2019 Google Study Jam - 두번째이야기 Hello Node Kubernetes

Created in January 10, 2019

2019   ·   Kubernetes

오늘은 두번째 스터디를 진행했습니다.

주제는 Hello Node Kubernetes

주제에서 알 수 있듯이 kubenetes의 기본을 살펴보는 시간이었습니다.

kubenetes의 pod, service, deployments 의 개념을 명령어를 직업 쳐보면서 배우고 Kubernetes cluster, up scaling 작업을 통해 clustering의 동작 방식을 알게되었습니다.

마지막으로 쿠버네티스가 잘 동작하는지 시각적으로 보여주는 Dashboard설정법을 배웁니다.

기존에 kubernetes를 개념적으로 알고 있었지만 이렇게 실습해보니 더욱 쉽게 이해가 되고 좋았습니다.

그럼 오늘의 스터디를 시작해보겠습니다!!

오늘 스터디를 진행하면 아래의 아키텍처를 만나실수 있습니다. 오늘의-스터디-아키텍처

오늘 진행할 사항

  • Create a Node.js server.
  • Create a Docker container image.
  • Create a container cluster
  • Create a Kubernetes pod.
  • Scale up your services.

위 5가지를 수행하며 쿠버네티스 스터디를 진행했습니다.

리눅스 기반으로 스터디가 진행되며 vim, emacs, nano와 같은 리눅스기반 편집기를 사용할 줄 알아야합니다.

저는 vim을 예전부터 사용해왔기 때문에 vim을 사용했습니다.

Create a Node.js server

쿠버네티스내에서 동작할 Nodejs server를 제작합니다.

$ vi server.js

var http = require("http");
var handleRequest = function (request, response) {
  response.writeHead(200);
  response.end("Hello World!");
};
var www = http.createServer(handleRequest);
www.listen(8080);

위와같이 파일을 만듭니다. 위파일을 실행한 NodeJS 서버는은 8080포트로 접근가능하며 “Hello World!” 를 리턴합니다.

Create a Docker container image.

쿠버네티스내 continer에서 동작할 Docker Image 를 만들어봅니다.

$ vi Dockerfile

FROM node:6.9.2     # Nodejs V6.9.2 이미지를 불러온다.
EXPOSE 8080         # 외부와 8080 포트를 연결한다.
COPY server.js .    # Build 위에 있는 server.js를 복사해온다.
CMD node server.js  # 컨테이너 실행시 node server.js를 실행한다.

# Dockerfile을 Image로 빌드합니다.
# image 이름은 gcr.io/PROJECT_ID/hello-node 입니다.
# 이미지에 v1 태그를 등록합니다.
$ docker build -t gcr.io/PROJECT_ID/hello-node:v1 .
Sending build context to Docker daemon  20.99kB
Step 1/4 : FROM node:6.9.2
6.9.2: Pulling from library/node
75a822cd7888: Pull complete
57de64c72267: Pull complete
4306be1e8943: Pull complete
871436ab7225: Pull complete
0110c26a367a: Pull complete
1f04fe713f1b: Pull complete
ac7c0b5fb553: Pull complete
Digest: sha256:2e95be60faf429d6c97d928c762cb36f1940f4456ce4bd33fbdc34de94a5e043
Status: Downloaded newer image for node:6.9.2
 ---> faaadb4aaf9b
Step 2/4 : EXPOSE 8080
 ---> Running in 3b9b5a4618dc
Removing intermediate container 3b9b5a4618dc
 ---> 46752700727f
Step 3/4 : COPY server.js .
 ---> 7a9afff80010
Step 4/4 : CMD node server.js
 ---> Running in 61bdb8747991
Removing intermediate container 61bdb8747991
 ---> 4e86e4736626
Successfully built 4e86e4736626
Successfully tagged gcr.io/qwiklabs-gcp-a22c2abe0529392c/hello-node:v1

build된 이미지를 실행해보고 결과를 확인해봅니다.

