shell脚本监控k8s集群job状态，若出现error通过触发阿里云的进程监控报警

#!/bin/bash

while [ 1 ]
   
do
 
   job_error_no=`kubectl get pod -n weifeng |grep -i "job"|grep -ci error`
   
   
 
   if [ $job_error_no -gt 0  ];then
      ps -fe|grep k8s_job_status_monitor|grep -v grep|awk '{print $2}'|xargs kill -9
      echo "k8s job running  is not stable " >> /tmp/k8s_job_error_no.log
 
   fi
   sleep 60
 
done

若k8s集群job状态出现error , 脚本自动kill 掉自己的 montior进程，通过阿里云的云监控进程监控来触发报警

阿里云云监控进程监控文档 https://www.cnblogs.com/weifeng1463/p/11591796.html