破解故事——我是如何破解1亿2200万SHA1和Md5哈希密码
原文：http://blog.thireus.com/cracking-story-how-i-cracked-over-122-million-sha1-and-md5-hashed-passwords/
这是关于我如何破解密码的哈希值1亿2200万* John the Ripper和oclHashcat-plus故事。
这是几个月前，当我看到一条推特（https://twitter.com/m3g9tr0n）从korelogic约含共1亿4600万个密码的密码哈希表的torrent文件（https://www.korelogic.com/InfoSecSouthwest2012_Ripe_Hashes.html）。这个非常大的数量的密码哈希起初让我很沮丧，因为我自己只有一台经典配置的AMD Phenom II 4核3.2 MHz的电脑,外加一个ATI/AMD 5770显卡。但我真的很想给尝试一下，因为该领域的密码破解让我着迷。
在这漫长的旅程中我用的密码破解工具是John the Ripper和oclHashcat-puls。这篇文章是关于破解korelogic提供的MD5哈希值，但同样的策略也适用于SHA1哈希破解。
更新：
08/29/2012 –新的例子在John the Ripper部分: “在dict2hash.pl脚本的帮助下破解双MD5密码值“
08/29/2012 – 所有的都进行过排序和整理新的版本下载。
处理哈希值：
首先，所有的korelogic torrent文件必须被解压缩，它包含一个命名为“校验”文件夹。让我们检查这个文件夹的内容…
root@m3g9tr0n:~/hashes$ ls
longer_salts raw-md5.hashes.txt salted_with_md5 SHA1 vBulletin-v3.8.4
从现在起我们集中在raw-md5.hashes.txt文件列表，这个文件是4.3GB，通过wc工具可以看出它包括139444502行。
root@m3g9tr0n:~/hashes$ wc -lraw-md5.hashes.txt
139444502 raw-md5.hashes.txt
正如你可以假设，无论是John the Ripper和oclHashcat-plus都无法加载该文件，因为它太大了。出于这个原因，我们需要分割该文件。在Linux下，我们有一个很好的工具称为split 做这件工作得很好：
root@m3g9tr0n:~$ split –help
Usage: split [OPTION]… [INPUT[PREFIX]]
Output fixed-size pieces of INPUT toPREFIXaa, PREFIXab, …; default
size is 1000 lines, and default PREFIXis `x’. With no INPUT, or when INPUT
is -, read standard input.
Mandatory arguments to long options aremandatory for short options too.
-a, –suffix-length=N usesuffixes of length N (default 2)
-b, –bytes=SIZE put SIZEbytes per output file
-C, –line-bytes=SIZE put atmost SIZE bytes of lines per output file
-d, –numeric-suffixes usenumeric suffixes instead of alphabetic
-l, –lines=NUMBER put NUMBERlines per output file
–verbose print adiagnostic just before each
output file isopened
–help display this help andexit
–version output versioninformation and exit
SIZE may be (or may be an integeroptionally followed by) one of following:
KB 1000, K 1024, MB 1000*1000, M1024*1024, and so on for G, T, P, E, Z, Y.
We can use the –lines=NUMBER parameter tosplit our raw-md5.hashes.txt file.
我们可以使用“–lines=NUMBER”参数来分割我们的原始md5.hashes.txt文件。
root@m3g9tr0n:~/hashes$ split -l 3000000raw-md5.hashes.txt part
请注意，考虑到每个MD5哈希是32字节长，我们还可以基于MB来进行文件分割。
使用oclHashcat-plus来破解密码
我开始玩oclHashcat-plus，因为它包含了-remove选项，一旦被破解它从hashfile删除破解成功的哈希值，这是真的很方便。oclHashcat-plus唯一的限制，是它对密码长度的限制。换句话说，它仅能够破解密码最多15个字符。我使用oclHashcat-plus规则是base64.rule，passwordspro.rule，T0XlC.rule在某些情况下d3ad0ne.rule。这些规则可以直接从oclHashcat-plus中找到。
暴力破解技术是不是我的第一选择。我用的字典是从g0tm1lk的BlogSpot下载的（http://blog.g0tmi1k.com/2011/06/dictionaries-wordlists/?redirect）。在g0tmi1k的文章其他外部链接中你会发现更多的字典。破解过程中最大的一部分是通过使用这些字典与上述规则进行。让我们来看看一些例子
