How do I set the imagePullPolicy with Minikube

Question

I am new to Kubernetes. I am trying to follow this tutorial that instructs me on how to use minikube to setup a local service. I was able to get things running with the $ kubectl run hello-minikube --image=k8s.gcr.io/echoserver:1.10 --port=8080 service from the tutorial. Huzzah!

Now I want to run a server with a locally tagged-and-built Docker image. According to this post all I need to do is tell my computer to use the minikube docker daemon, build my image, and set the imagePullPolicy to never.

How and where do I set the imagePullPolicy with minikube? I've googled around and while there's plenty of results, my "babe in the woods" status with K8 leads to information overload. (i.e. the simpler your answer the better)

@AndrewSavinykh MacOS -- which I think means things are running behind the scenes in a linux VM? — Alana Storm, Dec 21 '18 at 00:32
Right, sorry asked because I like to test stuff out before I set people on the wrong route and I don't have MacOS, so hopefully someone else can help. — Andrew Savinykh, Dec 21 '18 at 00:36

score 19 · Accepted Answer · answered Dec 21 '18 at 01:41

You have to edit your Deployment (kubectl run creates a deployment). The spec would look something like this:

spec:
  progressDeadlineSeconds: 600
  replicas: 1
  revisionHistoryLimit: 2
  selector:
    matchLabels:
      run: hello-minikube
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        run: hello-minikube
    spec:
      containers:
      - image: k8s.gcr.io/echoserver:1.10 <-- change to the right image
        imagePullPolicy: IfNotPresent <-- change to Always
        name: hello-minikube
        ports:
        - containerPort: 8080
          protocol: TCP
        resources: {}
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30

Edit with:

$ kubectl edit deployment hello-minikube

Ah, got it. I'd seen the `deployment` used in the tutorial but didn't understand it. You answered both my original question, and one I didn't realize I had. Thank you! — Alana Storm, Dec 21 '18 at 04:49

How do I set the imagePullPolicy with Minikube

1 Answers1