Service Type

[metrue]

The type property in the Service's spec determines how the service is exposed to the network. It changes where a Service is able to be accessed from. The possible types are ClusterIP, NodePort, and LoadBalancer

• ClusterIP, The default value. The service is only accessible from within the Kubernetes cluster – you can’t make requests to your Pods from outside the cluster!
• NodePort, This makes the service accessible on a static port on each Node in the cluster. This means that the service can handle requests that originate from outside the cluster.
• LoadBalancer, The service becomes accessible externally through a cloud provider's load balancer functionality. GCP, AWS, Azure, and OpenStack offer this functionality. The cloud provider will create a load balancer, which then automatically routes requests to your Kubernetes Service
• ExternalName, it maps the Service to the contents of the externalName field (e.g. foo.bar.example.com), by returning a CNAME record with its value. No proxying of any kind is set up.

[metrue]

NodePort usage

For example, you have a running pod test-fx-pod, and expose to service test-fx-pod-svc with NodePort type,

$
kubectl expose deployment test-fx-pod  --name=local_access --port=3000 --target-port=3000 node-port=30001 --type=NodePort

Then you can access the pod with following different ways,

• Access pod with pod's IP

$ kubectl describe pod test-fx-pod

Name:               test-fx-pod
Namespace:          default
Priority:           0
PriorityClassName:  <none>
Node:               fx-test-control-plane/172.17.0.3
Start Time:         Sat, 28 Sep 2019 18:27:28 +0800
Labels:             fx-app=fx-app
Annotations:        <none>
Status:             Running
IP:                 10.244.0.12
Containers:
  fx-placeholder-container-name:
    Container ID:   containerd://04d235365c1fdf2fce22a2ab8a327a207bbda2782d1fab08b3035d9aed32ccdd
    Image:          metrue/kube-hello
    Image ID:       docker.io/metrue/kube-hello@sha256:514e0802941c6608b1d2efe4d676e5edff43871544211b95b234e18a9f921ef7
    Port:           3000/TCP
    Host Port:      3000/TCP
    State:          Running
      Started:      Sat, 28 Sep 2019 18:27:31 +0800
    Ready:          True
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-xwg5b (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             True
  ContainersReady   True
  PodScheduled      True
Volumes:
  default-token-xwg5b:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-xwg5b
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:          <none>

Then you can access it with,

$ kubectl exec -it test-fx-pod curl 10.244.0.12:3000

or

kubectl exec -it test-fx-pod curl 127.0.0.1:3000

• Access it with serivce IP

$ kubectl describe svc  test-fx-pod-svc

Name:                     test-fx-pod-svc
Namespace:                default
Labels:                   <none>
Annotations:              <none>
Selector:                 fx-app=fx-app
Type:                     NodePort
IP:                       10.104.28.208
Port:                     fx-function-as-an-api  3000/TCP
TargetPort:               3000/TCP
NodePort:                 fx-function-as-an-api  30001/TCP
Endpoints:                10.244.0.12:3000
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>

then,

$ kubectl exec -it test-fx-pod curl 10.244.0.12:3000

• Access it with cluster IP

$ kubectl get svc test-fx-pod-svc

NAME              TYPE       CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
test-fx-pod-svc   NodePort   10.104.28.208   <none>        3000:30001/TCP   15h

then you can,

kubectl exec -it test-fx-pod curl 10.104.28.208:3000

[metrue]

But how to access the pod's functionality with kubectl exec, like cURL for instance?

[metrue]

The reason discussed here,

https://github.com/kubernetes-sigs/kind/issues/99