[K8s] 서비스(Service)

Aiden·2021년 7월 18일

DKOS k8s kubernetes

K8S Study

목록 보기

8/9

1. 서비스

1-1. 서비스 란?

파드에서 실행중인 애플리케이션을 네트워크 서비스로 노출시키는 방법

1-2. 서비스 종류

Cluster IP 타입

쿠버네티스 내부에서만 포드들에 접근할 때 사용한다. 외부로 포드를 노출하지 않기 때문에 쿠버네티스 클러스터 내부에서만 사용하는 포드에 적합하다.

NodePort 타입

파드에 접근할 수 있는 포트를 클러스터의 모든 노드에 동일하게 개방한다. 따라서, 외부에서 포드에 접근할 수 있는 서비스 타입이다.
접근할 수 있는 포트는 랜덤하게 정해지지만, 특정 포트로 접근하도록 설정할 수 있다.

LoadBalancer 타입

클라우드 플랫폼에서 제공하는 로드 밸런서를 동적으로 프로비저닝해 포드에 연결한다.
NodePort 타입과 마찬가지로 외부에서 포드에 접근할 수 있는 서비스 타입이다. 그렇지만 일반적으로 AWS, GCP 등과 같은 클라우드 플랫폼 환경에서만 사용할 수 있다.

2. Cluster IP Services

2-1. 파드 생성

YAML 템플릿

pod.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: deploy-cndk
spec:
  replicas: 3
  selector:
    matchLabels:
      app: deploy-websrv
  template:
    metadata:
      labels:
        app: deploy-websrv
    spec:
      containers:
      - name: cndk-websrv
        image: gcr.io/google-samples/kubernetes-bootcamp:v1
        ports:
        - containerPort: 8080

생성 및 확인

생성

# kubectl apply -f pod.yaml
deployment.apps/deploy-cndk created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 03:47:37 2021

NAME                               READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-cndk-6cbc8c75db-fjd4v   1/1     Running   0          4m27s   172.16.189.102   worker2   <none>           <none>
pod/deploy-cndk-6cbc8c75db-nssqs   1/1     Running   0          4m27s   172.16.182.38    worker3   <none>           <none>
pod/deploy-cndk-6cbc8c75db-zs2lg   1/1     Running   0          4m28s   172.16.235.136   worker1   <none>           <none>

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   12m   <none>

NAME                   ENDPOINTS            AGE
endpoints/kubernetes   192.168.1.211:6443   12m

2-2. 접속 테스트용 파드 배포(Master)

YAML 템플릿

podnet.yaml

nodeName 에는 마스터 노드의 이름을 입력

apiVersion: v1
kind: Pod
metadata:
  name: netshoot-pod
spec:
  **nodeName: master**
  containers:
  - name: netshoot-pod
    image: nicolaka/netshoot
    command: ["tail"]
    args: ["-f", "/dev/null"]

생성 및 확인

생성

# kubectl apply -f podnet.yaml
pod/netshoot-pod created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 03:48:42 2021

NAME                               READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-cndk-6cbc8c75db-fjd4v   1/1     Running   0          5m32s   172.16.189.102   worker2   <none>           <none>
pod/deploy-cndk-6cbc8c75db-nssqs   1/1     Running   0          5m32s   172.16.182.38    worker3   <none>           <none>
pod/deploy-cndk-6cbc8c75db-zs2lg   1/1     Running   0          5m33s   172.16.235.136   worker1   <none>           <none>
**pod/netshoot-pod                   1/1     Running   0          15s     172.16.219.70    master    <none>           <none>**

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   13m   <none>

NAME                   ENDPOINTS            AGE
endpoints/kubernetes   192.168.1.211:6443   13m

# kubectl exec -it netshoot-pod -- zsh
                    dP            dP                           dP   
                    88            88                           88   
88d888b. .d8888b. d8888P .d8888b. 88d888b. .d8888b. .d8888b. d8888P 
88'  `88 88ooood8   88   Y8ooooo. 88'  `88 88'  `88 88'  `88   88   
88    88 88.  ...   88         88 88    88 88.  .88 88.  .88   88   
dP    dP `88888P'   dP   `88888P' dP    dP `88888P' `88888P'   dP   
                                                                    
Welcome to Netshoot! (github.com/nicolaka/netshoot)

