CoreDNS multicluster plugin manual testing

manual_test.md

To check that the plugin builds with the current version of coredns

1. Ran local kind cluster with the ServiceImport CRD installed from the MCS API repo.
2. Built local docker image against coredns/Dockerfile of latest coredns master commit c9eedcb, with multicluster plugin installed via refrencing it in coredns/plugin.cfg.

   docker build $MYDEVHOME/coredns/Dockerfile -t coredns-193-withmc
   

3. Uploaded local image to kind cluster (did you know you needed to do that? I didn't. See KiND - How I Wasted a Day Loading Local Docker Images by Ivan Velichko)

   kind load docker-image coredns-193-withmc:latest
   

4. Patched coredns deployment in kind cluster based on directions from these deployment directions, short version below.

   lauralorenz@lauralorenz:coredns$ kubectl patch deployment coredns -n kube-system -p '{"spec":{"template":{"spec":{"containers":[{"name":"coredns", "image":"coredns-193-withmc:latest"}]}}}}'
   deployment.apps/coredns patched
   #
   # I also updated the coredns/kube-dns Configmap to actually configure multicluster plugin, see multicluster plugin README
   #
   

5. Gave the coredns SA, now that I'm using a version with the multicluster plugin installed, RBAC privileges to the ServiceImport CRD.

   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRole
   metadata:
     name: system:coredns-multicluster
   rules:
   - apiGroups:
     - "multicluster.x-k8s.io"
     resources:
     - serviceimports
     verbs: ["*"]
   ---
   apiVersion: rbac.authorization.k8s.io/v1
   kind: ClusterRoleBinding
   metadata:
     name: system:coredns-multicluster
   roleRef:
     apiGroup: rbac.authorization.k8s.io
     kind: ClusterRole
     name: system:coredns-multicluster
   subjects:
   - kind: ServiceAccount
     name: coredns
     namespace: kube-system
   

6. Confirmed coredns redeployed, logs were healthy using kubectl.

More steps when you want to check that the multicluster plugin does what you think it does

7. Deployed a fake ServiceImport in demo namespace.

   apiVersion: multicluster.x-k8s.io/v1alpha1
   kind: ServiceImport
   metadata:
     name: myservice
     namespace: demo
   spec:
     type: ClusterSetIP
     ips:
     - 1.2.3.4
     ports:
     - port: 80
       protocol: TCP
   

8. Deployed dnsutils pod in demo namespace.

   apiVersion: v1
   kind: Pod
   metadata:
    name: dnsutils
    namespace: demo
   spec:
    containers:
    - name: dnsutils
      image: k8s.gcr.io/e2e-test-images/jessie-dnsutils:1.3
      command:
        - sleep
        - "3600"
      imagePullPolicy: IfNotPresent
    restartPolicy: Always
   

9. Used dnsutils pod to confirm that the DNS query for the serviceimport responds with the IP I set in the fake ServiceImport.

   lauralorenz@lauralorenz:multicluster$ kubectl exec -it dnsutils -n demo -- bash
   root@dnsutils:/# nslookup myservice.demo.svc.clusterset.local
   Server:         10.96.0.10
   Address:        10.96.0.10#53
   
   Name:   myservice.demo.svc.clusterset.local
   Address: 1.2.3.4