Deploy Todo App locally via Minikube

Minikube를 활용하여 로컬환경에 Todo App 배포

이번 실습은 쿠버네티스에 Python flask 어플리케이션과 mariaDB를 배포 해보고 접근 해보는 실습 입니다. 쿠버네티스에 DB용 영구볼륨(PersistentVolume)구성 방법 및 어플리케이션 배포 방법과 서비스 노출 방법에 대해 이해 할 수 있습니다.

사전 준비 사항

Minikube 구성

Minikube 구성: 관련 링크

Architecture

Star

Minikube에 Application 배포

Minikube 실행

Run minikube start to start minikube locally

Todo Data Volume 생성

Run kubectl apply -f todo-mariadb-pv.yaml to create persistent data volume for the mariadb.

todo-mariadb-pv.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: task-data-volume
  labels:
    classname: mini
spec:
  storageClassName: mini
  accessModes:
    - ReadWriteOnce
  capacity:
    storage: 1Gi
  hostPath:
    path: /data/task-data/
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: task-data-claim
  labels:
    classname: mini
spec:
  storageClassName: mini
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 500Mi

Run kubectl get pv,pvc -l classname=mini to verify the pv and pvc are created properly

$ kubectl get pv,pvc -l classname=mini
NAME                                CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM                     STORAGECLASS   REASON   AGE
persistentvolume/task-data-volume   1Gi        RWO            Retain           Bound    default/task-data-claim   mini                    11s

NAME                                    STATUS   VOLUME             CAPACITY   ACCESS MODES   STORAGECLASS   AGE
persistentvolumeclaim/task-data-claim   Bound    task-data-volume   1Gi        RWO            mini           11s

Todo App용 Mariadb 배포

Source code: cloudacode/coolstuff/todo-mariadb

Run kubectl apply -f todo-mariadb-app.yaml to deploy mariadb and service endpoint

todo-mariadb-app.yaml

---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: mysql
spec:
  selector:
    matchLabels:
      app: mysql
  strategy:
    type: Recreate
  template:
    metadata:
      labels:
        app: mysql
    spec:
      containers:
      - image: cloudacode/mariadb-todo:v1.0.1
        name: mysql
        ports:
        - containerPort: 3306
          name: mysql
        volumeMounts:
        - name: task-data-storage
          mountPath: /var/lib/mysql
      volumes:
      - name: task-data-storage
        persistentVolumeClaim:
          claimName: task-data-claim
---
apiVersion: v1
kind: Service
metadata:
  name: mysql
spec:
  ports:
  - port: 3306
  selector:
    app: mysql
  clusterIP: None

Todo App용 Flask 배포

Source code: cloudacode/coolstuff/todo-flask-app

Run kubectl apply -f todo-flask-app.yaml to deploy flask app and service endpoint

todo-flask-app.yaml

---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: todo-app
  labels:
    app: todo-app
spec:
  replicas: 1
  selector:
    matchLabels:
      app: todo-app
  strategy:
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 0
    type: RollingUpdate
  template:
    metadata:
      labels:
        app: todo-app
    spec:
      containers:
      - image: cloudacode/python-todo:v1.1.0
        imagePullPolicy: Always
        name: todo-app
        ports:
        - containerPort: 5000
          protocol: TCP
        env:
        - name: DB_USER
          value: root
        - name: DB_PASSWORD
          value: mysecret
        - name: DB_NAME
          value: todo
        - name: DB_HOST
          value: mysql
---
apiVersion: v1
kind: Service
metadata:
  name: todo-app-svc
spec:
  selector:
    app: todo-app
  ports:
   -  protocol: TCP
      port: 80
      targetPort: 5000
  type: LoadBalancer

Run kubectl get svc in check the exposed service. you may see the EXTERNAL-IP is pending status due to minikube doesn't have extra LB.

kubectl get svc todo-app-svc
NAME           TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
todo-app-svc   LoadBalancer   10.102.161.248   <pending>     80:31289/TCP   54s

You should use the minikube tunnel to get the external-ip for the service. Loadbalancer Access

Minikube tunnel 설정

Run minikube tunnel in a seperate terminal to enable external-ip

minikube tunnel
❗  The service todo-app-svc requires privileged ports to be exposed: [80]
🔑  sudo permission will be asked for it.
🏃  Starting tunnel for service todo-app-svc.
Password:

Run kubectl get svc todo-app-svc in check the external-ip and it will be

kubectl get svc todo-app-svc
NAME           TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
todo-app-svc   LoadBalancer   10.102.161.248   127.0.0.1     80:31289/TCP   5m41s

Todo App 접근

Open http://127.0.0.1 in your favorite web browser and check the flask todo app

Task Web

Todo MariaDB Pod 삭제 및 재배포

위에서 세팅한 Persistent Volume이 Pod의 내부 볼륨이 아닌 외부 스토리지 볼륨을 사용하는 것을 검증 하기 위해 pod를 삭제

Run kubectl delete pods -l app=todo-db to delete the pod resource. Deployment resource will monitor the pod status and will spawn the pod immediately

$ kubectl get pods -l app=todo-db
NAME                       READY   STATUS    RESTARTS   AGE
todo-db-65848c57db-hwqnp   1/1     Running   0          17s

A new todo-db pod is up and mounted node's /data/task-data/ to /var/lib/mysql/ volume on the pod. This means that DB data volume is managed in an external volume space.

확인을 위해 open http://127.0.0.1 and make sure the todo data exists.

Todo App 리소스 확장

Run kubectl scale --replicas=3 deployment/todo-app to scale out the app

$ kubectl scale --replicas=3 deployment/todo-app
deployment.apps/todo-app scaled
$ kubectl get pods
NAME                        READY   STATUS              RESTARTS   AGE
mysql-8bb5f69f8-cs55w       1/1     Running             0          10m
todo-app-576d68cf7f-cf74w   1/1     Running             0          5s
todo-app-576d68cf7f-mskbm   1/1     Running             0          7m16s
todo-app-576d68cf7f-w8x8q   0/1     ContainerCreating   0          5s

Clean Up

배포한 리소스(deployments, service) 정리

$ kubectl delete -f todo-flask-app.yaml
$ kubectl delete -f todo-mariadb-app.yaml

이 글이 유용하였다면 ⭐ Star를, 💬 1:1 질문이나 기술 관련 문의가 필요하신 분들은 클라우드어코드 카카오톡 채널 추가 부탁드립니다.🤗

Star