2025-07-17 AWS EKS

eks에서 입력
sudo su -

eksctl create cluster --name 4gl0717 --region ap-northeast-2 --node-type t2.small

앤서블 인스턴스

vi /etc/ansible/hosts

• [ansible]
  10.0.0.9 ansible_user=ansibleadmin ansible_ssh_pass=1234 ansible_become=yes ansible_become_method=sudo ansible_become_password=1234

• [k8s]
  10.0.4.100

eks 인스턴스

vi /etc/ssh/sshd_config

• 65행 yes (PasswordAuthentication yes)

systemctl restart sshd
useradd ansibleadmin
passwd ansibleadmin(1234)

visudo

• ansibleadmin ALL=(ALL) NOPASSWD: ALL

앤서블 인스턴스

sudo su ansibleadmin
ssh-keygen
ssh-copy-id ansibleadmin@10.0.4.100(eks 인스턴스의 프라이빗IP)
ansible all -m ping

eks 인스턴스

vi k8s_app.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: my4glapp
  labels:
    app: my4glapp
spec:
  replicas: 2
  selector:
    matchLabels:
      app: my4glapp
  template:
    metadata:
      labels:
        app: my4glapp
    spec:
      containers:
      - name: my4glapp
        image: sohyun2223/tomcatfinal
        imagePullPolicy: Always
        ports:
          - containerPort: 8080

  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 1

vi k8s_svc.yaml

apiVersion: v1
kind: Service
metadata:
  name: elbk8s
  labels:
    app: my4glapp
spec:
  selector:
    app: my4glapp
  ports:
    - port: 8080
      targetPort: 8080
  type: LoadBalancer

앤서블 인스턴스

cd /opt/devopsjob/
ansible-playbook 4glapp_build.yaml

cat 4glapp_build.yaml

---
- hosts: ansible

  tasks:

    - name: create docker images
      shell: docker build -t tomcatfinal .
      args:
        chdir: /opt/devopsjob

    - name: tag docker images
      shell: docker tag tomcatfinal sohyun2223/tomcatfinal

    - name: push docker images
      shell: docker push sohyun2223/tomcatfinal

eks 인스턴스

kubectl apply -f k8s_app.yaml
kubectl apply -f k8s_svc.yaml
watch kubectl get pod
kubectl get svc

pod는 상태가 running으로 잘 뜨는것을 확인

svc도 로드밸런서의 주소 확인 가능

AWS 페이지에서 로드밸런서에 생성된 내역을 확인 가능

EKS 인스턴스

확인했으니 삭제

kubectl delete -f k8s_app.yaml
kubectl delete -f k8s_svc.yaml

앤서블 인스턴스

앤서블에서 해당 매니페스트들을 실행할 플레이북 작성

sudo su - ansibleadmin
cd /opt/devopsjob
vi k8s_deploy.yaml

---
- hosts: k8s
  become: yes

  tasks:
    - name: deploy app on k8s
      shell: /root/bin/kubectl apply -f /root/k8s_app.yaml

    - name: deploy svc on k8s
      shell: /root/bin/kubectl apply -f /root/k8s_svc.yaml

젠킨스 퍼블릭ip:8080 접속

https://github.com/nohsohyun0128/4glmyapp

clean package

target/*.war
target
//opt/devopsjob
ansible-playbook /opt/devopsjob/4glapp_build.yaml

sudo vi k8s_deploy.yaml

---
- hosts: k8s
  become: yes

  tasks:
    - name: deploy app on k8s
      shell: /root/bin/kubectl apply -f /root/k8s_app.yaml

    - name: deploy svc on k8s
      shell: /root/bin/kubectl apply -f /root/k8s_svc.yaml

    - name: rollout update on k8s
      shell: /root/bin/kubectl rollout restart deployment.apps/my4glapp

마지막에는 뭘 하는지도 모르겠다 걍 놓침

eks 인스턴스

eksctl delete cluster --name 4gl0717 --region ap-northeast-2

이제 돈나가는거 다 끄자