2-3. 서비스(Cluster IP 타입) 배포

YAML 템플릿

svc-clusterip.yaml

apiVersion: v1
kind: Service
metadata:
  name: svc-clusterip
spec:
  ports:
    - name: svc-webport
      **port: 9000
      targetPort: 8080**
  selector:
    app: deploy-websrv
  **type: ClusterIP**

생성 및 확인

생성

# kubectl apply -f svc-clusterip.yaml
service/svc-clusterip created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 03:57:19 2021

NAME                               READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-cndk-6cbc8c75db-fjd4v   1/1     Running   0          14m     172.16.189.102   worker2   <none>           <none>
pod/deploy-cndk-6cbc8c75db-nssqs   1/1     Running   0          14m     172.16.182.38    worker3   <none>           <none>
pod/deploy-cndk-6cbc8c75db-zs2lg   1/1     Running   0          14m     172.16.235.136   worker1   <none>           <none>
pod/netshoot-pod                   1/1     Running   0          8m52s   172.16.219.70    master    <none>           <none>

NAME                    TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE   SELECTOR
service/kubernetes      ClusterIP   10.96.0.1       <none>        443/TCP    22m   <none>
**service/svc-clusterip   ClusterIP   10.104.187.17   <none>        9000/TCP   29s   app=deploy-websrv**

NAME                      ENDPOINTS                                                    AGE
endpoints/kubernetes      192.168.1.211:6443                                           22m
**endpoints/svc-clusterip   172.16.182.38:8080,172.16.189.102:8080,172.16.235.136:8080   29s**

[root@master dkos-w05 (⎈ |kube:default)]# curl 10.104.187.17:9000
Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-nssqs | v=1
[root@master dkos-w05 (⎈ |kube:default)]# curl 10.104.187.17:9000
Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-fjd4v | v=1
[root@master dkos-w05 (⎈ |kube:default)]# curl 10.104.187.17:9000
Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-zs2lg | v=1
[root@master dkos-w05 (⎈ |kube:default)]# curl 10.104.187.17:9000
Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-zs2lg | v=1
[root@master dkos-w05 (⎈ |kube:default)]# curl 10.104.187.17:9000
Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-nssqs | v=1
[root@master dkos-w05 (⎈ |kube:default)]#

생성한 서비스(ClusterIP)를 통해 클러스터 내부에서 3개의 파드로 접속이 가능하다.
하지만, 클러스터 외부에서는 접속할 수 없다는 단점이 있다.
이를 극복하기 위해, NodePort 서비스 타입을 사용해본다.

삭제

# kubectl delete deploy,svc --all
deployment.apps "deploy-cndk" deleted
service "kubernetes" deleted
service "svc-clusterip" deleted

3. NodePort Services

3-1. 파드 생성

YAML 템플릿

nginx11.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: deploy-nginx11
spec:
  replicas: 3
  selector:
    matchLabels:
      app: deploy-nginx11
  template:
    metadata:
      labels:
        app: deploy-nginx11
    spec:
      containers:
      - name: deploy-nginx11
        image: nginx:1.11

생성 및 확인

생성

# kubectl apply -f nginx11.yaml
deployment.apps/deploy-nginx11 created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 04:27:14 2021

NAME                                  READY   STATUS    RESTARTS   AGE   IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-nginx11-656d5495b7-9rndv   1/1     Running   0          94s   172.16.235.137   worker1   <none>           <none>
pod/deploy-nginx11-656d5495b7-km865   1/1     Running   0          94s   172.16.189.103   worker2   <none>           <none>
pod/deploy-nginx11-656d5495b7-xw8sk   1/1     Running   0          94s   172.16.182.39    worker3   <none>           <none>

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   10m   <none>

NAME                   ENDPOINTS            AGE
endpoints/kubernetes   192.168.1.211:6443   10m

3-2. 접속 테스트용 파드 배포(Master)

위 실습에서 사용한 podnet.yaml 생성하거나, 삭제하지 않았다면 그대로 사용

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 04:27:53 2021

NAME                                  READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-nginx11-656d5495b7-9rndv   1/1     Running   0          2m13s   172.16.235.137   worker1   <none>           <none>
pod/deploy-nginx11-656d5495b7-km865   1/1     Running   0          2m13s   172.16.189.103   worker2   <none>           <none>
pod/deploy-nginx11-656d5495b7-xw8sk   1/1     Running   0          2m13s   172.16.182.39    worker3   <none>           <none>
**pod/netshoot-pod                      1/1     Running   0          31m     172.16.219.70    master    <none>           <none>**

