[系统运维]linux服务监控shell脚本实例

# func:监控BPC相关的服务
# author:zhonghuanhuan
# date:2021-11-1
# DEFINE

# 服务名称
ServerName=(bam bcc bpcplatform bpccenter)

# 获取服务相关的进行ID(其中[grep -w 'bim']代码中的是文件夹名)
BimID=$(ps -ef |grep "bam"|grep -v 'grep' |awk '{printf $2}')
BCCID=$(ps -ef |grep "bim-configuration"|grep -v 'grep' |awk '{printf $2}')
BPCPlatformID=$(ps -ef | grep "home/bpc/apache-tomcat-8.0.53" | grep -v grep | awk '{print $2}')
BPCCenterID=$(ps -ef | grep "/bpc/app-center/" | grep -v grep | awk '{print $2}')

IDS=($BimID $BCCID $BPCPlatformID $BPCCenterID)

# 启动程序的实际安装的路径
StartBim=/usr/local/tomcat/apache-tomcat-8.0.53/bin/startup.sh
BimCache=/usr/local/tomcat/apache-tomcat-8.0.53/work
   
StartBCC=/app/home/bcc/nohup-run.sh
BCCCache=/app/home/bcc/\*.gz

StartBPCPlatform=/app/home/bpc/apache-tomcat-8.0.53/bin/startup.sh 
BPCPlatformCache=/app/home/bpc/apache-tomcat-8.0.53/work

StartBPCCenter=/app/home/bpc/app-center/tomcat-center/bin/startup.sh 
BPCCenterCache=/app/home/bpc/app-center/tomcat-center/work

Starts=($StartBim $StartBCC $StartBPCPlatform $StartBPCCenter)
Caches=($BimCache $BCCCache $BPCPlatformCache $BPCCenterCache)

# 要监控的页面地址
BimWebUrl=http://172.16.20.51:8083/business-ui/login
BCCWebUrl=http://172.16.20.51:8021
BPCPlatformWebUrl=http://172.16.20.51:8015
BPCCenterWebUrl=http://172.16.20.51:8035

WebUrl=($BimWebUrl $BCCWebUrl $BPCPlatformWebUrl $BPCCenterWebUrl)

#日志输出
BimGetPageInfo=/tmp/BimMonitor.Info
BimMontorLog=/tmp/BimMonitor.log

BCCGetPageInfo=/tmp/BCCMonitor.Info
BCCMontorLog=/tmp/BCCMonitor.log

BPCPlatformGetPageInfo=/tmp/BPCPlatformMonitor.Info
BPCPlatformMontorLog=/tmp/BPCPlatformMonitor.log

BPCCenterGetPageInfo=/tmp/BPCCenterMonitor.Info
BPCCenterMontorLog=/tmp/BPCCenterMonitor.log

GetPageInfo=($BimGetPageInfo $BCCGetPageInfo $BPCPlatformGetPageInfo $BPCCenterGetPageInfo)
MontorLog=($BimMontorLog $BCCMontorLog $BPCPlatformMontorLog $BPCCenterMontorLog)

Monitor()
{
	echo "[info]开始监控${ServerName[$i]}...[$(date +'%F %H:%M:%S')]"
	if [ "${IDS[$i]}" ];then #这里判断进程是否存在
		echo "[info]当前${ServerName[$i]}进程ID为:${IDS[$i]},继续检测页面..."
		# 检测是否启动成功(成功的话页面会返回状态"200")
		ServiceCode=$(curl -s -o ${GetPageInfo[$i]} -m 10 --connect-timeout 10 ${WebUrl[$i]} -w %{http_code})
		if [ $ServiceCode -eq 200 ];then
			echo "[info]页面返回码为$ServiceCode，${ServerName[i]}启动成功，测试页面正常"
		else
			echo "[error]${ServerName[$i]}页面出错，请注意...状态码为$ServiceCode，错误日志已输出到${GetPageInfo[$i]}"
			echo "[error]页面访问出错，开始重启${ServerName[$i]}"
			kill -9 ${IDS[$i]} # 杀掉原进程
			if [ ${ServerName[$i]} = "bam" ];then
				sleep 600
			else
				sleep 3
			fi
			rm -rf ${Caches[$i]} # 清理缓存
			${Starts[i]}
		fi
	else 
		echo "[error]${ServerName[$i]}进程不存在!${ServerName[$i]}开始自动重启..."
		echo "[info]${Starts[$i]}，请稍候..."
		rm -rf ${Caches[$i]}
		${Starts[$i]}
	fi
		echo "--------------------------"
}

i=0
while(( $i < 4  ))
do
	Monitor
	let "i++"
done