[系统运维] 使用ansible巡检Linux系统并汇总成csv文件

开发: C++知识库 Java知识库 JavaScript Python PHP知识库人工智能区块链大数据移动开发嵌入式开发工具数据结构与算法开发测试游戏开发网络协议系统运维
教程: HTML教程 CSS教程 JavaScript教程 Go语言教程 JQuery教程 VUE教程 VUE3教程 Bootstrap教程 SQL数据库教程 C语言教程 C++教程 Java教程 Python教程 Python3教程 C#教程
数码: 电脑笔记本显卡显示器固态硬盘硬盘耳机手机 iphone vivo oppo 小米华为单反装机图拉丁

-> 系统运维 -> 使用ansible巡检Linux系统并汇总成csv文件 -> 正文阅读

[系统运维]使用ansible巡检Linux系统并汇总成csv文件

collectLog.sh脚本

#!/bin/sh
###############################################################################################################
#Script name: collectLog.sh
#Script description: for Ansible Server collectLog 
#Current Release Version: 1.0.0
#Script Owner: He ,Haibo
#Latest editor: He, Haibo
#Support platform:  Linux OS for redhat and centos.
#Date: 2021/4/21---first Version for collectLog.sh.
#Change log:
#Descript:sh collectLog.sh
#
###############################################################################################################

excuteUser(){
	excuteCurrentUser=`whoami`
	if [[ $excuteCurrentUser == "sysinsp" ]];then
		continue
	else
		echo "Current excute user is $excuteCurrentUser,scripts will exit."
		exit 1
	fi
}

mkdirLogPath(){
    if [[ ! -d /home/sysinsp/LinuxCheck/logs ]];then
        mkdir -p /home/sysinsp/LinuxCheck/logs
        chmod 767 /home/sysinsp/LinuxCheck/logs
    fi
    ansibleLinuxLogPath="/home/sysinsp/LinuxCheck/logs"
    ansibleLinuxLog=ansible.log	
}


copyLinuxShell(){
	ansible -i hosts webservers -m ping >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
	if [[ $? == 0 ]];then
		ansible webservers -i hosts -m copy -a "src=/home/sysinsp/linuxHealthCheck.sh dest=/home/sysinsp/linuxHealthCheck.sh mode=0755" >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
		if [[ $? -ne 0 ]];then
			echo "Cannot copy /home/sysinsp/linuxHealthCheck.sh to dest." "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
			exit 1
		fi
	else
		echo "ansible -i hosts webservers -m copy failed." >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
		exit 1
	fi
}

excuteLinuxShell(){
	ansible webservers -i hosts -m shell -a "timeout -k 10s 1m /home/sysinsp/linuxHealthCheck.sh" >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
#	if [[ $? == 0 ]];then
#		echo "ansible webservers -i hosts -m shell -a /home/sysinsp/linuxHealthCheck.sh excute successfully." >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
#	else
#		echo "ansible webservers -i hosts -m shell -a /home/sysinsp/linuxHealthCheck.sh excute failed." >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
#	fi
}

copyLogFileToLocal(){
	ansible webservers -i hosts -m synchronize -a "mode=pull src=/tmp/log/*_linuxHealthCheck_$(date +%y%m%d).csv dest=$ansibleLinuxLogPath" >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
	if [[ $? == 0 ]];then
		echo "ansible webservers -i hosts -m synchronize -a 'mode=pull src=/tmp/log/*_linuxHealthCheck_$(date +%y%m%d).csv dest=$ansibleLinuxLogPath' successfully." >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
	else
		echo "ansible webservers -i hosts -m synchronize -a 'mode=pull src=/tmp/log/*_linuxHealthCheck_$(date +%y%m%d).csv dest=$ansibleLinuxLogPath' failed." >> "$ansibleLinuxLogPath/$ansibleLinuxLog" 2>&1
	fi
}