# -d : 백그라운드 모드
# -p : 포트 포워딩, 컨테이너의 8080 포트를 외부 8080와 연결
$ docker run -d -p 8080:8080 gcr.io/PROJECT_ID/hello-node:v1
fa0d785f94cb9eaf9fa73b0211057fe482e6c8f0a2265e503a904285c9a8e3c5

# GET 방식으로 요청합니다.
$ curl -XGET localhost:8080
Hello World!

# 실행중인 컨테이너를 확인합니다.
$ docker ps -a
CONTAINER ID        IMAGE                             COMMAND
fa0d785f94cb        gcr.io/PROJECT_ID/hello-node:v1   "/bin/sh -c…"

# 컨테이너를 중지합니다.
$ docker stop fa0d785f94cb
fa0d785f94cb # 중지된 컨테이너 ID를 반환합니다.

생성된 이미지를 Kubernetes에서 사용하기 위해 GCP registry에 저장해 봅니다.

# 생성된 이미지를 Google Container Registry에 push 합니다.
# 해당 이미지는 gcp web console -> Container Registry에서 확인가능합니다.
$ gcloud docker -- push gcr.io/PROJECT_ID/hello-node:v1
WARNING: `gcloud docker` will not be supported for Docker client versions above 18.03.

As an alternative, use `gcloud auth configure-docker` to configure `docker` to
use `gcloud` as a credential helper, then use `docker` as you would for non-GCR
registries, e.g. `docker pull gcr.io/project-id/my-image`. Add
`--verbosity=error` to silence this warning: `gcloud docker
--verbosity=error -- pull gcr.io/project-id/my-image`.

See: https://cloud.google.com/container-registry/docs/support/deprecation-notices#gcloud-docker

The push refers to repository [gcr.io/qwiklabs-gcp-a22c2abe0529392c/hello-node]
026f0f5e78a3: Pushed
381c97ba7dc3: Pushed
604c78617f34: Pushed
fa18e5ffd316: Pushed
0a5e2b2ddeaa: Pushed
53c779688d06: Pushed
60a0858edcd5: Pushed
b6ca02dfe5e6: Pushed
v1: digest: sha256:af6ecd3b8087d1fc49c34ff02910a1a63131e74510922e0c3353e930249d7999 size: 2002

GCP Registry에 저장된 Docker Image를 확인해봅니다. Container-Registry-Position Container-Registry-List

Create a container cluster

쿠버네티스 내의 컨테이너 클러스터를 생성합니다.

# gcloud 명령어를 사용하기 위해 자신의 PROJECT_ID를 세팅합니다.
$ gcloud config set project PROJECT_ID
Updated property [core/project].

# gcloud명령어로 클러스터를 생성합니다.
# 클러스터명 : hello-world
# 노드수 : 2
# 머신유형: n1-standard-1
# 클러스터 위치: us-central1-a
$ gcloud container clusters create hello-world \
                --num-nodes 2 \
                --machine-type n1-standard-1 \
                --zone us-central1-a
Creating cluster hello-world in us-central1-a...done.
Created [https://container.googleapis.com/v1/projects/qwiklabs-gcp-a22c2abe0529392c/zones/us-central1-a/clusters/hello-wor
ld].
To inspect the contents of your cluster, go to: https://console.cloud.google.com/kubernetes/workload_/gcloud/us-central1-a
/hello-world?project=qwiklabs-gcp-a22c2abe0529392c
kubeconfig entry generated for hello-world.
NAME         LOCATION       MASTER_VERSION  MASTER_IP       MACHINE_TYPE   NODE_VERSION  NUM_NODES  STATUS
hello-world  us-central1-a  1.10.9-gke.5    35.188.172.182  n1-standard-1  1.10.9-gke.5  2          RUNNING

생성된 클러스터를 확인합니다. Created-Container

Create your pod

컨테이너 그룹인 포드를 생성하고 확인해봅니다.

pod를 실행하고 실행상태를 확인해봅니다.

