This page shows how to deploy an application in your user cluster.
SSH into your admin workstation
SSH into your admin workstation:
ssh -i ~/.ssh/vsphere_workstation ubuntu@[IP_ADDRESS]
where [IP_ADDRESS] is the IP address of your admin workstation.
Do all of the remaining steps in this topic on your admin workstation.
Creating a Deployment
Here is a manifest for a Deployment:
apiVersion: apps/v1
kind: Deployment
metadata:
name: my-deployment
spec:
selector:
matchLabels:
app: metrics
department: sales
replicas: 3
template:
metadata:
labels:
app: metrics
department: sales
spec:
containers:
- name: hello
image: "gcr.io/google-samples/hello-app:2.0"
Copy the manifest to a file named my-deployment.yaml, and create the
Deployment:
kubectl apply --kubeconfig [USER_CLUSTER_KUBECONFIG] -f my-deployment.yaml
where [USER_CLUSTER_KUBECONFIG] is the path of the kubeconfig file for your user cluster.
Get basic information about your Deployment:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get deployment my-deployment
The output shows that the Deployment has three Pods that are all available:
NAME READY UP-TO-DATE AVAILABLE AGE my-deployment 3/3 3 3 27s
List the Pods in your Deployment:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get pods
The output shows that your Deployment has three running Pods:
NAME READY STATUS RESTARTS AGE my-deployment-54944c8d55-4srm2 1/1 Running 0 6s my-deployment-54944c8d55-7z5nn 1/1 Running 0 6s my-deployment-54944c8d55-j62n9 1/1 Running 0 6s
Get detailed information about your Deployment:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get deployment my-deployment --output yaml
The output shows details about the Deployment spec and status:
kind: Deployment
metadata:
...
generation: 1
name: my-deployment
namespace: default
...
spec:
...
replicas: 3
revisionHistoryLimit: 10
selector:
matchLabels:
app: metrics
department: sales
...
spec:
containers:
- image: gcr.io/google-samples/hello-app:2.0
imagePullPolicy: IfNotPresent
name: hello
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
status:
availableReplicas: 3
conditions:
- lastTransitionTime: "2019-11-11T18:44:02Z"
lastUpdateTime: "2019-11-11T18:44:02Z"
message: Deployment has minimum availability.
reason: MinimumReplicasAvailable
status: "True"
type: Available
- lastTransitionTime: "2019-11-11T18:43:58Z"
lastUpdateTime: "2019-11-11T18:44:02Z"
message: ReplicaSet "my-deployment-54944c8d55" has successfully progressed.
reason: NewReplicaSetAvailable
status: "True"
type: Progressing
observedGeneration: 1
readyReplicas: 3
replicas: 3
updatedReplicas: 3
Describe your Deployment:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] describe deployment my-deployment
The output shows nicely formatted details about the Deployment, including the associated ReplicaSet:
Name: my-deployment Namespace: default CreationTimestamp: Mon, 11 Nov 2019 10:43:58 -0800 Labels:... Selector: app=metrics,department=sales Replicas: 3 desired | 3 updated | 3 total | 3 available | 0 unavailable StrategyType: RollingUpdate MinReadySeconds: 0 RollingUpdateStrategy: 25% max unavailable, 25% max surge Pod Template: Labels: app=metrics department=sales Containers: hello: Image: gcr.io/google-samples/hello-app:2.0 Port: Host Port: Environment: Mounts: Volumes: Conditions: Type Status Reason ---- ------ ------ Available True MinimumReplicasAvailable Progressing True NewReplicaSetAvailable OldReplicaSets: NewReplicaSet: my-deployment-54944c8d55 (3/3 replicas created)
Creating a Service of type LoadBalancer
One way to expose your Deployment to clients outside your cluster is to create
a Kubernetes Service of type
LoadBalancer.
Here's a manifest for a Service of type LoadBalancer:
apiVersion: v1
kind: Service
metadata:
name: my-service
spec:
selector:
app: metrics
department: sales
type: LoadBalancer
loadBalancerIP: [SERVICE_IP_ADDRESS]
ports:
- port: 80
targetPort: 8080
For the purpose of this exercise, these are the important things to understand about the Service:
Any Pod that has the label
app: metricsand the labeldepartment: salesis a member of the Service. Note that the Pods inmy-deploymenthave these labels.When a client sends a request to the Service on TCP port 80, the request is forwarded to a member Pod on TCP port 8080.
Every member Pod must have a container that is listening on TCP port 8080.
It happens that by default, the hello-app container listens on TCP port
8080. You can see this by looking at the
Dockerfile and the source code
for the app.
Replace [SERVICE_IP_ADDRESS] with an address that you own that is not already in use. For example, you could set this to a public IP address that your company owns. Or you could set it to a private address in your company network.
The address you choose must be routable from the location of any client that sends requests to the Service. For example, if you choose a private address, then external clients will not be able to send requests to the Service.
Save the manifest to a file named my-service.yaml, and create the Service:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] apply -f my-service.yaml
where [USER_CLUSTER_KUBECONFIG] is the path of your user cluster's kubeconfig file.
When you create the Service, GKE on-prem automatically configures the
loadBalancerIP address on your F5 BIG-IP load balancer.
View your Service:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get service my-service --output yaml
The output is similar to this:
apiVersion: v1
kind: Service
metadata:
...
name: my-service
namespace: default
...
spec:
clusterIP: 10.107.84.202
externalTrafficPolicy: Cluster
loadBalancerIP: 21.0.133.48
ports:
- nodePort: 31919
port: 80
protocol: TCP
targetPort: 8080
selector:
app: metrics
department: sales
sessionAffinity: None
type: LoadBalancer
status:
loadBalancer:
ingress:
- ip: 203.0.113.1
In the preceding output, you can see that your Service has a clusterIP, and
a loadBalancerIP. It also has a nodePort, a port, and a targetPort.
The clusterIP is not relevant to this exercise. The loadBalancerIP is the
IP address that you provided in my-service.yaml.
As an example, take the values shown in the preceding output. That is, suppose
your Service has loadBalancerIP = 203.0.113.1, port = 80,
nodePort = 31919, and targetPort = 8080.
A client sends a request to 203.0.113.1 on TCP port 80. The request gets routed to your F5 BIG-IP load balancer. The load balancer chooses one of your user cluster nodes, and forwards the request to [NODE_ADDRESS] on TCP port 31919. The iptables rules on the node forward the request to a member Pod on TCP port 8080.
Call your Service:
curl [SERVICE_IP_ADDRESS]
where [SERVICE_IP_ADDRESS] is the address that you specified for
loadBalancerIP.
The output shows a Hello, world! message:
curl 21.0.133.48 Hello, world! Version: 2.0.0 Hostname: my-deployment-dbd86c8c4-9wpbv
Deleting your Service
Delete your Service:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] delete service my-service
Verify that your Service has been deleted:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get services
The output no longer shows my-service.
Deleting your Deployment
Delete your Deployment:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] delete deployment my-deployment
Verify that your Deployment has been deleted:
kubectl --kubeconfig [USER_CLUSTER_KUBECONFIG] get deployments
The output no longer shows my-deployment.