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
1 #！/usr/bin/env bash 2 # 3 # Author : Jack zhao 4 # Data : 2017/12/08 5 # Description: This script is used to monitor the use of local system resources. 6 7 ###The global variable### 8 export PATH=$PATH:/usr/kerberos/sbin:/usr/kerberos/bin:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/root/bin 9 export LANG=C 10 INITDIR=`pwd` 11 LOCAL_TIME=`date '+%F %H:%M:%S'` 12 BEFORE_DAY=`date -d "-1 days" +%F` 13 WHOLE_TIME=`date '+%H'` 14 15 ###Mailbox configuration### 16 ACCENT_MAIL="123456@126.com" 17 18 ###The alarm threshold### 19 LOAD_AVERAGE_ALARM=2 20 CPU_ALARM=70 21 MEMORY_ALARM=70 22 SWAP_ALARM=30 23 ROOT_PART_ALARM=95 24 IN_NETWORK_TRAFFIC_ALARM=167772160 25 OUT_NETWORK_TRAFFIC_ALARM=167772160 26 27 ###Log variable### 28 ACCESS_LOG_PATH=localhost_access_log.${BEFORE_DAY}.txt 29 30 31 ####Timed task### 32 TIMING=08 33 34 ###Public function### 35 out_format(){ 36 printf "%-22s %-5s %-5s \n" "$1" ":" "$2" 37 } 38 39 ###Send an alarm mail### 40 send_mail(){ 41 python mail.py "$ACCENT_MAIL" "[Alarm]:$1" "`cat $INITDIR/Monitoring.log`" 42 } 43 44 ###Get load information 15 minutes### 45 get_load_average(){ 46 LOAD_AVERAGE=`uptime | awk -F "," '{print$NF}' | sed 's#[[:space:]]##g' ` 47 out_format "Load average" "$LOAD_AVERAGE" >> $INITDIR/Monitoring.log 48 return 0 49 } 50 51 ###Get CPU usage### 52 get_cpu_usage(){ 53 CPU_FREE=`vmstat 1 5 |sed -n '3,$p' |awk '{x = x + $15} END {print x/5}' |awk -F. '{print $1}'` 54 CPU_USAGE=$((100 - $CPU_FREE)) 55 out_format "CPU usage" "${CPU_USAGE}%" >> $INITDIR/Monitoring.log 56 return 0 57 } 58 59 ###Get memory usage### 60 get_memory_usage(){ 61 MEMORY_USED=`free -m | grep -w "buffers\/cache:" | awk '{print$3}'` 62 MEMORY_TOTAL=`free -m | grep "^Mem" | awk '{print$2}'` 63 MEMORY_USAGE=`echo "scale=2;${MEMORY_USED}/${MEMORY_TOTAL}*100;" | bc -l` 64 out_format "Memory usage" "${MEMORY_USAGE}%" >> $INITDIR/Monitoring.log 65 return 0 66 } 67 68 ###Get Swap usage### 69 get_swap_usage(){ 70 SWAP_USED=`free -m | grep -w "^Swap" | awk '{print$3}'` 71 SWAP_TOTAL=`free -m | grep -w "^Swap" | awk '{print$2}'` 72 SWAP_USAGE=`echo "scale=2;${SWAP_USED}/${SWAP_TOTAL}*100;" | bc -l` 73 out_format "Swap usage" "${SWAP_USAGE}%" >> $INITDIR/Monitoring.log 74 return 0 75 } 76 77 ###Get disk usage### 78 get_disk_usage(){ 79 MOUNT_POINT=`df -hP | egrep -wv '^tmpfs|Filesystem|boot' | awk '{print$NF}'` 80 for i in `echo $MOUNT_POINT`;do 81 DISK_USAGE=`df -hP | grep -w "$i" | awk '{print$5}'` 82 out_format "Disk usage $i" "${DISK_USAGE}" >> $INITDIR/Monitoring.log 83 done 84 ROOT_PART=`df -hP | grep -w "/" | awk '{print$5}' | awk -F "%" '{print$1}'` 85 return 0 86 } 87 88 ###Get network traffic### 89 get_network_traffic(){ 90 NETWORK_TRAFFIC=`sar -n DEV 1 60|grep Average|grep eth0|awk '{print "Input:",$5*1000*8,"bps","Output:",$6*1000*8,"bps"}'` 91 out_format "Network traffic" "${NETWORK_TRAFFIC}" >> $INITDIR/Monitoring.log 92 return 0 93 } 94 95 ###Get website PV,UV### 96 get_website_pu(){ 97 WEBSITE_PV=`grep -v "^127.0.0.1" ${ACCESS_LOG_PATH} | wc -l` 98 out_format "Before Day Website PV" "$WEBSITE_PV" >> $INITDIR/Monitoring.log 99 WEBSITE_UV=`grep -v "^127.0.0.1" ${ACCESS_LOG_PATH} | awk '{print$1}' | sort | uniq | wc -l` 100 out_format "Before Day Website UV" "$WEBSITE_UV" >> $INITDIR/Monitoring.log 101 } 102 103 ###Server information### 104 IP=`ip addr show eth0 | grep -w "inet" | awk -F "[ /]+" '{print$3}'` 105 HOSTNAME=`hostname` 106 SN=`sudo -b /usr/sbin/dmidecode | grep -A5 "System Information" | grep "Serial Number"| awk -F ":" '{print$2}' | sed 's#^[[:space:]]##g'` 107 OS=`cat /etc/redhat-release` 108 109 ###Program entrance### 110 echo "############### Monitoring information Date:$LOCAL_TIME ###############" > $INITDIR/Monitoring.log 111 out_format "IP" "$IP" >> $INITDIR/Monitoring.log 112 out_format "Hostname" "$HOSTNAME" >> $INITDIR/Monitoring.log 113 out_format "SN" "$SN" >> $INITDIR/Monitoring.log 114 out_format "Os" "$OS" >> $INITDIR/Monitoring.log 115 116 ###Call function### 117 get_load_average 118 get_cpu_usage 119 get_memory_usage 120 get_swap_usage 121 get_disk_usage 122 get_network_traffic 123 get_website_pu 124 125 ###variable format### 126 LOAD_AVERAGE_FORMAT=`echo $LOAD_AVERAGE | awk -F "." '{print$1}'` 127 MEMORY_USAGE_FORMAT=`echo $MEMORY_USAGE | awk -F "." '{print$1}'` 128 SWAP_USAGE_FORMAT=`echo $SWAP_USAGE | awk -F "." '{print$1}'` 129 IN_NETWORK_TRAFFIC=`echo $NETWORK_TRAFFIC | awk '{print$2}'` 130 OUT_NETWORK_TRAFFIC=`echo $NETWORK_TRAFFIC | awk '{print$5}'` 131 132 ##Monitoring threshold judgment### 133 if [ ${LOAD_AVERAGE_FORMAT} -ge ${LOAD_AVERAGE_ALARM} ];then 134 send_mail "$IP Load:High load average." 135 elif [ $CPU_USAGE -ge $CPU_ALARM ]; then 136 send_mail "$IP CPU:usage greater than ${CPU_ALARM}%." 137 elif [ $MEMORY_USAGE_FORMAT -ge $MEMORY_ALARM ];then 138 send_mail "$IP Memory:usage greater than ${MEMORY_ALARM}%." 139 elif [ $SWAP_USAGE_FORMAT -ge $SWAP_ALARM ];then 140 send_mail "$IP Swap:usage greater than ${SWAP_ALARM}%." 141 elif [ $ROOT_PART -ge $ROOT_PART_ALARM ]; then 142 send_mail "$IP Root Partition:usage greater than ${ROOT_PART_ALARM}%." 143 elif [ $IN_NETWORK_TRAFFIC -ge $IN_NETWORK_TRAFFIC_ALARM ]; then 144 send_mail "$IP Network Traffic:Input Greater than ${IN_NETWORK_TRAFFIC_ALARM}bps." 145 elif [ $OUT_NETWORK_TRAFFIC -ge $OUT_NETWORK_TRAFFIC_ALARM ]; then 146 send_mail "$IP Network Traffic:Output Greater than ${OUT_NETWORK_TRAFFIC_ALARM}bps." 147 fi 148 149 ###Record history log### 150 cat $INITDIR/Monitoring.log >> $INITDIR/Monitor_history.log 151 152 ###Send a notification email at a time### 153 if [ "$WHOLE_TIME" == "$TIMING" ];then 154 python mail.py "$ACCENT_MAIL" "[Notice]:$IP Monitoring information" "`cat $INITDIR/Monitoring.log`" 155 fi 156 157 ###Delete temporary files### 158 INITDIR_CONFIRM=${INITDIR:-/tmp} 159 rm -f $INITDIR_CONFIRM/Monitoring.log2.发送邮件脚本
