K8S日常操作手册

kubectl 创建对象

• 创建资源 $ kubectl create -f ./my-manifest.yaml
• 使用多个文件创建资源 $ kubectl create -f ./my1.yaml -f ./my2.yaml
• 使用目录下的所有清单文件（yaml）来创建资源 $ kubectl create -f ./dir
• 使用url创建资源 $ kubectl create -f https://git.io/vPieo
• 启动一个nginx实例 $ kubectl run nginx --image=nginx
• 获取pod和svc的文档 $ kubectl explain pods

2. kubectl Scale 资源

• 扩展pod下容器数量

root@bx-florencetest-k8smaster1[ K8S-ADMIN ]:~   kubectl scale --replicas=0 -n florence-4qjxvbk2 deployment moddpap
deployment.extensions/moddpap scaled

root@bx-florencetest-k8smaster1[ K8S-ADMIN ]:~   kubectl scale --replicas=1 -n florence-4qjxvbk2 deployment moddpap
deployment.extensions/moddpap scaled

• 扩展名称为foo的资源到3个，是否使用rs取决于yaml中的编写

$ kubectl scale --replicas=3 rs/foo

例如yaml中kind: Deployment ，则应通过下面方法扩展

$ kubectl scale --replicas=3 deployment/foo

或者直接通过创建资源的yaml文件扩展

$ kubectl scale --replicas=3 -f foo.yaml

kubectl 与运行中的pod交互

• 查看pod日志， -f 持续查看

kubectl logs -n florence-4qjxvbk2 moddpap-844d4fd9d5-9n2wk

• 转发pod中的80端口到本地的5000端口

root@bx-florencetest-k8smaster1[ K8S-ADMIN ]:~  kubectl port-forward -n florence-4qjxvbk2 moddpap-844d4fd9d5-9n2wk 5000:80
Forwarding from 127.0.0.1:5000 -> 80
Forwarding from [::1]:5000 -> 80

• 在已存在的容器中执行命令

kubectl exec -n florence-4qjxvbk2 moddpap-844d4fd9d5-9n2wk -- ls /home/

• 获取namespace

[root@bx-florence-k8smaster1 ~]*  kubectl get ns

• 生成ingress配置

[root@bx-florence-k8smaster1 ~]*  kubectl get ingresses -n prod -oyaml > all-prod-ingress.yaml
[root@bx-florence-k8smaster1 ~]*  ll all-prod-ingress.yaml
-rw-r--r-- 1 root root 83 May 23 01:19 all-prod-ingress.yaml

• 获取线上pod list

[root@bx-florence-k8smaster1 ~]*  kubectl get pods -n florence-8mu2rlvw -o wide

• 进入指定pod的容器内

[root@bx-florence-k8smaster1 ~]*  kubectl exec -it florence-5cb7ffdf68-77gfb -c florence -n florence-8mu2rlvw bash

• 获取pod的分配的内部IP

[root@bx-florence-k8smaster1 ~]*  kubectl describe pod florence-5cb7ffdf68-77gfb -n florence-8mu2rlvw|grep IP
IP:                 10.254.58.48

• 查看k8S某个pod的信息

[root@bx-florence-k8smaster1 ~]*  kubectl get pods florence-5cb7ffdf68-77gfb  -n florence-8mu2rlvw -o wide
NAME                        READY   STATUS    RESTARTS   AGE     IP             NODE          NOMINATED NODE   READINESS GATES
florence-5cb7ffdf68-77gfb   1/1     Running   0          6h36m   10.254.58.48   10.11.22.52   <none>           <none>
 
[root@bx-florence-k8smaster1 ~]*  kubectl get pods florence-5cb7ffdf68-77gfb -n florence-8mu2rlvw
NAME                        READY   STATUS    RESTARTS   AGE
florence-5cb7ffdf68-77gfb   1/1     Running   0          6h58m

• 详细信息查询
• JSON格式显示Pod的完整信息

[root@bx-florence-k8smaster1 ~]*  kubectl get pod florence-5cb7ffdf68-77gfb -n florence-8mu2rlvw --output yaml

