Expand storage capacity by using external storage for your Raspberry Pi Cluster

External Storage for Pi Cluster

(Total Setup Time: 35 mins)

With the MicroK8s cluster in place, I decided to expand the External Storage for Raspberry Pi cluster. For this to work, I will use my external 640GB USB hard disk and configure MicroK8s default storage.

Mounting External Storage

(5 mins)

First, I plug my external 640GB USB hard disk into the Raspberry Pi 4 USB3 port (next to the LAN). For displaying information about the available block devices, issue this command and take note of the UUID:

sudo blkid

Since I would like to mount the hard disk automatically to the /mnt/hdd folder, modify fstab:

sudo mkdir /mnt/hdd
sudo nano /etc/fstab

As my external storage is a ntfs hard disk, replace the UUID and add the following to /etc/fstab:

UUID=[UUID] /mnt/hdd ntfs defaults,auto,users,rw,nofail,umask=000,x-systemd.device-timeout=30 0 0

To test mount the new entry before a reboot, run this:

sudo mount -a

And that’s it! After a reboot, the hard disk will be recognised by the Raspberry Pi.

Expanding Storage for MicroK8s

(25 mins)

Because of the errors I encountered, highlighted in the Troubleshooting section, I decided to perform a reset to my existing cluster:

microk8s reset --destroy-storage

After preparing the required folders and I opened the fstab for edit:

sudo mkdir -p /var/snap/microk8s/common/default-storage
sudo mkdir -p /mnt/hdd/microk8s/common/default-storage
sudo nano /etc/fstab

For the external storage to perform a bind mount permanently, add these to the /etc/fstab:

/mnt/hdd/microk8s/common/default-storage /var/snap/microk8s/common/default-storage none bind,user 0 2

And I re-enabled all the previous addons which I had before:

microk8s enable dashboard dns registry

Again, to access dashboard without a token, add –enable-skip-login option to the spec section (microk8s edit uses the vim editor):

sudo microk8s kubectl -n kube-system edit deploy kubernetes-dashboard -o yaml

spec:
  progressDeadlineSeconds: 600
  replicas: 1
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      k8s-app: kubernetes-dashboard
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        k8s-app: kubernetes-dashboard
    spec:
      containers:
      - args:
        - --auto-generate-certificates
        - --namespace=kube-system
        - --enable-skip-login

For serving the dashboard in the background (the skip button will appear only after a reboot or by restarting the MicroK8s):

microk8s kubectl proxy --accept-hosts=.\* --address=0.0.0.0 &

Finally, for viewing dashboard, hit the skip button:

http://192.168.xx.xx:8001/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy

Adding leaf nodes for MicroK8s

(5 mins)

In the previous post, each of the leaf nodes is added by running these:

# on the master node
sudo microk8s add-node

# copy output and run on leaf node
microk8s join 192.168.xx.xx:25000/TphlaAAVchZxMFGpFmrNVFLALxcvNFyn

# check if the nodes are added successfully
microk8s kubectl get nodes

Troubleshooting

After microK8s is started and dashboard is served, no endpoints are available for dashboard:

{
  "kind": "Status",
  "apiVersion": "v1",
  "metadata": {    
  },
  "status": "Failure",
  "message": "no endpoints available for service \"kubernetes-dashboard\"",
  "reason": "ServiceUnavailable",
  "code": 503
}

For my case, when running the below commands, it shows that all pods are having unknown status:

microk8s kubectl get pods -o wide --all-namespaces

When I tried removing addons, MicroK8s also suggests to perform microk8s disable storage:destroy-storage for deleting any storage.

microk8s disable dashboard dns registry