最近用shell写了个监控haproxy状态的脚本,记录一下以备后用。
1、首先明确需求。我们需要什么功能,常规方式是怎么实现的,使用脚本又该怎么实现。
只需要监控5个状态:当前连接数Session rate下的Cur;错误状态Errors下的三种状态Req,Conn,Resp;服务状态Status。状态页面如图:那么如何用脚本的形式把这些展示出来呢?重要的是,如何让nagios能够获取到这些数据?这就是接下来需要做的事情。
获取这些数据有两种方法:其一,通过wget或者curl来访问状态页面,然后筛选数据;其二,通过sock来获取状态,然后筛选。我先采用第二种方式来获取数据。
从sock获取数据,需要安装socat这个软件,具体使用省略。
echo "show stat" | socat /var/lib/haproxy/stats stdio
#/var/lib/haproxy/stats是haproxy路径
可以看到以csv格式的状态
pxname,svname,qcur,qmax,scur,smax,slim,stot,bin,bout,dreq,dresp,ereq,econ,eresp,wretr,wredis,status,weight,act,bck,chkfail,chkdown,lastchg,downtime,qlimit,pid,iid,sid,throttle,lbtot,tracked,type,rate,rate_lim,rate_max,check_status,check_code,check_duration,hrsp_1xx,hrsp_2xx,hrsp_3xx,hrsp_4xx,hrsp_5xx,hrsp_other,hanafail,req_rate,req_rate_max,req_tot,cli_abrt,srv_abrt,
admin_status,FRONTEND,,,1,1,20480,31,12757,354568,0,0,0,,,,,OPEN,,,,,,,,,1,1,0,,,,0,1,0,10,,,,0,29,1,0,0,0,,1,10,31,,,
admin_status,BACKEND,0,0,0,0,20480,0,12757,354568,0,0,,0,0,0,0,UP,0,0,0,,0,773,0,,1,1,0,,0,,1,0,,0,,,,0,0,0,0,0,0,,,,,0,0,
http_80_in,FRONTEND,,,0,1,20480,4,348,848,0,0,0,,,,,OPEN,,,,,,,,,1,2,0,,,,0,0,0,1,,,,0,0,0,0,4,0,,0,1,4,,,
web,web1,0,0,0,0,,0,0,0,,0,,0,0,0,0,UP,1,1,0,0,0,773,0,,1,3,1,,0,,2,0,,0,L7OK,200,0,0,0,0,0,0,0,0,,,,0,0,
web,web2,0,0,0,0,,0,0,0,,0,,0,0,0,0,DOWN,2,1,0,0,1,771,771,,1,3,2,,0,,2,0,,0,L4TOUT,,1502,0,0,0,0,0,0,0,,,,0,0,
web,BACKEND,0,0,0,0,0,0,0,0,0,0,,0,0,0,0,UP,1,1,0,,0,773,0,,1,3,0,,0,,1,0,,0,,,,0,0,0,0,0,0,,,,,0,0,
需要的数据:当起连接数rate在第34域;err相关的ereq、econ、eresp在第13、14、15域;状态status在第18域。用awk筛选数据:
rate=`echo "show stat" | socat /var/lib/haproxy/stats stdio | awk -F, '$2=="web1" {print $34}'`
ereq=`echo "show stat" | socat /var/lib/haproxy/stats stdio | awk -F, '$2=="web1" {print $34}'`
econ=`echo "show stat" | socat /var/lib/haproxy/stats stdio | awk -F, '$2=="web1" {print $34}'`
eresp=`echo "show stat" | socat /var/lib/haproxy/stats stdio | awk -F, '$2=="web1" {print $34}'`
status=`echo "show stat" | socat /var/lib/haproxy/stats stdio | awk -F, '$2=="web1" {print $34}'`
数据现在已经可以获取了,接下来就要让nagios能够接受到数据。
2、了解插件规范。
nagios的插件规范有两点:
(1)、需要一个返回值来确定状态
0 | OK |
1 | WARNING |
2 | CRITICAL |
3 | UNKNOWN |
(2)、在报警处输出定义的说明,默认大小4K,见红框位置:了解了编写规范,现在就开始写脚本。
3、脚本实例
#########################################################################
# File Name: check_haproxy.sh
# Author: jc
# Created Time: 2014年03月01日 星期六 14时36分23秒
#########################################################################
#!/bin/bash
#定义返回状态
ST_OK=0
ST_WR=1
ST_CR=2
ST_UK=3
#sock的默认路径
sock_path=/var/lib/haproxy/stats
#默认检查sock是否存在
#sock_check=1
#帮助信息函数
print_help() {
echo " --sock|-s)"
echo " haproxy的sock路径,默认路径:/var/lib/haproxy/stats"
echo " -m/--mode)"
echo " haproxy的状态项,目前只支持:rate,ereq,econ,eresp,status"
# echo " -n/--no-check-sock)"
# echo " 检查sock存在与否,1检查,0不检查"
exit $ST_UK
}
#获取输入的选项
while test -n "$1"; do
case "$1" in
-help|-h)
print_help
exit $ST_UK
;;
--sock|-s)
sock=$2
shift
;;
--mode|-m)
mode=$2
shift
;;
--no-sock-check|-n)
