Kubernetes实用技巧

kubectl命令太多太长记不住？

• 查看资源缩写

kubectl describe 回车

• 配置kubectl自动补全命令

source <(kubectl completion bash)

kubectl写yaml太累，找样例太麻烦？

• 用run命令生成

kubectl run --image=nginx my-deploy -o yaml --dry-run > my-deploy.yaml

• 用get命令导出

kubectl get statefulset/foo -o=yaml --export > new.yaml

• Pod亲和性下面字段的拼写忘记了

kubectl explain pod.spec.affinity.podAffinity

监控集群组件

• 集群整体状态

kubectl cluster-info 

• 更多集群信息

kubectl cluster-info dump

• 组件metrics

curl localhost:10250/stats/summary

• 组件监控状况

curl localhost:10250/healthz

管理K8s组件日志

# 组件日志
/var/log/kube-apiserver.log
/var/log/kube-proxy.log
/var/log/kube-controller-manager.log
/var/log/kubelet.log

• 使用systemd管理

 journalctl -u kubelet 

• 使用K8s插件部署

 kubectl logs -f kube-proxy

管理K8s应用日志

• 从容器标准输出截获

 kubectl logs -f {pod name} -c {container name}
 docker logs -f {docker name}

• 日志文件挂载到主机目录

 apiVersion: v1
 kind: Pod
 metadata:
   name: test-pod
 spec:
   containers:
   - image: test-webserver
     name: test-container
     volumeMounts:
     - mountPath: /log
       name: log-volume
   volumes:
   - name: log-volume
     hostPath:
       path: /var/k8s/log

Deployment升级与回滚

• 创建Deployment

kubectl run {deployment} --image={image} --replicas={rep.}

• 升级Deployment

kubectl set image deployment/nginx-deployment nginx=nignx:1.9.1
kubectl set resources deployment/nginx-deployment -c=nginx --limits=cpu=200m,memory=512Mi

• 升级策略

minReadySeconds: 5
strategy:
  type: RollingUpdata
    maxSurge: 1 #默认25%
    maxUnavailable: 1 #默认25%

• 暂停Deployment

kubectl rollout pause deployment/nginx-deployment

• 恢复Deployment

kubectl rollout resume deployment/nginx-deployment

• 查询升级状态

kubectl rollout status deployment/nginx-deployment

• 查询升级历史

kubectl rollout history deploy/nginx-deployment
kubectl rollout history deploy/nginx-deployment --revision=2

• 回滚

kubectl rollout undo deployment/nginx-deployment --to-revision=2

• 应用弹性伸缩

kubectl scale deployment nginx-deployment --replicas=10

• 对接了Heapster,和HPA联动后

kubectl autoscale deployment nginx-deployment --min=10 --max=15 --cpu-percent=80