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   11m   <none>

NAME                   ENDPOINTS            AGE
endpoints/kubernetes   192.168.1.211:6443   11m

3-3. NodePort Service 배포

YAML 템플릿

svc-nodeport.yaml

apiVersion: v1
kind: Service
metadata:
  name: svc-nodeport
spec:
  ports:
    - name: svc-webport
      port: 9000
      targetPort: 80
****        ports:
        - containerPort: 8080
  selector:
    app: deploy-nginx11
  **type: NodePort**

생성 및 확인

생성

# kubectl apply -f svc-nodeport.yaml
service/svc-nodeport created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 04:30:09 2021

NAME                                  READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-nginx11-656d5495b7-9rndv   1/1     Running   0          4m30s   172.16.235.137   worker1   <none>           <none>
pod/deploy-nginx11-656d5495b7-km865   1/1     Running   0          4m30s   172.16.189.103   worker2   <none>           <none>
pod/deploy-nginx11-656d5495b7-xw8sk   1/1     Running   0          4m30s   172.16.182.39    worker3   <none>           <none>
pod/netshoot-pod                      1/1     Running   0          41m     172.16.219.70    master    <none>           <none>

NAME                   TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)          AGE   SELECTOR
service/kubernetes     ClusterIP   10.96.0.1      <none>        443/TCP          13m   <none>
**service/svc-nodeport   NodePort    10.97.24.204   <none>        9000:30000/TCP   23s   app=deploy-nginx11**

NAME                     ENDPOINTS                                              AGE
endpoints/kubernetes     192.168.1.211:6443                                     13m
**endpoints/svc-nodeport   172.16.182.39:80,172.16.189.103:80,172.16.235.137:80   22s**

# curl localhost:30000
# curl 192.168.1.211:30000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
... 생략 ...

# curl 10.97.24.204:9000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
... 생략 ...

# Worker Node 1
# curl 192.168.1.212:30000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
... 생략 ...

# Worker Node 2
# curl 192.168.1.213:30000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
... 생략 ...

# Worker Node 3
# curl 192.168.1.214:30000
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
... 생략 ...

<서비스IP내부포트(9000)> 을 이용해 접속이 가능하다.
<마스터노드:외부포트(30000)> 를 이용해 접속이 가능하다.
뿐만 아니라, <워커노드:외부포트(30000)> 를 이용해 접속이 가능하다.

삭제

# kubectl delete deploy,svc --all
deployment.apps "deploy-nginx11" deleted
service "kubernetes" deleted
service "svc-nodeport" deleted

4. LoadBalancer Services

외부 클라이언트 접속 시, kube-proxy에 바로 접근하는것이 아니라, 그 앞단에 위치하여 Load Balancer에 접근한다.

Cloud 사업자의 경우, 각각 구현한 LoadBalancer가 존재하며, 온프레미스의 경우 MetalLB와 같은 LoadBalancer를 사용하여 구현한다.

현 실습 환경이 온프레미스이므로, MetalLB를 사용해 실습한다.

4-1. MetalLB 소개

온프렘 환경에서 로드밸런서 서비스를 구현해준다.
MetalLB 동작(memberlist)을 위한 2가지 모드 : Layer2 모드(IPv4-ARP, IPv6-NDP), BGP 모드
Layer2 모드 : 멤버(노드) 중 1노드가 리더로 서비스 IP 1개 인입을 받음(ARP 응답) → 단점(single-node bottlenecking) ⇒ 상단 라우터에서 ECMP 경로로 부하 분산 가능하다.
BGP 모드 : failover 시 좀 더 빠른 절체가 가능하지만, 결국 인입은 1곳에서 처리 → L2 에 상단 라우터 ECMP가 더 효율적!
설치 : Kubernetes manifests, using Kustomize, or using Helm

4-2. MetalLB 설치(Kubernetes manifests)