sock_check=0
;;
--hostname|-H)
hostname=$2
shift
;;
--warning|-w)
warning=$2
shift
;;
--critical|-c)
critical=$2
shift
;;
*)
echo "Unknown argument: $1"
print_help
exit $ST_UK
;;
esac
shift
done
#sock检测
#check_sock() {
#echo "show stat" | socat /var/lib/haproxy/stats stdio >/tmp/hap_sta.txt
#if [ -s "/tmp/hap_sta.txt" ]
#then
# con=1
#else
# con=2
#fi
#}
#获取数据
get_val() {
case $mode in
rate)
val=`echo "show stat" | socat $sock_path stdio | awk -F, '$2=="web1" {print $34}'`
;;
ereq)
val=`echo "show stat" | socat $sock_path stdio | awk -F, '$2=="web1" {print $13}'`
if [ -z $val ]
then
val=0
fi
;;
econ)
val=`echo "show stat" | socat $sock_path stdio | awk -F, '$2=="web1" {print $14}'`
;;
eresp)
val=`echo "show stat" | socat $sock_path stdio | awk -F, '$2=="web1" {print $15}'`
;;
status)
val=`echo "show stat" | socat $sock_path stdio | awk -F, '$2=="web1" {print $18}'`
;;
*)
echo "暂时不能检测此项,请重新输入"
;;
esac
}
#显示的文字信息
out_pr() {
output="haproxy is running. $mode's key is $val"
}
#开始执行
##检查sock是否能够连上
#if [ $sock_check = 1 ]
#then
# check_sock
# if [ "$con" = 2 ]
# then
# echo "ERR!!连接haproxy被拒绝,请确认服务已经开启,或者检查sock路径/权限是否正确!"
# exit $ST_CR
# fi
#fi
get_val
out_pr
#get_stat() {
if [ "$val" -ge "$warning" ] && [ "$val" -lt "$critical" ]
then
echo "WARNING - $output"
exit $ST_WR
elif [ "$val" -ge "$critical" ]
then
echo "CRITICAL - $output"
exit $ST_CR
else
echo "OK - $output"
exit $ST_OK
fi
脚本完成,先在命令行测试:
[root@webtest-250 libexec]# /opt/nagios/libexec/check_haproxy.sh -m rate -w 10 -c 20
OK - haproxy is running. rate's key is 0
注意,以下两步相当重要,如果不给nagios用户添加权限,在页面会看不到值,如(1)、更改/etc/sudoers,把nagios启动用户添加进去
nagios ALL=(ALL) NOPASSWD:/usr/local/nagios/libexec/check_hap.sh
(2)、更改/etc/sudoers把这一行注释,否则在后台运行sudo会报错
#Defaults requiretty
现在切换到nagios的用户执行就正确了。
4、添加插件至nagios
(1)、修改commands.cfg,添加
##haproxy
define command{
command_name check_haproxy
command_line /opt/nagios/libexec/check_haproxy.sh -s $ARG1$ -m $ARG2$ -w $ARG3$ -C $ARG4$
}
(2)、修改服务配置文件,我是在本机监控故只改localhost.cfg,添加以下监控
##haproxy check
define service{
use local-service ; Name of service template to use
host_name localhost
service_description haproxy-Session-cur
check_command check_haproxy!/var/lib/haproxy/stats!rate!10!20
notifications_enabled 0
}
define service{
use local-service ; Name of service template to use
host_name localhost
service_description haproxy-Errors-req
check_command check_haproxy!/var/lib/haproxy/stats!ereq!3!5
notifications_enabled 0
}
define service{
use local-service ; Name of service template to use
host_name localhost
service_description haproxy-Errors-con
check_command check_haproxy!/var/lib/haproxy/stats!econ!3!5
notifications_enabled 0
}
define service{
use local-service ; Name of service template to use
host_name localhost
service_description haproxy-Errors-resp
check_command check_haproxy!/var/lib/haproxy/stats!eresp!3!5
notifications_enabled 0
}
(3)、添加进nrpe.cfg,这些报警阈值是为了方便测试设得比较小,根据实际应用更改