moveLogs(){
	if [[ ! -d "$ansibleLinuxLogPath"/$(date +%y%m%d) ]];then
		mkdir "$ansibleLinuxLogPath"/$(date +%y%m%d) && cd "$ansibleLinuxLogPath" && mv *_linuxHealthCheck_$(date +%y%m%d).csv "$ansibleLinuxLogPath"/$(date +%y%m%d)
	else
		cd "$ansibleLinuxLogPath" && mv *_linuxHealthCheck_$(date +%y%m%d).csv "$ansibleLinuxLogPath"/$(date +%y%m%d)
	fi
}

generateTotalCsv(){
	if [[ -d "$ansibleLinuxLogPath"/$(date +%y%m%d) ]];then
		cd "$ansibleLinuxLogPath"/$(date +%y%m%d)
		num=1
		echo " " > "$ansibleLinuxLogPath"/$(date +%y%m%d)/linuxHealthCheck_TOTAL_$(date +%y%m%d).csv
		sed -i "$num"c" HOSTNAME , IPADDR , CHECK_DATE , KERNEL , SSH_VERSION , OPENSSL_VERSION , OS_VERSION , UPTIME , SYSTEM_TIME , CPU , MEMORY , SWAP , INODE, DISK , IO , USERS , DEFUNCT , NTP_STATUS , TIMEZONE , DSA_STATUS \n"  linuxHealthCheck_TOTAL_$(date +%y%m%d).csv
		for filename in `ls *_linuxHealthCheck_$(date +%y%m%d).csv | xargs`
		do
			num=$(($num+1))
#			echo $num
			HOSTNAME=`cat $filename | awk -F ',' {'print $1'} | sed 's/[ \t]//g'`
			IPADDR=`cat $filename | awk -F ',' {'print $2'} | sed 's/[ \t]//g'`
			CHECK_DATE=`cat $filename | awk -F ',' {'print $3'} | sed 's/[ \t]//g'`
			KERNEL=`cat $filename | awk -F ',' {'print $4'} | sed 's/[ \t]//g'`
			SSH_VERSION=`cat $filename | awk -F ',' {'print $5'} | sed 's/[ \t]//g'`
			OPENSSL_VERSION=`cat $filename | awk -F ',' {'print $6'} | sed 's/[ \t]//g'`
			OS_VERSION=`cat $filename | awk -F ',' {'print $7'} | sed 's/[ \t]//g'`
			UPTIME=`cat $filename | awk -F ',' {'print $8'} | sed 's/[ \t]//g'`
			SYSTEM_TIME=`cat $filename | awk -F ',' {'print $9'} | sed 's/[ \t]//g'`
			CPU=`cat $filename | awk -F ',' {'print $10'} | sed 's/[ \t]//g'`
			MEMORY=`cat $filename | awk -F ',' {'print $11'} | sed 's/[ \t]//g'`
			SWAP=`cat $filename | awk -F ',' {'print $12'} | sed 's/[ \t]//g'`
			INODE=`cat $filename | awk -F ',' {'print $13'} | sed 's/[ \t]//g'`
			DISK=`cat $filename | awk -F ',' {'print $14'} | sed 's/[ \t]//g'`
			IO=`cat $filename | awk -F ',' {'print $15'} | sed 's/[ \t]//g'`
			USERS=`cat $filename | awk -F ',' {'print $16'} | sed 's/[ \t]//g'`
			DEFUNCT=`cat $filename | awk -F ',' {'print $17'} | sed 's/[ \t]//g'`
			NTP_STATUS=`cat $filename | awk -F ',' {'print $18'} | sed 's/[ \t]//g'`
			TIMEZONE=`cat $filename | awk -F ',' {'print $19'} | sed 's/[ \t]//g'`
			DSA_STATUS=`cat $filename | awk -F ',' {'print $20'} | sed 's/[ \t]//g'`
			#echo "HOSTNAME:$HOSTNAME"
			#echo "IPADDR:$IPADDR"
			#echo "CHECK_DATE:$CHECK_DATE"
			#echo "KERNEL:$KERNEL"
			#echo "SSH_VERSION:$SSH_VERSION"
			#echo "OPENSSL_VERSION:$OPENSSL_VERSION"
			#echo "OS_VERSION:$OS_VERSION"
			#echo "UPTIME:$UPTIME"
			#echo "SYSTEM_TIME:$SYSTEM_TIME"
			#echo "CPU:$CPU"
			#echo "MEMORY:$MEMORY"
			#echo "SWAP:$SWAP"
			#echo "INODE:$INODE"
			#echo "DISK:$DISK"
			#echo "IO:$IO"
			#echo "USERS:$USERS"
			#echo "DEFUNCT:$DEFUNCT"
			#echo "NTP_STATUS:$NTP_STATUS"
			#echo "TIMEZONE:$TIMEZONE"
			sed -i "$num"c" $HOSTNAME, $IPADDR, $CHECK_DATE, $KERNEL, $SSH_VERSION, $OPENSSL_VERSION, $OS_VERSION, $UPTIME, $SYSTEM_TIME, $CPU, $MEMORY, $SWAP, $DISK, $INODE, $IO, $USERS, $DEFUNCT, $NTP_STATUS, $TIMEZONE, $DSA_STATUS \n"  "$ansibleLinuxLogPath"/$(date +%y%m%d)/linuxHealthCheck_TOTAL_$(date +%y%m%d).csv
	done

