:38:34Z" level=info msg="Beginning cluster membership, namespace [kube-system], lock name [plndr-cp-lock], id [ha-demo]"
I0314 21:38:34.042420       1 leaderelection.go:250] attempting to acquire leader lease kube-system/plndr-cp-lock...
I0314 21:38:34.069509       1 leaderelection.go:260] successfully acquired lease kube-system/plndr-cp-lock
time="2024-03-14T21:38:34Z" level=info msg="Node [ha-demo] is assuming leadership of the cluster"
time="2024-03-14T21:38:34Z" level=info msg="Starting IPVS LoadBalancer"
time="2024-03-14T21:38:34Z" level=info msg="IPVS Loadbalancer enabled for 1.2.1"
time="2024-03-14T21:38:34Z" level=info msg="Gratuitous Arp broadcast will repeat every 3 seconds for [192.168.49.254]"
time="2024-03-14T21:38:34Z" level=info msg="Kube-Vip is watching nodes for control-plane labels"
time="2024-03-14T21:38:34Z" level=info msg="Added backend for [192.168.49.254:8443] on [192.168.49.3:8443]"
time="2024-03-14T21:38:48Z" level=info msg="Added backend for [192.168.49.254:8443] on [192.168.49.4:8443]"

minikube ssh -p ha-demo -- 'find /var/lib/minikube/binaries -iname kubectl -exec sudo {} --kubeconfig=/var/lib/minikube/kubeconfig logs -n kube-system pod/kube-vip-ha-demo-m02 \; -quit'

time="2024-03-14T21:38:25Z" level=info msg="Starting kube-vip.io [v0.7.1]"
time="2024-03-14T21:38:25Z" level=info msg="namespace [kube-system], Mode: [ARP], Features(s): Control Plane:[true], Services:[false]"
time="2024-03-14T21:38:25Z" level=info msg="prometheus HTTP server started"
time="2024-03-14T21:38:25Z" level=info msg="Starting Kube-vip Manager with the ARP engine"
time="2024-03-14T21:38:25Z" level=info msg="Beginning cluster membership, namespace [kube-system], lock name [plndr-cp-lock], id [ha-demo-m02]"
I0314 21:38:25.990817       1 leaderelection.go:250] attempting to acquire leader lease kube-system/plndr-cp-lock...
time="2024-03-14T21:38:34Z" level=info msg="Node [ha-demo] is assuming leadership of the cluster"

minikube ssh -p ha-demo -- 'find /var/lib/minikube/binaries -iname kubectl -exec sudo {} --kubeconfig=/var/lib/minikube/kubeconfig logs -n kube-system pod/kube-vip-ha-demo-m03 \; -quit'

time="2024-03-14T21:38:48Z" level=info msg="Starting kube-vip.io [v0.7.1]"
time="2024-03-14T21:38:48Z" level=info msg="namespace [kube-system], Mode: [ARP], Features(s): Control Plane:[true], Services:[false]"
time="2024-03-14T21:38:48Z" level=info msg="prometheus HTTP server started"
time="2024-03-14T21:38:48Z" level=info msg="Starting Kube-vip Manager with the ARP engine"
time="2024-03-14T21:38:48Z" level=info msg="Beginning cluster membership, namespace [kube-system], lock name [plndr-cp-lock], id [ha-demo-m03]"
I0314 21:38:48.856781       1 leaderelection.go:250] attempting to acquire leader lease kube-system/plndr-cp-lock...
time="2024-03-14T21:38:48Z" level=info msg="Node [ha-demo] is assuming leadership of the cluster"

Overview of multi-etcd instances

minikube ssh -p ha-demo -- 'find /var/lib/minikube/binaries -iname kubectl -exec sudo {} --kubeconfig=/var/lib/minikube/kubeconfig exec -ti pod/etcd-ha-demo -n kube-system -- /bin/sh -c "ETCDCTL_API=3 etcdctl member list --write-out=table --cacert=/var/lib/minikube/certs/etcd/ca.crt --cert=/var/lib/minikube/certs/etcd/server.crt --key=/var/lib/minikube/certs/etcd/server.key" \; -quit'

+------------------+---------+-------------+---------------------------+---------------------------+------------+
|        ID        | STATUS  |    NAME     |        PEER ADDRS         |       CLIENT ADDRS        | IS LEARNER |
+------------------+---------+-------------+---------------------------+---------------------------+------------+
| 3c464e4a52eb93c5 | started | ha-demo-m03 | https://192.168.49.4:2380 | https://192.168.49.4:2379 |      false |
| 59bde6852118b2a5 | started | ha-demo-m02 | https://192.168.49.3:2380 | https://192.168.49.3:2379 |      false |
| aec36adc501070cc | started |     ha-demo | https://192.168.49.2:2380 | https://192.168.49.2:2379 |      false |
+------------------+---------+-------------+---------------------------+---------------------------+------------+

Loosing a control-plane node - degrades cluster, but not a problem!

minikube node delete m02 -p ha-demo

🔥  Deleting node m02 from cluster ha-demo
✋  Stopping node "ha-demo-m02"  ...
🛑  Powering off "ha-demo-m02" via SSH ...
🔥  Deleting "ha-demo-m02" in docker ...
💀  Node m02 was successfully deleted.

kubectl get nodes -owide

NAME          STATUS   ROLES           AGE     VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION    CONTAINER-RUNTIME
ha-demo       Ready    control-plane   7m16s   v1.28.4   192.168.49.2   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m03   Ready    control-plane   6m32s   v1.28.4   192.168.49.4   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28

minikube profile list

|---------|-----------|------------|----------------|------|---------|----------|-------|--------|
| Profile | VM Driver |  Runtime   |       IP       | Port | Version |  Status  | Nodes | Active |
|---------|-----------|------------|----------------|------|---------|----------|-------|--------|
| ha-demo | docker    | containerd | 192.168.49.254 | 8443 | v1.28.4 | Degraded |     2 |        |
|---------|-----------|------------|----------------|------|---------|----------|-------|--------|