1 #!/usr/bin/env python 2 #-*- coding: UTF-8 -*- 3 import os,sys 4 reload(sys) 5 sys.setdefaultencoding('utf8') 6 import getopt 7 import smtplib 8 from email.MIMEText import MIMEText 9 from email.MIMEMultipart import MIMEMultipart 10 from subprocess import * 11 12 def sendqqmail(username,password,mailfrom,mailto,subject,content): 13 gserver = 'smtp.qq.com' 14 gport = 465 15 16 try: 17 msg = MIMEText(unicode(content).encode('utf-8')) 18 msg['from'] = mailfrom 19 msg['to'] = mailto 20 msg['Reply-To'] = mailfrom 21 msg['Subject'] = subject 22 23 smtp = smtplib.SMTP_SSL(gserver, gport) 24 smtp.set_debuglevel(0) 25 smtp.ehlo() 26 smtp.login(username,password) 27 28 smtp.sendmail(mailfrom, mailto, msg.as_string()) 29 smtp.close() 30 except Exception,err: 31 print "Send mail failed. Error: %s" % err 32 33 34 def main(): 35 to=sys.argv[1] 36 subject=sys.argv[2] 37 content=sys.argv[3] 38 sendqqmail('123456','gkhcvklnjyjsbgwic','123456@qq.com',to,subject,content) 39 if __name__ == "__main__": 40 main()3.readme配置
1 1.新建一个监控用户（monitor） 2 [root@mylab ~]# useradd monitor 3 [root@mylab ~]# passwd monitor 4 5 2.配置sudo 6 [root@mylab ~]# cp /etc/sudoers /etc/sudoers.bak.`date +'%F-%T'` 7 [root@mylab ~]# echo "monitor ALL=(root) NOPASSWD: /usr/sbin/dmidecode" >> /etc/sudoers 8 [root@mylab ~]# vim /etc/sudoers 9 # Defaults requiretty 10 将上面行注释掉 11 3.上传脚本文件至/home/monitor 12 mail.py #发送邮件脚本 13 monitor.sh #监控脚本 14 15 4.配置文件说明 16 mail.py #调用第三方邮箱发送邮件 17 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） 18 gserver = 'smtp.qq.com' 19 gport = 465 20 sendqqmail('123456','gkhcvklnjyjsbgic','123456@qq.com',to,subject,content) 21 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 22 23 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) 24 ###Mailbox configuration### 25 ACCENT_MAIL="123456@126.com" #告警接受邮箱 26 27 告警阀值，可根据需要修改 28 ###The alarm threshold### 29 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 30 CPU_ALARM=70 #CPU使用率 31 MEMORY_ALARM=70 #内存使用率 32 SWAP_ALARM=30 #虚拟内存使用率 33 ROOT_PART_ALARM=70 #/挂载点硬盘使用率 34 IN_NETWORK_TRAFFIC_ALARM=167772160 #网卡每分钟接收流量20M 35 OUT_NETWORK_TRAFFIC_ALARM=167772160 #网络每分钟发送流量20M 36 37 应用日志路径 38 ###Log variable### 39 ACCESS_LOG_PATH=localhost_access_log.${BEFORE_DAY}.txt 40 41 Monitor_history.log #日志历史记录 42 43 5.定时发送邮件 44 ####Timed task### 45 TIMING=08 #每天08点发生邮件 46 47 48 6.添加自动任务 49 [monitor@mylab ~]$ crontab -e 50 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....
转载于:https://www.cnblogs.com/changmingzhao/p/8012318.html
相关资源：shell脚本（监控日志发告警邮件）