	fi

}

main(){
	excuteUser
	mkdirLogPath
	copyLinuxShell
	excuteLinuxShell
	copyLogFileToLocal
	moveLogs
	generateTotalCsv
	exit 0
}


main

linuxHealthCheck.sh脚本

#!/bin/sh
###############################################################################################################
#Script name: linuxHealthCheck.sh
#Script description: for Linux Health Check
#Current Release Version: 1.0.0
#Script Owner: He ,Haibo
#Latest editor: He, Haibo
#Support platform:  Linux OS for redhat and centos.
#Date: 2022/4/20---first Version for linuxHealthCheck.sh.
#Change log:2022/4/22 add checkDsa
#Descript:sh linuxHealthCheck.sh
#
###############################################################################################################

export LANG=en_US
osbox=`uname`
RHversion=$(cat /proc/version | sed 's/[^0-9]//g' | cut -b -3)
IPADDR=`/sbin/ifconfig $(netstat -rn |grep ^0.0.0.0|awk '{print $NF}') |grep -w inet|awk '{print $2}'`

###打印日志函数
log_info(){
	DATE_N=`date "+%Y-%m-%d %H:%M:%S"`
	USER_N=`whoami`
	echo "${DATE_N} ${USER_N} execute $0 [INFO] $@"  
}

log_error(){
    DATE_N=`date "+%Y-%m-%d %H:%M:%S"`
    USER_N=`whoami`
    echo -e "\033[41;37m ${DATE_N} ${USER_N} execute $0 [ERROR] $@ \033[0m" 
}

fn_log(){
    if [[ $? -eq 0 ]];then
        log_info "$@ sucessed."
        echo -e "\033[32m $@ sucessed. \033[0m"
    else
        log_error "$@ failed."
        echo -e "\033[41;37m $@ failed. \033[0m"
    fi
}

checkExecuteUser(){
	uid=`id -u`
	if [[ $uid == 0 ]];then
		continue
	else
		fn_log "Current excute user is not root ,shell will exist."
		exit 1
	fi
}

###Create /tmp/log if not exist.
mkdirLogPath(){
    if [[ ! -d /tmp/log ]];then
        mkdir -p /tmp/log
        chmod 767 /tmp/log
    fi
    CheckLinuxLogPath="/tmp/log"
    CheckLinuxLog=`hostname`_${IPADDR}_linuxHealthCheck_$(date +%y%m%d).csv
}

