shell监控脚本

序言:

前几天一好友问我服务器监控怎么做？你们公司的监控是怎么做的？有什么开源的监控软件推荐？常见的开源的监控软件当然首先推荐ZABBIX，分布式够强大，而且很多公司都在用，我问他具体什么需求，能监控服务器负载、CPU使用率、内存使用率、硬盘使用率、网卡流量监控、网站PV，UV统计、异常错误日志，一旦超出设定的阈值就发邮件告警。确定需求后准备将ZABBIX的一些参考资料发给他看看，末了问了一句，你们有多少台机器，他说就3台，一个在线商城网站，哈哈...就3台机器整毛线ZABBIX，用shell脚本吧，用了一天时间给写了个初版的shell监控脚本，效果如下图：

监控项：

• 15分钟的系统平均负载
• CPU使用率
• 内存使用率
• 虚拟内存使用率
• 硬盘使用率
• 网卡流量
• 网站PV统计
• 网站UV统计

邮件发送设置：

• 每小时监控监控1次，超出设置的阀值就发送邮件通知。
• 每天08:00发送邮件，上班后可查看前天服务器运行情况。

1.监控脚本

 #！/usr/bin/env  bash
 #
 # Author     : Jack zhao
 # Data         : //
 # Description: This script is used to monitor the use of local system resources.

 ###The global variable###
 export PATH=$PATH:/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/root/bin
 export LANG=C
 INITDIR=`pwd`
 LOCAL_TIME=`date '+%F %H:%M:%S'`
 BEFORE_DAY=`date -d "-1 days" +%F`
 WHOLE_TIME=`date '+%H'`

 ###Mailbox configuration###
 ACCENT_MAIL="[e-mail]"

 ###The alarm threshold###
 LOAD_AVERAGE_ALARM=
 CPU_ALARM=
 MEMORY_ALARM=
 SWAP_ALARM=
 ROOT_PART_ALARM=
 IN_NETWORK_TRAFFIC_ALARM=
 OUT_NETWORK_TRAFFIC_ALARM=

 ###Log variable###
 ACCESS_LOG_PATH=localhost_access_log.${BEFORE_DAY}.txt

 ####Timed task###
 TIMING=

 ###Public function###
 out_format(){
     printf "%-22s %-5s %-5s \n" "$1" ":" "$2"
 }

 ###Send an alarm mail###
 send_mail(){
     python mail.py "$ACCENT_MAIL" "[Alarm]:$1" "`cat $INITDIR/Monitoring.log`"
 }

 ###Get load information  minutes###
 get_load_average(){
     LOAD_AVERAGE=`uptime | awk -F "," '{print$NF}' | sed 's#[[:space:]]##g' `
     out_format "Load average"  "$LOAD_AVERAGE" >> $INITDIR/Monitoring.log
     return
 }

 ###Get CPU usage###
 get_cpu_usage(){
     CPU_FREE=`vmstat   |sed -n '3,$p' |awk '{x = x + $15} END {print x/5}' |awk -F. '{print $1}'`
     CPU_USAGE=$(( - $CPU_FREE))
     out_format "CPU usage"  "${CPU_USAGE}%" >> $INITDIR/Monitoring.log
     return
 }

 ###Get memory usage###
 get_memory_usage(){
     MEMORY_USED=`free -m | grep -w "buffers\/cache:" | awk '{print$3}'`
     MEMORY_TOTAL=`free -m  | grep "^Mem" | awk '{print$2}'`
     MEMORY_USAGE=`echo "scale=2;${MEMORY_USED}/${MEMORY_TOTAL}*100;" | bc -l`
     out_format "Memory usage"  "${MEMORY_USAGE}%" >> $INITDIR/Monitoring.log
     return
 }

 ###Get Swap usage###
 get_swap_usage(){
     SWAP_USED=`free -m | grep -w "^Swap" | awk '{print$3}'`
     SWAP_TOTAL=`free -m | grep -w "^Swap" | awk '{print$2}'`
     SWAP_USAGE=`echo "scale=2;${SWAP_USED}/${SWAP_TOTAL}*100;" | bc -l`
     out_format "Swap usage"  "${SWAP_USAGE}%" >> $INITDIR/Monitoring.log
     return
 }

