【故障案例】一个Zabbix5.0 与Percona监控数据库的故障-ERROR: run the command manually to investigate the problem

背景

• 早上有个,刚刚入职的小伙伴,找李导提问,说Zabbix监控MySQL要用percona
• 于是有了下面的故障
• 故障环境是: CentOS 7.6,Zabbix是5.0
• OK, Let's start .

流程

• zabbix环境是OK的,目前主要问题是
• 运行percona脚本的时候报错

/var/lib/zabbix/percona/scripts/get_mysql_stats_wrapper.sh gg
ERROR: run the command manually to investigate the problem: /usr/bin/php -q /var/lib/zabbix/percona/scripts/ss_get_mysql_stats.php --host 127.0.0.1 --items gg

• 但是直接执行里面的php脚本是不报错的

/usr/bin/php -q /var/lib/zabbix/percona/scripts/ss_get_mysql_stats.php --host 127.0.0.1 --items gg

解决

• 定位问题,在脚本上面继续排查
• 问题为php生产的临时文件多个数字(随机)
• 脚本中并没有对应修改,

cat  /var/lib/zabbix/percona/scripts/get_mysql_stats_wrapper.sh 
#!/bin/sh
# The wrapper for Cacti PHP script.
# It runs the script every 5 min. and parses the cache file on each following run.
# Version: 1.1.8
#
# This program is part of Percona Monitoring Plugins
# License: GPL License (see COPYING)
# Copyright: 2018 Percona
# Authors: Roman Vynar

ITEM=$1
HOST=127.0.0.1
DIR=`dirname $0`
CMD="/usr/bin/php -q $DIR/ss_get_mysql_stats.php --host $HOST --items gg"
CACHEFILE="/tmp/${HOST}-mysql_cacti_stats.txt"

if [ "$ITEM" = "running-slave" ]; then
    # Check for running slave
    RES=`HOME=~zabbix mysql -e 'SHOW SLAVE STATUS\G' | egrep '(Slave_IO_Running|Slave_SQL_Running):' | awk -F: '{print $2}' | tr '\n' ','`
    if [ "$RES" = " Yes, Yes," ]; then
        echo 1
    else
        echo 0
    fi
    exit
elif [ -e ${CACHEFILE}* ]; then
    # Check and run the script
    TIMEFLM=`stat -c %Y ${CACHEFILE}*`
    TIMENOW=`date +%s`
    if [ `expr $TIMENOW - $TIMEFLM` -gt 300 ]; then
        rm -f $CACHEFILE
        $CMD 2>&1 > /dev/null
    fi
else
    $CMD 2>&1 > /dev/null
fi

# Parse cache file
if [ -e ${CACHEFILE}* ]; then
    cat ${CACHEFILE}* | sed 's/ /\n/g; s/-1/0/g'| grep $ITEM | awk -F: '{print $2}'
else
    echo "ERROR: run the command manually to investigate the problem: $CMD"
fi

总结

• 平时多多练习下shell编程,对于这种故障难度不大.
• 有各种问题欢迎留言与交流