• YAML方式显示Pod的完整信息

[root@bx-florence-k8smaster1 ~]*  kubectl get pod  florence-5cb7ffdf68-77gfb -n florence-8mu2rlvw --output json

• 状态和生命周期查询

[root@bx-florence-k8smaster1 ~]*  kubectl describe pods florence-5cb7ffdf68-77gfb -n florence-8mu2rlvw
Name:               florence-5cb7ffdf68-77gfb
Namespace:          florence-8mu2rlvw
Priority:           0
PriorityClassName:  <none>
Node:               10.11.22.52/10.11.22.52
Start Time:         Fri, 22 May 2020 18:31:52 +0800
Labels:             app=florence
                    department=ihospital
                    group=ihospital
                    pod-template-hash=5cb7ffdf68
                    scene=main
Annotations:        podpreset.admission.kubernetes.io/podpreset-paas-env: 11479649
Status:             Running
IP:                 10.254.58.48
Controlled By:      ReplicaSet/florence-5cb7ffdf68
Containers:
  florence:
    Container ID:   docker://688475bdad4cb9fc293c71e1b0366843521d6986c38f6eb1cf6fadb9b69d52e2
    Image:          insurance-medicine-florence-backend/florence:feature_522-232652
    Image ID:       docker-pullable://insurance-medicine-florence-backend/florence@sha256:e78c88b9ae532b642ad6bc2226643c99bfddda7c96a6560e33956c2092e5ee69
    Port:           <none>
    Host Port:      <none>
    State:          Running
      Started:      Fri, 22 May 2020 18:32:05 +0800
    Ready:          True
    Restart Count:  0
    Limits:
      cpu:     3
      memory:  3200Mi
    Requests:
      cpu:      3
      memory:   3200Mi
    Liveness:   tcp-socket :80 delay=120s timeout=10s period=10s * success=1 * failure=1
    Readiness:  tcp-socket :80 delay=50s timeout=1s period=10s * success=1 * failure=1
    Environment:
      PAAS_CLUSTER:             florence
      PAAS_TARGET:              prod
      PAAS_CLUSTER_KUBERNETES:  1
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-gpbhp (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             True
  ContainersReady   True
  PodScheduled      True
Volumes:
  default-token-gpbhp:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-gpbhp
    Optional:    false
QoS Class:       Guaranteed
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:          <none>
 
* * * * * * * * * * 
[root@bx-florence-k8smaster1 ~]*  kubectl get pods -n kafka log-kafka-0
NAME          READY   STATUS    RESTARTS   AGE
log-kafka-0   2/2     Running   3          79m
 
[root@bx-florence-k8smaster1 ~]*  kubectl get pods -n kafka log-kafka-0 -o wide
NAME          READY   STATUS    RESTARTS   AGE   IP            NODE          NOMINATED NODE   READINESS GATES
log-kafka-0   2/2     Running   3          79m   10.254.5.59   10.11.22.41   <none>           <none>

• 查看POD的log

[root@bx-florence-k8smaster1 ~]*  kubectl logs -f -n kafka log-archiver-79bb645dbb-9pvpr

• 获取florence Deployment部署的服务

[root@bx-florence-k8smaster1 ~]*  kubectl get deployment -n florence-8mu2rlvw | grep florence
florence                      5/5     5            5           102d
florence-agent                1/1     1            1           104d
florence-cms                  1/1     1            1           104d
florence-consult-service      2/2     2            2           74d
florence-doctor               1/1     1            1           103d
florence-messager             3/3     3            3           102d
florence-patient              3/3     3            3           98d
florence-pm-master            3/3     3            3           74d
florence-pm-schedule-master   1/1     1            1           74d
florence-search               2/2     2            2           103d
florence-update               1/1     1            1           102d
florence-wx-accesstoken       3/3     3            3           105d

• 更改deployment配置

[root@bx-florence-k8smaster1 ~]*  kubectl edit deployments -n kafka log-archiver

• 更新POD

kubectl replace -f hello-world-pod.yaml

• 删除POD

kubectl delete pod hello-world

• 重建POd

kubectl replace --force -f hello-world-pod.yaml