공식문서 : https://metallb.universe.tf/installation/

설치 및 확인

설치

# kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/namespace.yaml
namespace/metallb-system created
# kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.10.2/manifests/metallb.yaml
Warning: policy/v1beta1 PodSecurityPolicy is deprecated in v1.21+, unavailable in v1.25+
podsecuritypolicy.policy/controller created
podsecuritypolicy.policy/speaker created
serviceaccount/controller created
serviceaccount/speaker created
clusterrole.rbac.authorization.k8s.io/metallb-system:controller created
clusterrole.rbac.authorization.k8s.io/metallb-system:speaker created
role.rbac.authorization.k8s.io/config-watcher created
role.rbac.authorization.k8s.io/pod-lister created
role.rbac.authorization.k8s.io/controller created
clusterrolebinding.rbac.authorization.k8s.io/metallb-system:controller created
clusterrolebinding.rbac.authorization.k8s.io/metallb-system:speaker created
rolebinding.rbac.authorization.k8s.io/config-watcher created
rolebinding.rbac.authorization.k8s.io/pod-lister created
rolebinding.rbac.authorization.k8s.io/controller created
daemonset.apps/speaker created
deployment.apps/controller created

확인

# kubectl get namespaces
NAME              STATUS   AGE
default           Active   32d
ingress-nginx     Active   6d16h
kube-node-lease   Active   32d
kube-public       Active   32d
kube-system       Active   32d
**metallb-system    Active   44s**

# kubectl get all -n metallb-system
NAME                              READY   STATUS    RESTARTS   AGE
pod/controller-6b78bff7d9-9nl82   1/1     Running   0          46s
pod/speaker-68jzn                 1/1     Running   0          46s
pod/speaker-6xcvb                 1/1     Running   0          46s
pod/speaker-l9b8w                 1/1     Running   0          46s
pod/speaker-rpctf                 1/1     Running   0          46s

NAME                     DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR            AGE
daemonset.apps/speaker   4         4         4       4            4           kubernetes.io/os=linux   46s

NAME                         READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/controller   1/1     1            1           46s

NAME                                    DESIRED   CURRENT   READY   AGE
replicaset.apps/controller-6b78bff7d9   1         1         1       46s

설정

metallb-config.yaml

apiVersion: v1
kind: ConfigMap
metadata:
  namespace: metallb-system
  name: config
data:
  config: |
    address-pools:
    - name: default
      protocol: layer2
      addresses:
      - 192.168.1.221-192.168.1.229

configmap 설정

# kubectl apply -f metallb-config.yaml
configmap/config created

4-3. 파드 생성

위 실습에서 사용한 pod.yaml 템플릿을 그대로 사용

생성 및 확인

생성

# kubectl apply -f pod.yaml
deployment.apps/deploy-cndk created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 05:18:51 2021

NAME                               READY   STATUS    RESTARTS   AGE   IP               NODE      NOMINATED NODE   READINESS GATES
**pod/deploy-cndk-6cbc8c75db-cscvj   1/1     Running   0          8s    172.16.235.139   worker1   <none>           <none>
pod/deploy-cndk-6cbc8c75db-h5htx   1/1     Running   0          8s    172.16.182.40    worker3   <none>           <none>
pod/deploy-cndk-6cbc8c75db-jsbp7   1/1     Running   0          8s    172.16.189.104   worker2   <none>           <none>**
pod/netshoot-pod                   1/1     Running   0          90m   172.16.219.70    master    <none>           <none>

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   30m   <none>

NAME                   ENDPOINTS            AGE
endpoints/kubernetes   192.168.1.211:6443   30m

4-4. MetelLB Service 배포

YAML 템플릿

mtlb.yaml

apiVersion: v1
kind: Service
metadata:
  name: svc-mtlb-1
spec:
  ports:
    - name: svc-mtlb-webport
      port: 9000
      targetPort: 8080
      nodePort: 30001
  selector:
    app: deploy-websrv
  type: LoadBalancer
---
apiVersion: v1
kind: Service
metadata:
  name: svc-mtlb-2
spec:
  ports:
    - name: svc-mtlb-webport
      port: 9000
      targetPort: 8080
      nodePort: 30002
  selector:
    app: deploy-websrv
  type: LoadBalancer
---
apiVersion: v1
kind: Service
metadata:
  name: svc-mtlb-3
spec:
  ports:
    - name: svc-mtlb-webport
      port: 9000
      targetPort: 8080
      nodePort: 30003
  selector:
    app: deploy-websrv
  type: LoadBalancer

생성 및 확인

생성

# kubectl apply -f mtlb.yaml
service/svc-mtlb-1 created
service/svc-mtlb-2 created
service/svc-mtlb-3 created

확인

# watch -d 'kubectl get pods,svc,ep -o wide'
Every 2.0s: kubectl get pods,svc,ep -o wide                                                                                                          master: Sun Jul 18 05:32:34 2021

