How to use kubernetes

qomarullah

4 min readOct 23, 2020

simple tutorial

This tutorial use kubernetes in digital ocean.

Create k8s with minimum resource 2 nodes : 2CPU, 4GB Memmory, it takes $20/month

Install kubectl

https://kubernetes.io/docs/tasks/tools/install-kubectl/

Configuration

download cluster config from DO and put inside ~/.kube

$ cd ~/.kube && kubectl --kubeconfig="k8s-01-kubeconfig.yaml" get nodes

if you have one config, you can copy k8s cluster config as default, so your command is only kubectl

$cp k8s-01-kubeconfig.yaml config
$kubectl get nodes

now we have two pool nodes ready.

Deployment

Before you go to deployment and dont know anything about docker, you can check this first tutorial.

Running Golang with local Docker and Kubernetes

Simple tutorial

medium.com

We can use this sample deployment config

$ kubectl apply -f deployment.yml
deployment.apps/go-hello-world created$ kubectl get deployments
NAME             READY   UP-TO-DATE   AVAILABLE   AGE 
go-hello-world   3/3     3            3           107s$ kubectl get pods
NAME                             READY   STATUS    RESTARTS   AGE  
go-hello-world-fbfc48747-g9d8g   1/1     Running   0          2m32s
go-hello-world-fbfc48747-jk6vb   1/1     Running   0          2m32s
go-hello-world-fbfc48747-tngx9   1/1     Running   0          2m32s

Service

After deployment pods, you need access from outside cluster using services.

Test port forward kube to one pod

$ kubectl port-forward go-hello-world-fbfc48747-g9d8g 8080:8080
Forwarding from 127.0.0.1:8080 -> 8080
Forwarding from [::1]:8080 -> 8080$ curl localhost:8080
Hello, Guest

Creating service act as proxy to multiple pods inside cluster

$ kubectl apply -f deployment.yml
deployment.apps/go-hello-world unchanged
service/go-hello-world-service created$ kubectl get services
NAME                     TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
go-hello-world-service   NodePort    10.245.139.86   <none>        9090:31447/TCP   6m56s
kubernetes               ClusterIP   10.245.0.1      <none>        443/TCP          27h

we can see there is cluster IP type nodeport 10.245.139.86:9090, this is still private IP so then we need to additional IP public for i, it can be exposed using LB

Ingress with LB

update services to use type LoadBalancer instead NodePort

After re-apply deployment with Load Balancer

$ kubectl apply -f deployment-with-lb.yml 
deployment.apps/go-hello-world unchanged
service/go-hello-world-service configured$ kubectl get service
NAME                     TYPE           CLUSTER-IP      EXTERNAL-IP       PORT(S)        AGE
go-hello-world-service   LoadBalancer   10.245.139.86   165.227.241.128   80:31365/TCP   91m
kubernetes               ClusterIP      10.245.0.1      <none>            443/TCP        28h

Now we have ip public load balancer, and we also have new bill for load balancer with two droplet. It takes arround $10/month

$kubectl describe service go-hello-world-service
Name:                     go-hello-world-service
Namespace:                default
Labels:                   <none>
Annotations:              kubernetes.digitalocean.com/load-balancer-id: de209ac6-ea0f-46ea-a13c-066d84d24425
Selector:                 app=go-hello-world
Type:                     LoadBalancer
IP:                       10.245.199.8
LoadBalancer Ingress:     165.227.241.128
Port:                     tcp  80/TCP
TargetPort:               8080/TCP
NodePort:                 tcp  30431/TCP
Endpoints:                10.244.0.172:8080,10.244.0.224:8080,10.244.0.44:8080
Session Affinity:         None
External Traffic Policy:  Cluster
Events:

Now we can curl to load balancer

curl http://165.227.241.128
Hello, Guest

Dashboard

$ kubectl cluster-info
Kubernetes master is running at https://30543e5d-3037-4c59-9fe8-922c70bf303a.k8s.ondigitalocean.com
CoreDNS is running at https://30543e5d-3037-4c59-9fe8-922c70bf303a.k8s.ondigitalocean.com/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy$ kubectl -n default describe deployments
Name:                   go-hello-world
Namespace:              default
CreationTimestamp:      Sat, 17 Oct 2020 21:47:50 +0700
Labels:                 <none>
Annotations:            deployment.kubernetes.io/revision: 1
Selector:               app=go-hello-world
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=go-hello-world
  Containers:
   go-hello-world:
    Image:        qomarullah/go-hello-world:1.0.0
    Port:         8080/TCP
    Host Port:    0/TCP
    Liveness:     http-get http://:8080/health delay=5s timeout=5s period=15s #success=1 #failure=3
    Readiness:    http-get http://:8080/readiness delay=5s timeout=1s period=10s #success=1 #failure=3
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   go-hello-world-fbfc48747 (3/3 replicas created)
Events:          <none>

We can check dashboard kubernetes via digitalocean dashboard link

How to Scale

kubectl get deployment
NAME             READY   UP-TO-DATE   AVAILABLE   AGE
go-hello-world   3/3     3            3           74m$ kubectl scale --replicas=4 deployment/go-hello-world
deployment.apps/go-hello-world scaled$ kubectl get deployment
NAME             READY   UP-TO-DATE   AVAILABLE   AGE
go-hello-world   4/4     4            4           76m

How to Delete

kubectl delete pod {pod} 
kubectl delete service {service}
kubectl delete deployment {deployment}