http://hi.baidu.com/higkoo/blog/item/1a7aedb1fed499580823021d.html
市面上Memcached监控工具好多，譬如：
1、memcached.php
2、memcached-tool：
3、Brutis：
4、像Cacti 之类大型的监控系统也能支持：
5、telnet
外表在华丽，最终都是从Memcached的stats里拿到的数据，然后包装一下（包装其实也很重要！）。
最原始的方法是使用Telnet连接：
譬如：telnet 192.168.10.31 11211
然后输入stats回车，会拿到类似如下的信息：
STAT pid 3306
STAT uptime 80456
STAT time 1274499768
STAT version 1.4.0
STAT pointer_size 64
STAT rusage_user 247.557365
STAT rusage_system 381.562993
STAT curr_connections 12
STAT total_connections 495
STAT connection_structures 17
STAT cmd_get 28902205
STAT cmd_set 30107902
STAT cmd_flush 0
STAT get_hits 6139097
STAT get_misses 22763108
STAT delete_misses 0
STAT delete_hits 0
STAT incr_misses 0
STAT incr_hits 0
STAT decr_misses 0
STAT decr_hits 0
STAT cas_misses 0
STAT cas_hits 0
STAT cas_badval 0
STAT bytes_read 2492211224
STAT bytes_written 558817585
STAT limit_maxbytes 107374182
STAT accepting_conns 1
STAT listen_disabled_num 0
STAT threads 5
STAT conn_yields 529550
STAT bytes 183861479
STAT curr_items 1814816
STAT total_items 30107902
STAT evictions 0
END
这里介绍Bash Shel版本的监控方法：
一、原生态，绝大多数Linux/Unix都支持的方法：
#!/bin/bash# 文件名: monitormem.sh
# 用法: monitormem.sh 192.168.10.31 11211
exec 5<> /dev/tcp/$1/$2
if[$? -eq 0 ]; then
echo"stats" >&5
echo"quit" >&5
while read -u 5 -d $'\r' stat name value;
do
echo$value
done
exit 0
fi
exit 1
二、需要netcat支持，脚本稍简单一点，原理都是一样的：
#!/bin/bash# 文件名: monitormem.sh
# 用法: monitormem.sh 192.168.10.31 11211
echo -e "stats\nquit"|nc $1$2 | while read -d $'\r' stat name value;
do
echo$value
done
Memcached-Tool，也挺好的：
简单的状态信息：
# Item_Size Max_age 1MB_pages Count Full?1 96 B 3969 s 2 16384 no
2 120 B 77827 s 284 1798432 no
详细信息如下：
#192.168.10.31:11211 Field Valueaccepting_conns 1
bytes 183861479
bytes_read 2492211350
bytes_written 558827791
cas_badval 0
cas_hits 0
cas_misses 0
cmd_flush 0
cmd_get 28902205
cmd_set 30107902
conn_yields 529550
connection_structures 17
curr_connections 13
curr_items 1814816
decr_hits 0
decr_misses 0
delete_hits 0
delete_misses 0
evictions 0
get_hits 6139097
get_misses 22763108
incr_hits 0
incr_misses 0
limit_maxbytes 10737418240
listen_disabled_num 0
pid 3406
pointer_size 64
rusage_system 381.562993
rusage_user 247.557365
threads 5
time 1274500838
total_connections 499
total_items 30107902
uptime 81526
version 1.4.0
Brutis的最大特点是Ops/sec等指标是以秒为单位进行差量计算得到，其它如Memcached.php是累积数据。截图：
Brutis 2 Connections
Type Ops/sec Hits/sec Misses/sec Fails/sec Latency MB/sec
-------------------------------------------------------------------------------
Sets: 119.00 --- --- 0.00 0.000797 0.03
Gets: 1,189.98 208.60 981.38 0.00 0.000723 0.05
Totals: 1,308.98 208.60 981.38 0.00 0.000760 0.08
-------------------------------------------------------------------------------
SETS GETS
-------------------------------------------------------------------------------
Host Ops/sec Latency MB/sec Ops/sec Latency MB/sec
-------------------------------------------------------------------------------
higkoo.net 119.00 0.000797 0.03 1,189.98 0.000723 3.05
memcached.php截图：
OK，这里我们来写一个简单的监控Memcached点击率的Shell脚本：
#!/bin/baship=192.168.10.31
get_hits(){
exec 5<> /dev/tcp/$ip/11211
if [ $? -eq 0 ]; then
echo "stats" >&5
echo "quit" >&5
while read -u 5 -d $'\r' stat name value;
do
if [ $name == 'get_hits' ];then
echo $value
break
fi
done
fi
}
do
start=`get_hits`
sleep 1
end=get_hits`
echo " $end - $start " | bc
done