Junzizhiai的博客

Deployment升级与回滚kubectl set image deployment/nginx-deployment nginx=nginx:1.9.1kubectl set resources deployment/nginx-deployment -c=nginx--limits=cpu=200m,memory=512Mi#暂停kubectl rollout pause deployment/nginx-deployment#恢复kubectl rollout resume depl

k8s容器探针（保证容器存货）apiVersion: v1kind: ReplicaSetmetadata: name: kubia-manual labels: name: kubia-manualspec: containers: - name: kubia-manual image: luksa/kubia resources: limits: memory: "128Mi" cpu: "500m" p

K8S 证书过期时间查询#脚本方法#/bin/bashfor item in `find /etc/kubernetes/pki -maxdepth 2 -name "*.crt"`;echo ======================$item===================;do openssl x509 -in $item -text -noout| grep Not;done# 指令方法[root@qtxian-k8s-master-1 ~]# kubeadm alpha c

Kubernetes集群应用ipvs后长连接失效问题最近在k8s集群升级 采取ipvs代理 部署后在调用gRPC服务时小概率出现Connection reset by peer的错误。简单理解下：gRPC是基于HTTP/2协议的，gRPC的client和server在交互时会建立多条连接，为了性能，这些连接都是长连接并且是一直保活着的。参考问题：github有一篇关于docker swarm的连接网络长连接问题https://github.com/moby/moby/issues/31208究其原