DEMO

Pre-demo steps

1. Install kubectl tool for your OS.

2. Install any Kubernetes solution you want. I'm going to use minikube as an example.

3. Enable ingress addon, it will allow you to proceed with ingress part successfully.

minikube addons enable ingress

DEMO 1

Let's deploy pod, rs, deployment demo

# if you want to deploy in specific namespace
# use -n <namespace_name> option
kubectl apply -f demo1/pod.yaml

# expected result
kubectl get pods
NAME    READY   STATUS    RESTARTS   AGE
web   1/1     Running   0          14s

# describe application
# po stands for pods
kubectl describe po web

# check application logs
kubectl logs -f web

# delete one pod, what's happened?
kubectl delete po web

# let's deploy replicaset
# replicaset is going to deploy N-pods with specified template
kubectl apply -f demo1/rs.yaml
kubectl get pods
NAME             READY   STATUS    RESTARTS   AGE
web-rs-88bpx   1/1     Running   0          71s
web-rs-gwhgs   1/1     Running   0          71s
web-rs-wzg4q   1/1     Running   0          71s

# nice, let's describe replicaset object
# rs stands for replicaset
kubectl describe rs web-rs

# pay attention to the following lines:
# we can see current state of our replicas
Annotations:  Replicas:  3 current / 3 desired
Pods Status:  3 Running / 0 Waiting / 0 Succeeded / 0 Failed

# let's delete one of the pods, replace XXXX to one of your pods
kubectl delete po web-rs-XXXXX
kubectl get po
NAME             READY   STATUS    RESTARTS   AGE
web-rs-gwhgs   1/1     Running   0          3m35s
web-rs-rwm5c   1/1     Running   0          32s
web-rs-wzg4q   1/1     Running   0          3m35s

# here we are, there is a new pod which replaced the one we deleted
# try to delete replicaset
kubectl delete rs web-rs
kubectl get pods
NAME             READY   STATUS        RESTARTS   AGE
web-rs-gwhgs   0/1     Terminating   0          4m19s
web-rs-wzg4q   0/1     Terminating   0          4m19s

# you can catch terminating state of pods ^
# the last resource is Deployment, deploy it by yourself as we've done before
NAME                     READY   STATUS    RESTARTS   AGE
web-75b6496574-846x9   1/1     Running   0          67s
web-75b6496574-8p7v7   1/1     Running   0          47s
web-75b6496574-bbk9s   1/1     Running   0          58s

# let's describe deployment!
# deploy stands for deployment
# looks familiar, isn't it?
kubectl describe deploy web
Replicas:               3 desired | 3 updated | 3 total | 3 available | 0 unavailable

# let's test how it will be deployed if we change variable to LOG_LEVEL=error
web-7b9887855f-pzhhb   0/1     PodInitializing   0          4s

# new pods should be created, but why only one?
# describe deployment again
RollingUpdateStrategy:  25% max unavailable, 25% max surge

# the default update strategy is Rolling. it will update by 25 % and terminate by 25%
# for us it will round up update count: 3 / 4 ~ 1, and termination round down: 3 / 4 ~ 0
Events:
  Type    Reason             Age    From                   Message
  ----    ------             ----   ----                   -------
  Normal  ScalingReplicaSet  3m10s  deployment-controller  Scaled up replica set web-75b6496574 to 3
  Normal  ScalingReplicaSet  70s    deployment-controller  Scaled up replica set web-7b9887855f to 1
  Normal  ScalingReplicaSet  64s    deployment-controller  Scaled down replica set web-75b6496574 to 2
  Normal  ScalingReplicaSet  64s    deployment-controller  Scaled up replica set web-7b9887855f to 2
  Normal  ScalingReplicaSet  58s    deployment-controller  Scaled down replica set web-75b6496574 to 1
  Normal  ScalingReplicaSet  57s    deployment-controller  Scaled up replica set web-7b9887855f to 3
  Normal  ScalingReplicaSet  45s    deployment-controller  Scaled down replica set web-75b6496574 to 0

DEMO 2

Let's deploy Service and Ingress and play around with it.

kubectl apply -f demo2/service.yaml

# get svc list; svc stands for Service
kubectl get svc
web        ClusterIP   10.99.190.198   <none>        8080/TCP   15s

# let's access our service from another container in the Kubernetes
# all the network tools are already installed see more: https://github.com/nicolaka/netshoot
kubectl run tmp-shell --rm -i --tty --image nicolaka/netshoot -- /bin/bash

Name:      web
Address 1: 10.99.190.198 web.default.svc.cluster.local

ping web
PING web.default.svc.cluster.local (10.99.190.198) 56(84) bytes of data.

# and from our PC:
# <LOCAL_PORT>:<PORT_IN_SERVICE>
kubectl port-forward svc/web 8000:8080

# navigate into other terminal window and perform
curl 127.0.0.1:8000
# or from browser window

# final part. Ingress!
kubectl apply -f demo2/ingress.yaml
kubectl get ingress
NAME    HOSTS   ADDRESS         PORTS   AGE
web   *       192.168.64.15   80      35s

# make curl into IP address below ADDRESS column with /web at the end
curl 192.168.64.15/web

# you should hit the web main page =)

### CONGRATULATIONS!