Alternative installation scenarios

The installation can be performed using the following commands in order:

This will install Supervisor on the selected worker node as a pod.

1. Create a directory on the desired worker node, to be used to store the supervisor configuration and license:

mkdir -p /home/user/supervisor-data/

This is only a reference path used for this documentation. If a different path is used, edit the following commands accordingly.

2. Copy the provided license file in the data directory on the worker node:

cp SFM-010010-10.lic /home/user/supervisor-data/license.lic

Replace the name of the file in this command with the actual license file provided.

3. Load the provided Supervisor docker container on the worker node (replace X.Y.Z with the appropriate version number).

Kubernetes uses containerd's own image store, hence we import into containerd in order for Kubernetes to see it.

sudo ctr -n k8s.io images import profitap-supervisor-vX.Y.Z.tar

4. Create the following YAML file (e.g. sv.yml) on the master node, with image, hostPath path, and nodeSelector hostname adjusted as per your setup:

sv.yml

apiVersion: v1
kind: Pod
metadata:
  name: supervisor-pod
  labels:
    app: supervisor
spec:
  hostNetwork: true  # Use host networking
  containers:
    - name: supervisor
      image: profitap-supervisor:v1.1.0  # Use the image name after loading
      imagePullPolicy: IfNotPresent  # Use IfNotPresent since the image is local
#     env:
#       - name: SUPERVISOR_THREADS
#         value: "8"      
      volumeMounts:
        - mountPath: /data  # Path in the container
          name: supervisor-storage  # Name of the volume
  volumes:
    - name: supervisor-storage  # Volume name
      hostPath:
        path: /home/profitap/supervisor-data  # Updated local path on the host
  nodeSelector:
    kubernetes.io/hostname: workernode1  # Updated hostname

5. Run the sv.yml file with kubectl on the master node:

kubectl apply -f sv.yml

At this point, the Supervisor pod should be running. If you wish to verify that deployment has proceeded correctly, you can check the running containers using the following command:

kubectl get pods

The installation can be performed using the following commands in order:

This will install Supervisor on the selected worker node as a deployment.

1. Create a directory on the desired worker node, to be used to store the supervisor configuration and license:

mkdir -p /home/user/supervisor-data/

This is only a reference path used for this documentation. If a different path is used, edit the following commands accordingly.

2. Copy the provided license file in the data directory on the worker node:

cp SFM-010010-10.lic /home/user/supervisor-data/license.lic

Replace the name of the file in this command with the actual license file provided.

3. Load the provided Supervisor docker container on the worker node (replace X.Y.Z with the appropriate version number).

Kubernetes uses containerd's own image store, hence we import into containerd in order for Kubernetes to see it.

sudo ctr -n k8s.io images import profitap-supervisor-vX.Y.Z.tar

4. Create the following YAML file (e.g. sv-deployment.yml) on the master node, with image, hostPath path, and nodeSelector hostname adjusted as per your setup:

sv-deployment.yml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: supervisor-deployment
  labels:
    app: supervisor
spec:
  replicas: 1
  selector:
    matchLabels:
      app: supervisor
  template:
    metadata:
      labels:
        app: supervisor
    spec:
      hostNetwork: true  # Use host networking
      containers:
        - name: supervisor
          image: profitap-supervisor:v1.1.0
          imagePullPolicy: IfNotPresent  # Use local image if available
#         env:
#           - name: SUPERVISOR_THREADS
#             value: "8"
          volumeMounts:
            - mountPath: /data
              name: supervisor-storage
      volumes:
        - name: supervisor-storage
          hostPath:
            path: /home/profitap/supervisor-data
      nodeSelector:
        kubernetes.io/hostname: workernode1

5. Run the sv-deployment.yml file with kubectl on the master node:

kubectl apply -f sv-deployment.yml

At this point, the Supervisor deployment should be running. If you wish to verify that deployment has proceeded correctly, you can check the running containers using the following commands:

kubectl get pods
kubectl get deployments

The installation can be performed using the following commands in order:

This will install Supervisor on any worker node as a deployment with remote data directory. The remote data directory is an NFS share (CIFS is not supported).

1. Create the following YAML file (e.g. pv-nfs.yml), with server IP and path adjusted as per your setup:

pv-nfs.yml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: nfs-pv
spec:
  capacity:
    storage: 5Gi
  accessModes:
    - ReadWriteMany
  nfs:
    path: /volume1/remotedata
    server: 10.10.10.233  
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs-storage

Run the pv-nfs.yml file with kubectl:

kubectl apply -f pv-nfs.yml

This will create a remote persistent volume.

Note: NFS tools need to exist on worker nodes. Tools can be installed using:

sudo apt update
sudo apt install -y nfs-common

2. Create the following YAML file (e.g. pvc-nfs.yml):

pvc-nfs.yml

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: supervisor-pvc
spec:
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 5Gi
  storageClassName: nfs-storage
  volumeName: nfs-pv # Explicitly bind to the specific PV

Run the pvc-nfs.yml file with kubectl:

kubectl apply -f pvc-nfs.yml

This will create the PVC on kubernetes cluster.

3. Create the following YAML file (e.g. sv-deployment-remote.yml), replacing the image name with the appropriate image name:

sv-deployment-remote.yml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: supervisor-deployment
  labels:
    app: supervisor
spec:
  replicas: 1
  selector:
    matchLabels:
      app: supervisor
  template:
    metadata:
      labels:
        app: supervisor
    spec:
      hostNetwork: true  # Use host networking
      containers:
        - name: supervisor
          image: profitap-supervisor:v1.1.0
          imagePullPolicy: IfNotPresent  # Use local image if available
          # env:
          #   - name: SUPERVISOR_THREADS
          #     value: "8"
          volumeMounts:
            - name: supervisor-storage
              mountPath: /data  # Same container path
      volumes:
        - name: supervisor-storage
          persistentVolumeClaim:
            claimName: supervisor-pvc  # Using the PVC instead of hostPath

Run the sv-deployment-remote.yml file with kubectl:

kubectl apply -f sv-deployment-remote.yml

At this point, the Supervisor deployment should be running. If you wish to verify that deployment has proceeded correctly, you can check the running containers using the following commands:

kubectl get pods 
kubectl get deployments