How to use multi-instance monitoring mysql zabbix

agent played on more than mysql instance, takes a different port, agent only in the initial condition, into the script and key configuration, and then reboot. After the maintenance time (mysql port change), to do not move agent, and strive to end or only at the web server end can automatically make changes to the corresponding monitor port.

The main tool used: macro (macro template into the host macro, macro auto-discovery)

      Host macro format: $ MACRO (can be directly fill value) 
           automatically discovers the macro: #MACRO (need to cooperate to create a key value)

Broad framework

1.agent host - agent host to create a macro variable $ MYSQLPORT, the macro variable is Discovery rules of automatic discovery script parameters
2. Templates - template to create automatic discovery rules, relying on the automatic discovery script, macro variables get automatic discovery .
Item 3. Monitoring - Monitoring item created in the Discovery rules, the monitoring of key names and key values in the macro variables are automatically discover, monitor will automatically generate the relevant ports of entry
4. Create a client script and keys, to support the entire infrastructure running

Build step

1. Create a template

How multi-instance How to use multi-instance zabbix monitoring mysql monitoring mysql zabbix

Because this template is a clone of the default template mysql monitor, you can ignore items, triggers. Discovery rules are monitored item inside.

2. Add the template automatically discovers the macro (macro declares automatic discovery is under way to map the script, there is only a fixed format script generation, zabbix will be found that he was automatically discovers the macro)

The figure in the mysql.discovery is to get automatic discovery macro script, the script is on the agent side.

For this reason behind the $ MYSQLPORT execution of the script with the parameters of the time, $ MYSQLPORT is the host macro.

discovery rule 右边的 filters 是过滤规则的意思，比方你只想取脚本里的某某自动发现宏，就可以在里面设置。这里就只有一个自动发现宏，所有不做配置。

3.创建监控项（监控项和监控的脚本都是要改的，因为自动发现功能，至少脚本会多了一个参数）

4.创建主机宏

点开主机，添加主机宏，宏名称是自动发现脚本里的参数，宏的值就是此主机上需要监控的端口。

这样就实现了只要改动主机宏的值，就可以改变主机宏的监控项。

脚本展示

1.discovery_mysql.sh 自动发现端口脚本 (网上抄的脚本做了修改)

 res=`echo $1| sed "s/_/\n/g"`;
port=($res)
printf '{\n' printf '\t"data":[\n' for key in ${!port[@]} do if [[ "${#port[@]}" -gt 1 && "${key}" -ne "$((${#port[@]}-1))" ]]; then printf '\t {\n' printf "\t\t\t\"{#MYSQLPORT}\":\"${port[${key}]}\"},\n" else [[ "${key}" -eq "((${#port[@]}-1))" ]] printf '\t {\n' printf "\t\t\t\"{#MYSQLPORT}\":\"${port[${key}]}\"}\n" fi done printf '\t ]\n' printf '}\n'

执行实例：以_ 为分隔符，格式化输出自动发现宏
./discovery_mysql.sh 3306_3307_3308

{
              "data":[
              {
                                "{#MYSQLPORT}":"3306"},
              {
                                "{#MYSQLPORT}":"3307"},
              {
                                "{#MYSQLPORT}":"3308"} ] }

2.mysql_filestype.sh 监控mysql的一些挂载盘

var=$1
MYSQL_PORT=$2
MYSQL_NAME=`ps -ef |grep '/mysql/app/bin/mysqld'| grep "$MYSQL_PORT" |grep -v grep | awk -F" " '{print $11}' |awk -F"/" '{print $2}'` ; MYSQL_SOCk_DIR="/$MYSQL_NAME/" ; df -h "${MYSQL_SOCk_DIR}${var}" |grep -v Filesystem |awk -F" " '{print $5}' | awk -F"%" '{print $1}' ;

3.mysql_ping.sh 监控mysql状态

MYSQL_PORT=$1;