 ###Get disk usage###
 get_disk_usage(){
     MOUNT_POINT=`df -hP | egrep -wv  '^tmpfs|Filesystem|boot' | awk '{print$NF}'`
     for i in `echo $MOUNT_POINT`;do
         DISK_USAGE=`df -hP | grep -w "$i" | awk '{print$5}'`
         out_format "Disk usage $i"  "${DISK_USAGE}" >> $INITDIR/Monitoring.log
     done
     ROOT_PART=`df -hP  | grep -w "/" | awk '{print$5}' | awk -F "%" '{print$1}'`
     return
 }

 ###Get network traffic###
 get_network_traffic(){
     NETWORK_TRAFFIC=`sar -n DEV  |grep Average|grep eth0|awk '{print "Input:",$5*1000*8,"bps","Output:",$6*1000*8,"bps"}'`
     out_format "Network traffic"  "${NETWORK_TRAFFIC}" >> $INITDIR/Monitoring.log
     return
 }

 ###Get website PV,UV###
 get_website_pu(){
     WEBSITE_PV=`grep -v "^127.0.0.1" ${ACCESS_LOG_PATH}  | wc -l`
     out_format "Before Day Website PV" "$WEBSITE_PV" >> $INITDIR/Monitoring.log
     WEBSITE_UV=`grep -v "^127.0.0.1" ${ACCESS_LOG_PATH} | awk '{print$1}' | sort | uniq | wc -l`
     out_format "Before Day Website UV" "$WEBSITE_UV" >> $INITDIR/Monitoring.log
 }

 ###Server information###
 IP=`ip addr show eth0 | grep -w "inet" | awk -F "[ /]+" '{print$3}'`
 HOSTNAME=`hostname`
 SN=`sudo -b /usr/sbin/dmidecode | grep -A5 "System Information" | grep "Serial Number"| awk -F ":" '{print$2}' | sed 's#^[[:space:]]##g'`
 OS=`cat /etc/redhat-release`

 ###Program entrance###
 echo "############### Monitoring information Date:$LOCAL_TIME ###############" > $INITDIR/Monitoring.log
 out_format "IP"  "$IP" >> $INITDIR/Monitoring.log
 out_format "Hostname"  "$HOSTNAME" >> $INITDIR/Monitoring.log
 out_format "SN"  "$SN" >> $INITDIR/Monitoring.log
 out_format "Os"  "$OS" >> $INITDIR/Monitoring.log

 ###Call function###
 get_load_average
 get_cpu_usage
 get_memory_usage
 get_swap_usage
 get_disk_usage
 get_network_traffic
 get_website_pu

 ###variable format###
 LOAD_AVERAGE_FORMAT=`echo $LOAD_AVERAGE | awk -F "." '{print$1}'`
 MEMORY_USAGE_FORMAT=`echo $MEMORY_USAGE | awk -F "." '{print$1}'`
 SWAP_USAGE_FORMAT=`echo $SWAP_USAGE | awk -F "." '{print$1}'`
 IN_NETWORK_TRAFFIC=`echo $NETWORK_TRAFFIC | awk '{print$2}'`
 OUT_NETWORK_TRAFFIC=`echo $NETWORK_TRAFFIC | awk '{print$5}'`

 ##Monitoring threshold judgment###
 if [ ${LOAD_AVERAGE_FORMAT} -ge ${LOAD_AVERAGE_ALARM} ];then
     send_mail "$IP Load:High load average."
 elif [ $CPU_USAGE -ge $CPU_ALARM ]; then
     send_mail "$IP CPU:usage greater than ${CPU_ALARM}%."
 elif [ $MEMORY_USAGE_FORMAT -ge $MEMORY_ALARM ];then
     send_mail "$IP Memory:usage greater than ${MEMORY_ALARM}%."
 elif [ $SWAP_USAGE_FORMAT -ge $SWAP_ALARM ];then
     send_mail "$IP Swap:usage greater than ${SWAP_ALARM}%."
 elif [ $ROOT_PART -ge $ROOT_PART_ALARM ]; then
     send_mail "$IP Root Partition:usage greater than ${ROOT_PART_ALARM}%."
 elif [ $IN_NETWORK_TRAFFIC -ge $IN_NETWORK_TRAFFIC_ALARM ]; then
     send_mail "$IP Network Traffic:Input Greater than ${IN_NETWORK_TRAFFIC_ALARM}bps."
 elif [ $OUT_NETWORK_TRAFFIC -ge $OUT_NETWORK_TRAFFIC_ALARM ]; then
     send_mail "$IP Network Traffic:Output Greater than ${OUT_NETWORK_TRAFFIC_ALARM}bps."
 fi

