制作Linux程序监控脚本

程序监控脚本

guard_cmms.sh

#!/bin/bash

#需要守护的进程数组,将需要守护的进程填入数组中,如 PRO_NAMES=(./bin/pro1 ./pro2 ) 表示要守护pro1、pro2 进程
PRO_NAMES=(
"maincmms"
"haikangaccessdevice" 
"dahuaaccessdevice"
)


#不保存控制台输出日志
OUTPUT="/dev/null"

#守护进程的数量
PRO_COUNT=${#PRO_NAMES[*]}

echo "PRO_COUNT = ${PRO_COUNT}"

start()
{
  while true ; do
       for ((i=0; i<${PRO_COUNT}; i++))
        do
        #  默认关闭日志输出
        #  OUTPUT="${PRO_NAMES[i]}.log"
        #    用ps获取$PRO_NAME进程数量
          NUM=`ps aux | grep ${PRO_NAMES[i]} | grep -v grep |wc -l`
          echo -e "####${PRO_NAMES[i]} \t pro_count = $NUM"
        #    少于1，重启进程
          if [ "${NUM}" -lt "1" ];then
                cd /opt/cvos/cmms
                nohup ./${PRO_NAMES[i]} >${OUTPUT} 2>&1 &
                cd -
        #    大于1，杀掉所有进程，重启
          elif [ "${NUM}" -gt "1" ];then
                echo "##### more than 1 ${PRO_NAMES[i]},killall ${PRO_NAMES[i]}}"
                killall  ${PRO_NAMES[i]}
                cd /opt/cvos/cmms
                nohup ./${PRO_NAMES[i]} >${OUTPUT} 2>&1 &
                cd -
          fi
        #    kill僵尸进程
          NUM_STAT=`ps aux | grep ${PRO_NAMES[i]} | grep T | grep -v grep | wc -l`

          if [ "${NUM_STAT}" -gt "0" ];then
                killall  ${PRO_NAMES[i]}
                cd /opt/cvos/cmms
                nohup ./${PRO_NAMES[i]} >${OUTPUT} 2>&1 &
                cd -
          fi

        done


    sleep 2
  done
}

stop()
{  
        for ((i=0; i<${PRO_COUNT}; i++))
        do

        killall ${PRO_NAMES[i]}

        done

}

case "$1" in 
 start)  
        start 
        ;; 
 stop)   
        stop 
        ;; 
 restart) 
         echo $"Restaring " 
         stop
   echo "stop over!" 
         sleep 1 
         start 
         ;; 
 *) 
        echo $"Usage: $0 {start|stop|restart}" 
        exit 1 
        ;; 
esac 
  
exit 0

服务启停脚本

#!/bin/bash
#chkconfig: 2345 80 90
#description:guard_cmms.sh

start()
{
  cd /opt/cvos/cmms/
  nohup ./guard_cmms.sh  start  > /dev/null 2>&1 &
  cd -
}

stop()
{
  cd /opt/cvos/cmms/
  killall guard_cmms.sh
  ./guard_cmms.sh  stop 
  cd -
}

restart()
{
  cd /opt/cvos/cmms
  nohup ./guard_cmms.sh restart  > /dev/null 2>&1 &
  cd -
}


case "$1" in 
 start)  
        start 
        ;; 
 stop)   
        stop 
        ;; 
 restart) 
  restart 
         ;; 
 *) 
        echo $"Usage: $0 {start|stop|restart}" 
        exit 1 
        ;; 
esac 
  
exit 0

注意

编写程序作为守护监控，调用fork函数启动新进程，然后调用execl启动监控程序，发现主进程被阻塞