###Get OS Arch Linux or not
getOsArch(){
	if [[ "$osbox" == "Linux" ]];then
		continue
	else
		fn_log "Current OS is $osbox,shell is exit now." >> "$CheckLinuxLogPath/$CheckLinuxLog" 2>&1
		echo 0
		exit 0
	fi
}

###Get redhat or centos
getOsCentosOrRedhat(){
	cat /proc/version | grep -iE "redhat|centos" > /dev/null
	if [[ $? == 0 ]];then
		continue
	else
		echo "Current OS is not centos or redhat." >> "$CheckLinuxLogPath/$CheckLinuxLog" 2>&1
		echo 1
		exit 1
	fi
}


osSystem(){
	OS_HOSTNAME=`hostname`
#	IPADDR=`ifconfig $(netstat -rn |grep ^0.0.0.0|awk '{print $NF}') |grep -w inet|awk '{print $2}'`
	CHECKTIME=`date "+%Y-%m-%d"`
    KERNEL=`uname -r`
	OPENSSH_VERSION=`ssh -V 2>&1|awk '{print $1}' | awk -F ',' {'print $1'}`
	OPENSSL_VERSION=`ssh -V 2>&1|awk '{print $3}'`
	OS_VERSION=`cat /etc/redhat-release | sed 's/[ \t]//g'`
	RUN_TIME=`uptime | awk -F ',' {'print $1'} | awk -F ' ' {'print $3,$4'} | sed 's/[ \t]//g'`
}


#硬件信息
cpuInfo(){
	cpu_idle=$(top -b -n 1 | grep Cpu | awk '{print $8}'| cut -f 1 -d "%" | sed 's/[ \t]//g' | sed "s/\..*//g")
	if [[ $cpu_idle -lt 20 ]]; then
		cpu_idle_status="Failed"
	else
		cpu_idle_status="Success"
	fi
}
##内存信息
memInfo(){
	mem_total=`free -m | grep Mem | awk '{print $2}'`
#	echo "内存总大小为:$mem_total"
	mem_used=`free -m | grep Mem | awk '{print $3}'`
#	echo "已使用内存为:$mem_used"
	mem_free=`free -m | grep Mem | awk '{print $4}'`
#	echo "剩余内存为:$mem_free"
	
	#百分比
	which bc > /dev/null 2>&1
	if [[ $? == 0 ]];then
		mem_used_percent=`echo "scale=2;$mem_used/$mem_total*100"|bc`
		mem_used_percent=`echo $mem_used_percent | awk -F '.' {'print $1'}`
#		echo "已使用内存百分比为:$mem_used_percent"
		if [[ $mem_used_percent -gt 80 ]];then
			mem_used_percent_status="Failed"
		else
			mem_used_percent_status="Success"
		fi
	fi
}

swapInfo(){
	swap_total=`free -m | grep Swap | awk '{print $2}'`
	swap_used=`free -m | grep Swap | awk '{print $3}'`
	which bc > /dev/null 2>&1
	if [[ $? == 0 ]];then
		swap_used_percent=`echo "scale=2;$swap_used/$swap_total*100"|bc`
		swap_used_percent=`echo $swap_used_percent | awk -F '.' {'print $1'}`
		if [[ $swap_used_percent -gt 80 ]];then
			swap_used_percent_status="Failed"
		else
			swap_used_percent_status="Success"
		fi
	fi
}

##get df -h infomation
dfInfo(){
	df_percent=`timeout -k 10 30 df -Th | awk -F '[ |%]+' 'NR>1 {if($6>80) print $6}'`
	if [[ $? == 0 ]];then
		if [[ "$df_percent" == "" ]];then
			df_percent_status="Success"
		else
			df_percent_status="Failed"
		fi
	else
		df_percent_status="Failed"	
	fi
}


checkIOInfo(){
	which iostat > /dev/null 2>&1
	if [[ $? == 0 ]];then
		IOPercent=`iostat | head -4  | awk 'END {print}' | awk -F ' ' {'print $6'} | sed 's/[ \t]//g'`
		IOPercent=`echo $IOPercent | awk -F '.' {'print $1'}`
		if [[ $IOPercent -lt 70 ]];then
			IOPercent_status="Failed"
		else
			IOPercent_status="Success"
		fi
	fi
}