 ###Record history log###
 cat $INITDIR/Monitoring.log >> $INITDIR/Monitor_history.log

 ###Send a notification email at a time###
 if [ "$WHOLE_TIME" == "$TIMING" ];then
     python mail.py "$ACCENT_MAIL" "[Notice]:$IP Monitoring information" "`cat $INITDIR/Monitoring.log`"
 fi

 ###Delete temporary files###
 INITDIR_CONFIRM=${INITDIR:-/tmp}
 rm -f $INITDIR_CONFIRM/Monitoring.log

2.发送邮件脚本

 #!/usr/bin/env python
 #-*- coding: UTF-8 -*-
 import os,sys
 reload(sys)
 sys.setdefaultencoding('utf8')
 import getopt
 import smtplib
 from email.MIMEText import MIMEText
 from email.MIMEMultipart import MIMEMultipart
 from  subprocess import *

 def sendqqmail(username,password,mailfrom,mailto,subject,content):
     gserver = 'smtp.qq.com'
     gport = 465

     try:
         msg = MIMEText(unicode(content).encode('utf-8'))
         msg['from'] = mailfrom
         msg['to'] = mailto
         msg['Reply-To'] = mailfrom
         msg['Subject'] = subject

         smtp = smtplib.SMTP_SSL(gserver, gport)
         smtp.set_debuglevel(0)
         smtp.ehlo()
         smtp.login(username,password)

         smtp.sendmail(mailfrom, mailto, msg.as_string())
         smtp.close()
     except Exception,err:
         print "Send mail failed. Error: %s" % err

 def main():
     to=sys.argv[1]
     subject=sys.argv[2]
     content=sys.argv[3]
     sendqqmail('','gkhcvklnjyjsbgwic','[e-mail]',to,subject,content)
 if __name__ == "__main__":
     main()

3.readme配置

 1.新建一个监控用户（monitor）
     [root@mylab ~]# useradd monitor
     [root@mylab ~]# passwd monitor

 2.配置sudo
     [root@mylab ~]# cp /etc/sudoers /etc/sudoers.bak.`date +'%F-%T'`
     [root@mylab ~]# echo "monitor ALL=(root) NOPASSWD: /usr/sbin/dmidecode" >> /etc/sudoers
     [root@mylab ~]# vim /etc/sudoers
     # Defaults    requiretty
     将上面行注释掉
 3.上传脚本文件至/home/monitor
     mail.py #发送邮件脚本
     monitor.sh #监控脚本

 4.配置文件说明
     mail.py #调用第三方邮箱发送邮件
         以QQ邮件为例修改mail.py文件：（其他平台修改参数即可,一般公司都有邮件接口或者自己内网搭建一台邮件服务器）
         gserver = 'smtp.qq.com'
         gport = 465
         sendqqmail('123456','gkhcvklnjyjsbgic','[e-mail]',to,subject,content)
             QQ邮箱用户名    IMAP/SMTP第三方登录授权码（QQ邮箱设置生成）QQ邮箱

     monitor.sh #可监控负载、CPU、内存、SWAP、硬盘、网卡流量、网站PV/UV(需要根据需求修改)
         ###Mailbox configuration###
         ACCENT_MAIL="[e-mail]" #告警接受邮箱

         告警阀值，可根据需要修改
         ###The alarm threshold###
         LOAD_AVERAGE_ALARM=2 #15分钟内的平均负载，参考：一般小于机器物理CPU个数
         CPU_ALARM=70 #CPU使用率
         MEMORY_ALARM=70 #内存使用率
         SWAP_ALARM=30 #虚拟内存使用率
         ROOT_PART_ALARM=70 #/挂载点硬盘使用率
         IN_NETWORK_TRAFFIC_ALARM=167772160 #网卡每分钟接收流量20M
         OUT_NETWORK_TRAFFIC_ALARM=167772160 #网络每分钟发送流量20M

         应用日志路径
         ###Log variable###
         ACCESS_LOG_PATH=localhost_access_log.${BEFORE_DAY}.txt

     Monitor_history.log #日志历史记录

 5.定时发送邮件
     ####Timed task###
     TIMING=08 #每天08点发生邮件

 6.添加自动任务
     [monitor@mylab ~]$ crontab  -e
     1 * * * *  /bin/sh  -c  "/home/monitor/monitor.sh &>/dev/null" #每个小时的第一分钟执行一次

4.写的最后

目前在测试环境运行了一段时间，待优化后上生产。哈哈....