Add a control-plane node

minikube node add --control-plane -p ha-demo

😄  Adding node m04 to cluster ha-demo as [worker control-plane]
👍  Starting "ha-demo-m04" control-plane node in "ha-demo" cluster
🚜  Pulling base image v0.0.42-1710284843-18375 ...
🔥  Creating docker container (CPUs=2, Memory=5266MB) ...
📦  Preparing Kubernetes v1.28.4 on containerd 1.6.28 ...
🔎  Verifying Kubernetes components...
🏄  Successfully added m04 to ha-demo!

kubectl get nodes -owide

NAME          STATUS   ROLES           AGE     VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION    CONTAINER-RUNTIME
ha-demo       Ready    control-plane   8m34s   v1.28.4   192.168.49.2   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m03   Ready    control-plane   7m50s   v1.28.4   192.168.49.4   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m04   Ready    control-plane   36s     v1.28.4   192.168.49.5   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28

minikube profile list

|---------|-----------|------------|----------------|------|---------|--------|-------|--------|
| Profile | VM Driver |  Runtime   |       IP       | Port | Version | Status | Nodes | Active |
|---------|-----------|------------|----------------|------|---------|--------|-------|--------|
| ha-demo | docker    | containerd | 192.168.49.254 | 8443 | v1.28.4 | HAppy  |     3 |        |
|---------|-----------|------------|----------------|------|---------|--------|-------|--------|

Add a worker node

minikube node add -p ha-demo

😄  Adding node m05 to cluster ha-demo as [worker]
👍  Starting "ha-demo-m05" worker node in "ha-demo" cluster
🚜  Pulling base image v0.0.42-1710284843-18375 ...
🔥  Creating docker container (CPUs=2, Memory=5266MB) ...
📦  Preparing Kubernetes v1.28.4 on containerd 1.6.28 ...
🔎  Verifying Kubernetes components...
🏄  Successfully added m05 to ha-demo!

kubectl get nodes -owide

NAME          STATUS   ROLES           AGE     VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION    CONTAINER-RUNTIME
ha-demo       Ready    control-plane   9m35s   v1.28.4   192.168.49.2   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m03   Ready    control-plane   8m51s   v1.28.4   192.168.49.4   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m04   Ready    control-plane   97s     v1.28.4   192.168.49.5   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28
ha-demo-m05   Ready    <none>          22s     v1.28.4   192.168.49.6   <none>        Ubuntu 22.04.4 LTS   6.7.7-1-default   containerd://1.6.28

Test by deploying a hello service, which just spits back the IP address the request was served from:

kubectl apply -f - <<EOF
apiVersion: apps/v1
kind: Deployment
metadata:
  name: hello
spec:
  replicas: 3
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 100%
  selector:
    matchLabels:
      app: hello
  template:
    metadata:
      labels:
        app: hello
    spec:
      affinity:
        # ⬇⬇⬇ This ensures pods will land on separate hosts
        podAntiAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            - labelSelector:
                matchExpressions: [{ key: app, operator: In, values: [hello] }]
              topologyKey: "kubernetes.io/hostname"
      containers:
        - name: hello-from
          image: pbitty/hello-from:latest
          ports:
            - name: http
              containerPort: 80
      terminationGracePeriodSeconds: 1
---
apiVersion: v1
kind: Service
metadata:
  name: hello
spec:
  type: NodePort
  selector:
    app: hello
  ports:
    - protocol: TCP
      nodePort: 31000
      port: 80
      targetPort: http
EOF

deployment.apps/hello created

kubectl rollout status deployment/hello

deployment "hello" successfully rolled out

Check out the IP addresses of our pods, to note for future reference

kubectl get pods -owide

NAME                     READY   STATUS    RESTARTS   AGE   IP           NODE          NOMINATED NODE   READINESS GATES
hello-7bf57d9696-64v6m   1/1     Running   0          18s   10.244.3.2   ha-demo-m04   <none>           <none>
hello-7bf57d9696-7gtlk   1/1     Running   0          18s   10.244.2.2   ha-demo-m03   <none>           <none>
hello-7bf57d9696-99qsw   1/1     Running   0          18s   10.244.0.4   ha-demo       <none>           <none>

Look at our service, to know what URL to hit

minikube service list -p ha-demo

|-------------|------------|--------------|---------------------------|
|  NAMESPACE  |    NAME    | TARGET PORT  |            URL            |
|-------------|------------|--------------|---------------------------|
| default     | hello      |           80 | http://192.168.49.2:31000 |
| default     | kubernetes | No node port |                           |
| kube-system | kube-dns   | No node port |                           |
|-------------|------------|--------------|---------------------------|

Let’s hit the URL a few times and see what comes back

curl  http://192.168.49.2:31000

Hello from hello-7bf57d9696-99qsw (10.244.0.4)

curl  http://192.168.49.2:31000
Hello from hello-7bf57d9696-7gtlk (10.244.2.2)

curl  http://192.168.49.2:31000
Hello from hello-7bf57d9696-7gtlk (10.244.2.2)

curl  http://192.168.49.2:31000
Hello from hello-7bf57d9696-64v6m (10.244.3.2)

Feedback

Was this page helpful?

Glad to hear it! Please tell us how we can improve.

Sorry to hear that. Please tell us how we can improve.

Last modified September 28, 2024: use installed kubectl version in examples (2c6754b0c)