userLogin(){
	userNum=`uptime | awk -F ', ' {'print $3'} | awk -F ' ' {'print $1'} | sed 's/[ \t]//g'`
	if [[ $userNum -gt 10 ]];then
		userNum_status="Failed"
	else
		userNum_status="Success"
	fi
}

checkDefunct(){
	DefunctNum=`ps -A -o stat,ppid,pid,cmd | grep -e '^[Zz]' | wc -l`
	if [[ $DefunctNum == 0 ]];then
		DefunctNum_status="Success"
	else
		DefunctNum_status="Failed"
	fi
}

checkNtp(){
	ntpstat > /dev/null 2>&1
	if [[ $? == 0 ]];then
		ntpstat_status="Success"
	else
		ntpstat_status="Failed"
	fi
}

checkZone(){
	ZoneName=`timedatectl | grep "Time zone" | awk -F ':' {'print $2'} | awk -F ' ' {'print $1'} | sed 's/[ \t]//g'`
	if [[ $ZoneName == "Asia/Shanghai" ]];then
		ZoneName_status="Success"
	else
		ZoneName_status="Failed"
	fi
}

##get df -i infomation
checkInode(){
    inode_percent=`timeout -k 10 30 df -Ti | awk -F '[ |%]+' 'NR>1 {if($6>80) print $6}'`
    if [[ $? == 0 ]];then
        if [[ "$inode_percent" == "" ]];then
                inode_status="Success"
        else
                inode_status="Failed"
        fi
	else
		inode_status="Failed"
    fi
}


checkDsa(){
	rpm -qa | grep -i "ds_agent" > /dev/null
	if [[ $? == 0 ]];then
		dsa_status="Success"
	else
		dsa_status="Failed"
	fi
}


getSystemTime(){
	DATE_N=`date "+%Y-%m-%d-%H:%M:%S"`
}

totaltocsv(){
	echo "${OS_HOSTNAME} , ${IPADDR} , ${CHECKTIME} , ${KERNEL} , ${OPENSSH_VERSION} , ${OPENSSL_VERSION} , ${OS_VERSION} , ${RUN_TIME} , ${DATE_N}, ${cpu_idle_status} , ${mem_used_percent_status} , ${swap_used_percent_status} , ${df_percent_status} ,${inode_status}, ${IOPercent_status} , ${userNum_status} , ${DefunctNum_status} , ${ntpstat_status} , ${ZoneName_status} , ${dsa_status}" > $CheckLinuxLogPath/$CheckLinuxLog

}

main(){
	getOsArch
	getOsCentosOrRedhat
	mkdirLogPath
	osSystem
	cpuInfo
	memInfo
	swapInfo
	dfInfo
	checkIOInfo
	userLogin
	checkDefunct
	checkNtp
	checkZone
	checkInode
	getSystemTime
	checkDsa
	totaltocsv
	exit 0
}

main

发送邮件脚本

#!/bin/sh
###############################################################################################################
#Script name: sendmail.sh
#Script description: for Linux Health Check
#Current Release Version: 1.0.0
#Script Owner: He ,Haibo
#Latest editor: He, Haibo
#Support platform:  Linux OS for redhat and centos.
#Date: 2022/4/20---first Version for sendmail.sh
#Change log:2022/4/22 add checkDsa
#Descript:sh sendmail.sh
#
###############################################################################################################
mailaddress="1111111@qq.com"

###send linuxHealthCheck_TOTAL_$(date +%y%m%d).csv to mail in $mailaddress
mutt -s "Linux Health Check" -a /home/sysinsp/LinuxCheck/logs/$(date +%y%m%d)/linuxHealthCheck_TOTAL_$(date +%y%m%d).csv -- ${mailaddress} < aa.txt