NAME                               READY   STATUS    RESTARTS   AGE     IP               NODE      NOMINATED NODE   READINESS GATES
pod/deploy-cndk-6cbc8c75db-h5htx   1/1     Running   0          13m     172.16.182.40    worker3   <none>           <none>
pod/deploy-cndk-6cbc8c75db-jsbp7   1/1     Running   0          13m     172.16.189.104   worker2   <none>           <none>
pod/deploy-cndk-6cbc8c75db-pbgzz   1/1     Running   0          3m26s   172.16.182.41    worker3   <none>           <none>
pod/netshoot-pod                   1/1     Running   0          104m    172.16.219.70    master    <none>           <none>

NAME                 TYPE           CLUSTER-IP       EXTERNAL-IP     PORT(S)          AGE   SELECTOR
service/kubernetes   ClusterIP      10.96.0.1        <none>          443/TCP          44m   <none>
**service/svc-mtlb-1   LoadBalancer   10.99.211.249    192.168.1.221   9000:30001/TCP   8s    app=deploy-websrv
service/svc-mtlb-2   LoadBalancer   10.104.210.238   192.168.1.222   9000:30002/TCP   8s    app=deploy-websrv
service/svc-mtlb-3   LoadBalancer   10.97.123.170    192.168.1.223   9000:30003/TCP   8s    app=deploy-websrv**

NAME                   ENDPOINTS                                                   AGE
endpoints/kubernetes   192.168.1.211:6443                                          44m
**endpoints/svc-mtlb-1   172.16.182.40:8080,172.16.182.41:8080,172.16.189.104:8080   8s
endpoints/svc-mtlb-2   172.16.182.40:8080,172.16.182.41:8080,172.16.189.104:8080   8s
endpoints/svc-mtlb-3   172.16.182.40:8080,172.16.182.41:8080,172.16.189.104:8080   8s**

# for i in {1..100}; do curl -s 192.168.1.221:9000 ; done | sort | uniq -c | sort -nr
     42 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
     34 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1
     24 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1
# for i in {1..100}; do curl -s 192.168.1.222:9000 ; done | sort | uniq -c | sort -nr
     36 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1
     35 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1
     29 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
# for i in {1..100}; do curl -s 192.168.1.223:9000 ; done | sort | uniq -c | sort -nr
     42 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
     32 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1
     26 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1

# for i in {1..100}; do curl -s 192.168.1.211:30001 ; done | sort | uniq -c | sort -nr
     38 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1
     34 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1
     28 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
# for i in {1..100}; do curl -s 192.168.1.211:30002 ; done | sort | uniq -c | sort -nr
     38 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1
     33 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
     29 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1
# for i in {1..100}; do curl -s 192.168.1.211:30003 ; done | sort | uniq -c | sort -nr
     35 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-h5htx | v=1
     34 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-jsbp7 | v=1
     31 Hello Kubernetes bootcamp! | Running on: deploy-cndk-6cbc8c75db-pbgzz | v=1

이번에는 NodePort와 다르게 metallb-config.yaml 에서 정의한 주소 범위로 서비스의 External-IP 가 생성되었다.

Aiden

기억이 안되면 기록이라도🐳

이전 포스트

[K8s] 컨트롤러(Controller)

다음 포스트

[K8s] 서비스(Service)

K8S Study

1. 서비스

1-1. 서비스 란?

1-2. 서비스 종류

Cluster IP 타입

NodePort 타입

LoadBalancer 타입

2. Cluster IP Services

2-1. 파드 생성

YAML 템플릿

생성 및 확인

2-2. 접속 테스트용 파드 배포(Master)

YAML 템플릿

생성 및 확인

2-3. 서비스(Cluster IP 타입) 배포

YAML 템플릿

생성 및 확인

3. NodePort Services

3-1. 파드 생성

YAML 템플릿

생성 및 확인

3-2. 접속 테스트용 파드 배포(Master)

3-3. NodePort Service 배포

YAML 템플릿

생성 및 확인

4. LoadBalancer Services

4-1. MetalLB 소개

4-2. MetalLB 설치(Kubernetes manifests)

설치 및 확인

설정

4-3. 파드 생성

생성 및 확인

4-4. MetelLB Service 배포

YAML 템플릿

생성 및 확인

[K8s] 컨트롤러(Controller)

[K8s] 인그레스(Ingress)

0개의 댓글