[ "${MYSQL_USER}"     = '' ] &&  MYSQL_USER=xxx
[ "${MYSQL_PASSWORD}" = '' ] && MYSQL_PASSWORD=xxxxx mysqladmin=/mysql/app/bin/mysqladmin ; MYSQL_SOCk_DIR=`ps -ef |grep 'mysql'| grep "${MYSQL_PORT}" |grep -v 'grep' | awk -F" " '{print $16}'` ; ${mysqladmin} -u${MYSQL_USER} -p${MYSQL_PASSWORD} ${MYSQL_SOCk_DIR} ping 2> /dev/null |grep 'alive'|wc -l ;

注意: 这边连接实例都是 mysql -u -p -S /xxx/mysql.sock

但是后来发现部分客户端总是无法正确显示值，而且就是这条语句报错，感觉是环境变量的问题

所以拿 --socket=/xxx/mysql.sock 来代替 -S/xxx/mysql.sock 。这里的变量${MYSQL_SOCk_DIR} 就是--socket=/xxx/mysql.sock 。

4.mysql_repl.sh mysql主从状态监控

var=$1
MYSQL_PORT=$2
MYSQL_SOCk_DIR=`ps -ef |grep 'mysql'| grep "${MYSQL_PORT}" |grep -v 'grep' | awk -F" " '{print $16}'` ; mysql=/mysql/app/bin/mysql [ "${MYSQL_USER}" = '' ] && MYSQL_USER=xxx [ "${MYSQL_PASSWORD}" = '' ] && MYSQL_PASSWORD=xxxx ${mysql} -u${MYSQL_USER} -p${MYSQL_PASSWORD} ${MYSQL_SOCk_DIR} -e "show slave status\G;" 2> /dev/null |grep "\b${var}\b"|awk -F" " '{print $2}' ;

主从状态有几个端口的值是空的，这个得具体问题具体设置

5.mysql_status2 mysql性能之类的监控

mysql=/mysql/app/bin/mysql
var=$1
MYSQL_PORT=$2
[ "${MYSQL_USER}"     = '' ] &&  MYSQL_USER=xxx
[ "${MYSQL_PASSWORD}" = '' ] && MYSQL_PASSWORD=xxxxx MYSQL_SOCk_DIR=`ps -ef |grep 'mysql'| grep "${MYSQL_PORT}" |grep -v 'grep' | awk -F" " '{print $16}'` ; ${mysql} -u${MYSQL_USER} -p${MYSQL_PASSWORD} ${MYSQL_SOCk_DIR} -e "show global status;" 2> /dev/null |grep -v Variable_name|grep "\b${var}\b"|awk '{print $2}' ;

发现有些监控项或因为数字过长而无法显示，待解决

6.mysql_version.sh mysql版本监控

MYSQL_PORT=$1;

MYSQL_SOCk_DIR=`ps -ef |grep 'mysql'| grep "${MYSQL_PORT}" |grep -v 'grep' | awk -F" " '{print $16}'` ; mysql=/mysql/app/bin/mysql ; [ "${MYSQL_USER}" = '' ] && MYSQL_USER=xxx [ "${MYSQL_PASSWORD}" = '' ] && MYSQL_PASSWORD=xxxx ${mysql} -u${MYSQL_USER} -p${MYSQL_PASSWORD} ${MYSQL_SOCk_DIR} -e "select version();" 2> /dev/null |awk 'END {print}'

7.mysql_status_many.conf 生成可用键配置文件

UserParameter=mysql.discovery[*],/patrol/zabbix/bin/duoshili_discovery/discovery_mysql.sh $1  
UserParameter=mysql.status_many[*],/patrol/zabbix/bin/duoshili_discovery/mysql_status2 $1 $2
UserParameter=mysql.ping_many[*],/patrol/zabbix/bin/duoshili_discovery/mysql_ping.sh $1 UserParameter=mysql.version_many[*],/patrol/zabbix/bin/duoshili_discovery/mysql_version.sh $1 UserParameter=mysqlcheck_repl[*],/patrol/zabbix/bin/duoshili_discovery/mysql_repl.sh $1 $2 UserParameter=mysql.filestyle_many[*],/patrol/zabbix/bin/duoshili_discovery/mysql_filestype

How to use multi-instance monitoring mysql zabbix

Guess you like