command[check_haproxy]=/usr/bin/sudo /opt/nagios/libexec/check_haproxy.sh -s /var/lib/haproxy/stats -m rate --warning 10 --critical 20
command[check_haproxy]=/usr/bin/sudo /opt/nagios/libexec/check_haproxy.sh -s /var/lib/haproxy/stats -m ereq --warning 3 --critical 5
command[check_haproxy]=/usr/bin/sudo /opt/nagios/libexec/check_haproxy.sh -s /var/lib/haproxy/stats -m econ --warning 3 --critical 5
command[check_haproxy]=/usr/bin/sudo /opt/nagios/libexec/check_haproxy.sh -s /var/lib/haproxy/stats -m eresp --warning 3 --critical 5
重启nagios服务,登陆web界面,已经可以看到监控项了:现在来测试下是否能够正常报警,使用siege测试:
./bin/siege -c 50 -n 50 -t 30 http://192.168.1.250/stat.php
,等几分钟看界面,已经有报警了:
完毕
附:通过curl访问haproxy状态页面来获取数据脚本:
#########################################################################
# File Name: check_haproxy-url.sh
# Author: jc
# Created Time: 2014年03月01日 星期六 17时27分03秒
#########################################################################
#!/bin/bash
ST_OK=0
ST_WR=1
ST_CR=2
ST_UK=3
url_check=1
print_help() {
echo " --url|-u)"
echo " haproxy监控页面的链接"
echo " -m/--mode)"
echo " haproxy的状态项,如:rate,ereq,econ,eresp,status"
echo " -n/--no-url-check|)"
echo " 是否检查haproxy的url:1检测(默认),0不检查"
exit $ST_UK
}
while test -n "$1"; do
case "$1" in
-help|-h)
print_help
exit $ST_UK
;;
--mode|-m)
mode=$2
shift
;;
--no-url-check|-n)
url_check=0
;;
--url|-u)
hap_url=$2
shift
;;
--warning|-w)
warning=$2
shift
;;
--critical|-c)
critical=$2
shift
;;
*)
echo "Unknown argument: $1"
print_help
exit $ST_UK
;;
esac
shift
done
#check_url() {
#curl -o /tmp/stats.csv $url >/dev/null 2>&1
#if [ -s stats.csv ]
#then
#con=1
#else
#con=2
#fi
#}
curl -o /tmp/stats.csv $hap_url >/dev/null 2>&1
ch_val() {
if [ -z $val ]
then
val=0
else
val=$val
fi
}
get_val() {
case $mode in
rate)
val=`cat /tmp/stats.csv | grep web1 | awk -F '<tr class=' '{print $3}' | awk -F '</td><td>' '{print $5}'`
ch_val
;;
ereq)
val=`cat /tmp/stats.csv | grep web1 | awk -F '<tr class=' '{print $3}' | awk -F '</td><td>' '{print $16}'`
ch_val
;;
econ)
val=`cat /tmp/stats.csv | grep web1 | awk -F '<tr class=' '{print $3}' | awk -F '</td><td>' '{print $17}'| awk -F '<' '{print $1}'`
ch_val
;;
eresp)
val=`cat /tmp/stats.csv | grep web1 | awk -F '<tr class=' '{print $3}' | awk -F '</td><td>' '{print $18}'`
ch_val
;;
# status)
# val=`cat /tmp/stats.csv | grep web1 | awk -F '<tr class=' '{print $3}' | awk -F '</td><td>' '{print $21}'`
# ch_val
# ;;
*)
echo "mode输入错误,请重新输入"
;;
esac
}
out_pr() {
output="haproxy is running. $mode: $val"
}
get_stat() {
if [ $val -ge $warning ] && [ $val -lt $critical ]
then
echo "WARNING - ${output}"
exit $ST_WR
elif [ "$val" -ge "$critical" ]
then
echo "CRITICAL - ${output}"
exit $ST_CR
else
echo "OK - ${output}"
exit $ST_OK
fi
}
#if [ $url_check -eq 1 ]
#then
# check_url
# if [ "$con" -eq 2 ]
# then
# echo "ERR!!无法访问haproxy的状态页面,请确认url输入正确!"
# exit $ST_CR
# fi
#fi
get_val
out_pr
get_stat
亿速云「云服务器」,即开即用、新一代英特尔至强铂金CPU、三副本存储NVMe SSD云盘,价格低至29元/月。点击查看>>
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。