zabbix 监控 redis

Posted 2020-09-20

通过redis自带的info命令来监控redis的健康状态，通过redis-cli PING命令来监控redis的存活状态。

附件中有监控模板，将监控脚本放在redis服务器的自定义的/scripts/zabbix_redis/下:

#! /bin/bash
#Name: redismontior.sh
REDISCLI="/usr/bin/redis-cli"
HOST="127.0.0.1"
PORT=6379
if [[ $# == 1 ]];then
    case $1 in
        version)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "redis_version" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        uptime)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "uptime_in_seconds" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        connected_clients)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "connected_clients" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        blocked_clients)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "blocked_clients" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_memory)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_memory" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_memory_rss)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_memory_rss" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_memory_peak)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_memory_peak" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_memory_lua)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_memory_lua" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_cpu_sys)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_cpu_sys" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_cpu_user)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_cpu_user" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_cpu_sys_children)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_cpu_sys_children" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        used_cpu_user_children)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "used_cpu_user_children" | awk -F‘:‘ ‘{print $2}‘`
            echo $result
        ;;
        rdb_last_bgsave_status)
            result=`$REDISCLI -h $HOST -p $PORT info  | grep -w "rdb_last_bgsave_status" | awk -F‘:‘ ‘{print $2}‘ | grep -c ok`
            echo $result
        ;;
        aof_last_bgrewrite_status)
            result=`$REDISCLI -h $HOST -p $PORT info  | grep -w "aof_last_bgrewrite_status" | awk -F‘:‘ ‘{print $2}‘ | grep -c ok`
            echo $result
        ;;
        aof_last_write_status)
            result=`$REDISCLI -h $HOST -p $PORT info  | grep -w "aof_last_write_status" | awk -F‘:‘ ‘{print $2}‘ | grep -c ok`
            echo $result
        ;;
        *)
            echo -e "\033[33mUsage: $0 {connected_clients|blocked_clients|used_memory|used_memory_rss|used_memory_peak|used_memory_lua|used_cpu_sys|used_cpu_user|used_cpu_sys_children|used_cpu_user_children|rdb_last_bgsave_status|aof_last_bgrewrite_status|aof_last_write_status}\033[0m" 
        ;;
    esac
elif [[ $# == 2 ]];then
    case $2 in
        keys)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "$1" | grep -w "keys" | awk -F‘=|,‘ ‘{print $2}‘`
            echo $result
        ;;
        expires)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "$1" | grep -w "keys" | awk -F‘=|,‘ ‘{print $4}‘`
            echo $result
        ;;
        avg_ttl)
            result=`$REDISCLI -h $HOST -p $PORT info | grep -w "$1" | grep -w "avg_ttl" | awk -F‘=|,‘ ‘{print $6}‘`
            echo $result
        ;;
        *)
            echo -e "\033[33mUsage: $0 {db0 keys|db0 expires|db0 avg_ttl}\033[0m" 
        ;;
    esac
fi

zabbix参数key的设置：

# cat /etc/zabbix/zabbix_agentd.d/userparameter_redis.conf 

UserParameter=Redis.Status, redis-cli PING|grep -c "PONG"
UserParameter=Redis.Info[*], /bin/sh /scripts/zabbix_redis/zabbix_redis.sh $1

模板见附件.其中对脚本传入两个参数的部分，对于不同的环境可能会不太兼容，可以对模板和脚本进行修改，或者直接在模板中禁用对应的item。