redis bigkey检查

/apps/svr/redis-2.8.19/bin/redis-cli -p 6288 --bigkeys
从库直接执行，后台是使用scan命令的

# /apps/svr/redis/bin/redis-cli -p 6382 --bigkeys -i 0.1

# Scanning the entire keyspace to find biggest keys as well as
# average sizes per key type.  You can use -i 0.1 to sleep 0.1 sec
# per 100 SCAN commands (not usually needed).

[00.00%] Biggest hash   found so far 'user:001' with 2 fields
[00.00%] Biggest string found so far 'k4' with 1 bytes
[00.00%] Biggest string found so far 'k1' with 3 bytes

-------- summary -------

Sampled 6 keys in the keyspace!
Total key length in bytes is 18 (avg len 3.00)

Biggest string found 'k1' has 3 bytes
Biggest   hash found 'user:001' has 2 fields

5 strings with 7 bytes (83.33% of keys, avg size 1.40)
0 lists with 0 items (00.00% of keys, avg size 0.00)
0 sets with 0 members (00.00% of keys, avg size 0.00)
1 hashs with 2 fields (16.67% of keys, avg size 2.00)
0 zsets with 0 members (00.00% of keys, avg size 0.00)

 案例:取样查看，较快（在从库查看！）

/apps/svr/redis-2.8.19/bin/redis-cli  -h  10.20.54.17 -p 6379  --bigkeys 

从中可以看到BaseItem201601和BaseItem201512所占用的内存是最多的，所有有充分的理由怀疑这两个key不正常，但是这两个key具体有多大？有几种方式

1、投机取巧法：将该库的rdb文件在一个测试机器上恢复过来，然后删掉BaseItem201601看内存减少多少，减少的内存即为该KEY大小

2、初略估算法：可以先取出Hash表中的一个元素

至文本文件中，可以得到一个文件的元素大小为，而该列表总共有3056237个元素，所以可以初略估算出总大小为11G左右

3、工具法：使用redis-memory-for-key工具，可以计算单独一个Key 的大小，但是线上没有安装此工具。需要自行安装

 

全扫操作，较慢（在key数量特别多或者特别大的时候运行会比较慢,建议在从库跑,如果存在Hash列表等情况，不一定能正确反映内存的实际大小情况）

#!/usr/bin/env bash
export PATH=/apps/svr/redis-2.8.17/bin/:$PATH
# This script prints out all of your Redis keys and their size in a human readable format
# Copyright 2013 Brent O'Connor
# License: http://www.apache.org/licenses/LICENSE-2.0
human_size() {
        awk -v sum="$1" ' BEGIN {hum[1024^3]="Gb"; hum[1024^2]="Mb"; hum[1024]="Kb"; for (x=1024^3; x>=1024; x/=1024) { if (sum>=x) { printf "%.2f %s\n",sum/x,hum[x]; break; } } if (sum<1024) print "1kb"; } '
}
redis_cmd='redis-cli'
# get keys and sizes
for k in `$redis_cmd keys "*"`; do key_size_bytes=`$redis_cmd debug object $k | perl -wpe 's/^.+serializedlength:([\d]+).+$/$1/g'`; size_key_list="$size_key_list$key_size_bytes $k\n"; done
# sort the list
sorted_key_list=`echo -e "$size_key_list" | sort -n`
# print out the list with human readable sizes
echo -e "$sorted_key_list" | while read l; do
    if [[ -n "$l" ]]; then
        size=`echo $l | perl -wpe 's/^(\d+).+/$1/g'`; hsize=`human_size "$size"`; key=`echo $l | perl -wpe 's/^\d+(.+)/$1/g'`; printf "%-10s%s\n" "$hsize" "$key";
    fi
done