Web服务自动监控shell _v2

在线上运行了一段时间发现有时候后台的进程没有down掉，但是web服务不能正常的解析jsp。把web服务重启后，一切恢复正常。

现在，把以前[b]扫描端口[/b]的方式修改[b]监控http请求返回的状态码[/b]来实现。

[b]操作步骤：[/b]
具体实现参考 [url=http://toptree.iteye.com/blog/1750957]Web服务自动监控shell[/url]

#!/bin/sh
#####################################################
#desc:扫描后台服务器的应用服务器，若不能正常访问则重启
#author：gaozhonghui
#mail：gaozhh@cdeledu.com
#date：20121127
#######################################################
year=`date -d "today" +"%Y"`
monthday=`date -d "today" +"%m"`
date=$(date -d "today" +"%Y%m%d")

#被监控服务器、端口列表
#str = web服务文件夹:端口号
server_all_list=(
	'www.test.com^http://127.0.0.1:8050/portal'
)

#应用服务器基路径
serverBasePath="/web/webserver/jboss/"
#日志路径
logBasePath="/web/webserver/logs/$year/$monthday/"

function_getPID(){
   local PID=`ps -ef|grep $1|grep java |awk '{print $2}'`
   echo $PID
}

function_jstackinfo(){
 PID=`function_getPID $1`
 if [ x$PID != x"" ] ;then
    if [ ! -d ${logBasePath} ];then
       mkdir -p ${logBasePath}  	
    fi

    jstack -l ${PID} >> ${logBasePath}"jstack_$1_${date}.log"	
 fi
}

function_shutdown(){
  local shutdownSh=${serverBasePath}$1"/bin/shutdown.sh"
  if [ -f $shutdownSh ];then
        local PID=`function_getPID $1`

        if [ x$PID != x"" ] ;then
                sh $shutdownSh > /dev/null 2>&1
                sleep 2
        fi

        local PID2=`function_getPID $1`

        if [ x$PID2 != x"" ];then
                kill -9 $PID2
        fi
  fi	
}

function_startup(){
 local startupSh=${serverBasePath}$1"/bin/startup.sh"
 if [ -f $shutdownSh ];then
	sh $startupSh > /dev/null 2>&1
 fi
}


function_restart(){
  function_shutdown $1
  sleep 5
  function_startup $1
}


#获得http 状态码
function_getHttpCode(){
  local http_code=`curl -o /dev/null -s -m $1 --connect-timeout $1 -w %{http_code} $2`
  local first_http_code=${http_code:0:1}
  echo $first_http_code
}


for server in ${server_all_list[@]}; do
   server_ip=127.0.0.1
   server_name=$(echo ${server} | awk -F '^' '{print $1}')
   server_url=$(echo ${server} | awk -F '^' '{print $2}')   
   http_code_10s=`function_getHttpCode 10 $server_url`

   if [ "$http_code_10s" != "0" ] && [ "$http_code_10s" != "4" ] && [ "$http_code_10s" != "5" ];then
       #status:    0,http down    1,http ok    2,http down but ping ok 
       status=1
   else
       http_code_30s=`function_getHttpCode 30 $server_url`	

       if [ "$http_code_30s" != "0" ] && [ "$http_code_10s" != "4" ] && [ "$http_code_10s" != "5" ];then
           status=1
       else
           if ping -c 1 $server_ip > /dev/null 2>&1;then
               status=2
           else
               status=0
           fi
       fi
   fi

   #根据不同状态操作服务
   if [ x$status != x"1" ];then
	PID=`function_getPID  ${server_name}`

	if [ x$PID != x ];then
		function_jstackinfo  ${server_name}
		function_restart  ${server_name}
	else
		function_startup ${server_name}
	fi	
   fi

done