$ kubectl run hello-node \
    --image=gcr.io/PROJECT_ID/hello-node:v1 \
    --port=8080
deployment.apps "hello-node" created

$ kubectl get deployments
NAME         DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
hello-node   1         1         1            1           40s

$ kubectl get pods
NAME                          READY     STATUS    RESTARTS   AGE
hello-node-7db8fd78df-jz4pv   1/1       Running   0          1m

$ kubectl cluster-info
Kubernetes master is running at https://35.188.172.182
GLBCDefaultBackend is running at https://35.188.172.182/api/v1/namespaces/kube-system/services/default-http-backend:http/proxy
Heapster is running at https://35.188.172.182/api/v1/namespaces/kube-system/services/heapster/proxy
KubeDNS is running at https://35.188.172.182/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy
Metrics-server is running at https://35.188.172.182/api/v1/namespaces/kube-system/services/https:metrics-server:/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.

$ kubectl config view
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: REDACTED
    server: https://35.188.172.182
  name: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
contexts:
- context:
    cluster: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
    user: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
  name: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
current-context: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
kind: Config
preferences: {}
users:
- name: gke_qwiklabs-gcp-a22c2abe0529392c_us-central1-a_hello-world
  user:
    auth-provider:
      config:
        access-token: ya29.GqMBjQZLvWCTZr0MQ4s5TCQK3NhaZFbpRqsMLG_11fyEs62wVbKfb4fnfFwA9hjT80eRUleTtwKH-lIhsIext2ewa9XPvFkno5eSJGQ5VOramQNB
JeiROmuAczSL5Tbgs4MviP_hva4X6CERtpKrOG4hjqzykjes2wYBVYibz8EKadcj1djlhsUI1DqenSQ5FjSRSsj4-5cAvAnsp1ZeLLCKeRmLKw
        cmd-args: config config-helper --format=json
        cmd-path: /google/google-cloud-sdk/bin/gcloud
        expiry: 2019-01-10T03:30:04Z
        expiry-key: '{.credential.token_expiry}'
        token-key: '{.credential.access_token}'
      name: gcp

$ kubectl get events

$ kubectl logs hello-node-7db8fd78df-jz4pv

Allow external traffic

외부 네트워크와의 연결을 허용하며 IP를 부여받습니다.

$ kubectl expose deployment hello-node --type="LoadBalancer"
service "hello-node" exposed

$ kubectl get services
NAME         TYPE           CLUSTER-IP     EXTERNAL-IP      PORT(S)          AGE
hello-node   LoadBalancer   10.31.241.99   35.193.233.235   8080:32509/TCP   1m
kubernetes   ClusterIP      10.31.240.1    <none>           443/TCP          11m

Extenal-Ip

Scale up your service

배포된 pod를 scale up 하여 수를 증가시켜봅니다.

$ kubectl scale deployment hello-node --replicas=4
deployment.extensions "hello-node" scaled
$ kubectl get deployment
NAME         DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
hello-node   4         4         4            4           9m
$ kubectl get pods
NAME                          READY     STATUS    RESTARTS   AGE
hello-node-7db8fd78df-6vn6h   1/1       Running   0          57s
hello-node-7db8fd78df-jz4pv   1/1       Running   0          10m
hello-node-7db8fd78df-r245c   1/1       Running   0          57s
hello-node-7db8fd78df-zdc9v   1/1       Running   0          57s

Roll out an upgrade to your service

기존 배포된 nodejs를 버전업하여 배포해봅니다.

$ vi server.js

var http = require("http");
var handleRequest = function (request, response) {
  response.writeHead(200);
  // response.end("Hello World!");
  response.end("Hello Kubernetes World!");
};
var www = http.createServer(handleRequest);
www.listen(8080);

$ docker build -t gcr.io/PROJECT_ID/hello-node:v2 .
Sending build context to Docker daemon  2.397MB
Step 1/4 : FROM node:6.9.2
 ---> faaadb4aaf9b
Step 2/4 : EXPOSE 8080
 ---> Using cache
 ---> 46752700727f
Step 3/4 : COPY server.js .
 ---> 82af2e80dcaf
Step 4/4 : CMD node server.js
 ---> Running in e54605d268f2
Removing intermediate container e54605d268f2
 ---> 71124401746d
Successfully built 71124401746d
Successfully tagged gcr.io/qwiklabs-gcp-a22c2abe0529392c/hello-node:v2

