LB > Node Port > Cluster IP
Node Port를 만들면 Cluster IP가 자동으로 만들어진다
*SVC 지우기
1 2 3 4 5 6 7 8 9 | [root@host01-4 hk]# k get service NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE headless-svc ClusterIP 10.96.102.169 <none> 80/TCP 27m kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 48m test-svc ClusterIP 10.96.25.31 <none> 80/TCP 45m [root@host01-4 hk]# k delete svc test-svc service "test-svc" deleted [root@host01-4 hk]# | cs |
*NodePort 생성하기(생성뒤에 해당 Port로 curl을 날릴 수 있다) :
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 | vi NodePort.yaml apiVersion: v1 kind: Service metadata: name: test-svc spec: type: NodePort selector: type: test ports: - port: 80 targetPort: 8080 [root@host01-4 hk]# k create -f nodePort.yaml service "test-svc" created [root@host01-4 hk]# [root@host01-4 hk]# k get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE headless-svc ClusterIP 10.96.102.169 <none> 80/TCP 31m kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 52m test-svc NodePort 10.107.213.48 <none> 80:32045/TCP 6s #NodePort에서 80포트는 cluster IP port이고 32045는 Node Port | cs |
*Load Balancer 생성
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 | k edit svc test-app # Please edit the object below. Lines beginning with a '#' will be ignored, # and an empty file will abort the edit. If an error occurs while saving this file will be # reopened with the relevant failures. # apiVersion: v1 kind: Service metadata: creationTimestamp: 2018-05-25T07:44:57Z name: test-svc namespace: default resourceVersion: "29844" selfLink: /api/v1/namespaces/default/services/test-svc uid: 85b2c6ea-5fef-11e8-8e09-005056b28b62 spec: clusterIP: 10.107.213.48 externalTrafficPolicy: Cluster ports: - nodePort: 32045 port: 80 protocol: TCP targetPort: 8080 selector: type: test sessionAffinity: None type: LoadBalancer #LoadBalancer로 수정해주면... status: loadBalancer: {} [root@host01-4 hk]# k get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE headless-svc ClusterIP 10.96.102.169 <none> 80/TCP 38m kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 59m test-svc LoadBalancer 10.107.213.48 <pending> 80:32045/TCP 7m #Node Port에 Cluster IP까지 자동 [root@host01-4 hk]# | cs |
상위 네트워크 모듈을 생성하면 하위 매핑 정보는 알아서 구성하게 됨
*TargetPort 설정 이유
1 2 3 4 5 6 7 | ports: - nodePort: 32045 targetPort: 8080 | cs |
=> Cluster의 port와 pod Port 정보가 다르기 때문
*비즈니스 관점에서 Probe가 정상적으로 동작하는지 여부를 체크할 수 있어야 한다....
Client가 접속한다!!
Svc > Pod(X)
정상동작 X 다고 생각하게 된다.
이때 readinessProbe를 활용하면 cluster 정상동작 여부를 체크할 수 있다 :
1 2 3 4 5 6 7 8 9 | containers: - image: reg.cloud.com/kubia name: kubia readinessProbe: exec: command: - ls - /var/ready | cs |
*ReadinessProbe 생성 예제 :
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 | #vi rc-probe.yaml # /var/ready가 pod 안에 있으면 probe에서 통과 시킨다(Ready 상태로 만들어줌) apiVersion: v1 kind: ReplicationController metadata: name: rc-readiness spec: replicas: 3 template: metadata: labels: type: test spec: containers: - image: reg.cloud.com/kubia name: kubia readinessProbe: exec: command: - ls - /var/ready ports: - containerPort: 8080 protocol: TCP #vi service.yaml apiVersion: v1 kind: Service metadata: name: test-svc spec: selector: type: test ports: - port: 80 targetPort: 8080 [root@host01-4 hk]# k create -f rc-probe.yaml replicationcontroller "rc-readiness" created [root@host01-4 hk]# k get po -o wide NAME READY STATUS RESTARTS AGE IP NODE rc-readiness-22hxs 0/1 Running 0 10s 10.36.0.2 host01-3.cloud.com rc-readiness-2pt29 0/1 Running 0 10s 10.44.0.1 host01-2.cloud.com rc-readiness-h74t2 0/1 Running 0 10s 10.36.0.1 host01-3.cloud.com #Running 상태이지만 Ready가 아무것도 없다 [root@host01-4 hk]# k create -f service.yaml service "test-svc" created [root@host01-4 hk]# k get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 6m test-svc ClusterIP 10.98.99.22 <none> 80/TCP 4s [root@host01-4 hk]# k get ep NAME ENDPOINTS AGE kubernetes 10.10.12.14:6443 6m test-svc 9s #Endpoint가 아무것도 안 떠있다. [root@host01-4 hk]# k exec rc-readiness-22hxs -- touch /var/ready #1번 pod에 /var/ready를 만들어준다 (Ready 상태로 만들어주기 위해) [root@host01-4 hk]# k get ep NAME ENDPOINTS AGE kubernetes 10.10.12.14:6443 8m test-svc 10.36.0.2:8080 1m #test-svc의 ENDPOINT가 enabled된다 [root@host01-4 hk]# k get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 8m test-svc ClusterIP 10.98.99.22 <none> 80/TCP 1m [root@host01-4 hk]# k get po -o wide NAME READY STATUS RESTARTS AGE IP NODE rc-readiness-22hxs 1/1 Running 0 4m 10.36.0.2 host01-3.cloud.com rc-readiness-2pt29 0/1 Running 0 4m 10.44.0.1 host01-2.cloud.com rc-readiness-h74t2 0/1 Running 0 4m 10.36.0.1 host01-3.cloud.com #1번 Pod가 Ready 상태로 바뀐다 [root@host01-4 hk]# k exec rc-readiness-2pt29 -- touch /var/ready [root@host01-4 hk]# k get po -o wide NAME READY STATUS RESTARTS AGE IP NODE rc-readiness-22hxs 1/1 Running 0 5m 10.36.0.2 host01-3.cloud.com rc-readiness-2pt29 1/1 Running 0 5m 10.44.0.1 host01-2.cloud.com rc-readiness-h74t2 0/1 Running 0 5m 10.36.0.1 host01-3.cloud.com #2번 Pod Ready 상태로 바뀐다 | cs |
'Docker(도커) Kubernetes' 카테고리의 다른 글
| Kubernetes Namespace 생성 및 리소스 사용 (0) | 2018.05.25 |
|---|---|
| Pod 개념, 멀티 컨테이너 생성, kubectl 명령어 사용법 (0) | 2018.05.25 |
| Kubernetes 설치 및 사용법, Yaml으로 pod 및 rc 생성 (0) | 2018.05.25 |
| 도커 관련 설정 파일 모음 (2) | 2018.05.24 |
| docker swarm 사용법 (0) | 2018.05.24 |