$ gcloud docker -- push gcr.io/PROJECT_ID/hello-node:v2
The push refers to repository [gcr.io/qwiklabs-gcp-a22c2abe0529392c/hello-node]
d7dd5a8d4a66: Pushed
381c97ba7dc3: Layer already exists
604c78617f34: Layer already exists
fa18e5ffd316: Layer already exists
0a5e2b2ddeaa: Layer already exists
53c779688d06: Layer already exists
60a0858edcd5: Layer already exists
b6ca02dfe5e6: Layer already exists
v2: digest: sha256:c2d1066f18f99955c9ab57557ae1aa2210cdfbed2ec55cfceb86b36bfbb378d8 size: 2002

$ kubectl edit deployment hello-node

spec.template.spec.containers.image 위치에 있는 gcr.io/PROJECT_ID/hello-node:v1 -> gcr.io/PROJECT_ID/hello-node:v2로 바꿈

# 저장시 아래와 같은 메세지를 확인
deployment.extensions "hello-node" edited

$ kubectl get deployments
NAME         DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
hello-node   4         4         4            4           19m

$ kubectl get services
NAME         TYPE           CLUSTER-IP     EXTERNAL-IP      PORT(S)          AGE
hello-node   LoadBalancer   10.31.241.99   35.193.233.235   8080:32509/TCP   16m
kubernetes   ClusterIP      10.31.240.1    <none>           443/TCP          25m

새로운 메세지로 바뀐것을 확인할 수있습니다. Extenal-Ip-Review

Kubernetes graphical dashboard (optional)

대쉬보드 설정법을 알아봅니다.

$ kubectl create clusterrolebinding cluster-admin-binding --clusterrole=cluster-admin --user=$(gcloud config get-value account)
Your active configuration is: [cloudshell-2007]
clusterrolebinding.rbac.authorization.k8s.io "cluster-admin-binding" created

$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v1.10.1/src/deploy/recommended/kubernetes-dashboard.yaml
secret "kubernetes-dashboard-certs" created
serviceaccount "kubernetes-dashboard" created
role.rbac.authorization.k8s.io "kubernetes-dashboard-minimal" created
rolebinding.rbac.authorization.k8s.io "kubernetes-dashboard-minimal" created
deployment.apps "kubernetes-dashboard" created
service "kubernetes-dashboard" created

$ kubectl -n kube-system edit service kubernetes-dashboard
Edit cancelled, no changes made.

$ kubectl -n kube-system describe $(kubectl -n kube-system \
get secret -n kube-system -o name | grep namespace) | grep token:
token:      eyJhbGciOiJSUzI1NiIsImtpZCI6IiJ9.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.U7IGwSM66LwfjJQYVdvHPzK0DcK4t6wXGdvw55Vlq_SVFTeo7Uq1uYTDeQ-dvwX5JHsvYKrRMOwNfGB1PLYknAo_62PdzBwuZ7Nf7zMsOqJ1A25HC7JZQ6H2VOESfHBWrHYI-M6O3F7cagoTqFT8HAkGh_YRG6X_oY0OjMA5f5FyVr07AERtqXGqsx9w6SYi2rLQ9NcO2Y0cSWRBBIRE4MxsX35gldzjd1YrhE8cHkUvtj8HfsHKMwl6F39CtDVQAec4ZuOJr_Ols8CVxxFkcJdIHDvWW81TkdixFIza7RpKddGCiwyndHOnBP-LEwTPrPz5z8vdWGOCdDa2G97qtw

$ kubectl proxy --port 8081
Starting to serve on 127.0.0.1:8081

위와 같이 실행하면 대쉬보드가 실행됩니다.

기본 링크에 /api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/#!/cronjob?namespace=default를 이어 붙이면 대쉬보드로 이동가능합니다.

위에서 얻은 토큰으로 로그인을 시도하면 대쉬보드로 이동한다. Extenal-Ip-Review

Extenal-Ip-Review



Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • 백엔드 개발자가 리액트 네이티브로 달력 앱을 배포해보았습니다.
  • MySQL 실시간 쿼리 확인하기
  • MySQL 트랜잭션 격리수준 Isolation level 알아보기
  • 레디스에 대한 간단한 설명과 성능향상시키기
  • Apple login 사용해보기