使用单一规则：
./oclHashcat-plus64.bin -m 0 ~/hashes/md5_1~/Wordlists/d3ad0ne.dic -r rules/best64.rule -o Ultimate_Crack/eNtr0pY_1–remove
使用组合规则：
./oclHashcat-plus64.bin -m 0 ~/hashes/md5_1~/Wordlists/d3ad0ne.dic -r rules/best64.rule r rules/passwordspro.rule -oUltimate_Crack/eNtr0pY_1 –remove
Bruteforce attack with mask (you canspecify whichever charset you want):
使用掩码进行暴力破解***（可以指定任何你想要的字符集）：
./oclHashcat-plus64.bin -a 3 -1 ?l?d?u?s -m0 ~/hashes/md5_1 ?1?1?1?1?1?1?1?1 -o Ultimate_Crack/eNtr0pY_1 –remove
组合破解：
./oclHashcat-plus64.bin -a 1 -m 0~/hashes/md5_1 ~/Wordlists/d3ad0ne.dic ~/Wordlists/list -oUltimate_Crack/eNtr0pY_1 –remove
使用规则进行组合破解：
./oclHashcat-plus64.bin -a 1 -m 0~/hashes/md5_1 ~/Wordlists/d3ad0ne.dic ~/Wordlists/list -r rules/passwordspro.rule-o Ultimate_Crack/eNtr0pY_1 –remove
排列组合破解：
./oclHashcat-plus64.bin -a 4 -m 0~/hashes/md5_1 ~/Wordlists/d3ad0ne.dic -o Ultimate_Crack/eNtr0pY_1 –remove
排列组合规则破解：
./oclHashcat-plus64.bin -a 4 -m 0~/hashes/md5_1 ~/Wordlists/d3ad0ne.dic -r rules/best64.rule -oUltimate_Crack/eNtr0pY_1 –remove
In some cases, I used the hybrid + maskattack technique:
在某些情况下，我使用了混合 + 掩码***技术：
./oclHashcat-plus64.bin -a 6 -1 ?l?d -m 0 ~/hashes/md5_1~/Wordlists/d3ad0ne.dic ?1?1 -o Ultimate_Crack/eNtr0pY_1 –remove
混合掩码规则破解：
./oclHashcat-plus64.bin -a 6 -1 ?l?d -m 0~/hashes/md5_1 ~/Wordlists/d3ad0ne.dic ?1?1 -r rules/best64.rule -oUltimate_Crack/eNtr0pY_1 –remove
在这一点上，我没有用最后两个方法，因为他们是非常耗时的。我宁愿为John the Ripper 找到一个更好的KoreLogic的规则，将John the Ripper 的输出管道到oclHashcat-plus。正如我所说，oclHashcat-plus能够破解的密码最多15个字符。出于这个原因，我不得不通过“–stdout”选项限定每一次所产生的单词长度。如果你拥有一个非常快速的GPU可以跳过下面的例子。
./john –wordlist=~/Wordlists/all.lst-rules:KoreLogicRulesPrependYears –stdout=10 | ./oclHashcat-plus64.bin -m 0~/hashes/md5_1 -o Ultimate_Crack/eNtr0pY_1 –remove
当然你也可以使用来自Korelogic创建的其他前置的规则，像KoreLogicRulesPrependNumNum，甚至更好的创建自己的规则！是时候从破解密码的单词表中生成字典来破解剩余哈希值了。从eNtr0pY_1，使用以下命令我删除了MD5哈希值。
cut -b34- eNtr0pY_1 >eNtr0pY_1.dic
通过使用上述制造的字典，大量MD5哈希值被这种指纹***破解。你可以阅读更多关于马丁博斯@purehate这种***，我向你保证，这个技术是非常成功的！
当然，你也可以使用hashcat-utils中的二进制文件以及每一个管道到oclHashcat-plus的输出。
root@m3g9tr0n:~/oclHashcat-plus-0.08/hashcat-utils$ls
combinator.bin expander.bin gate.bin len.bin mp32.bin permute.bin prepare.bin req.bin splitlen.bin
使用John the Ripper破解密码
测试完所有我收集的字典以及几天后，是时候转移到john the Ripper来破解密码哈希值的剩余部分…
我用的Magnum-ripper编译OpenCL支持ATI / AMD显卡，因为我想用“–format=raw-md5-opencl参数”。相比“–format=raw-md5”，这是更快的方式为它使用你的CPU和GPU。
与john the Ripper中使用的规则是：
wordlist
Single
NT
Extra
KoreLogicRulesAppendNumbersandSpecials_Simple
KoreLogicRulesAppend6Num
KoreLogicRulesPrependAndAppendSpecial
KoreLogicRulesAppendNumNum_AddSpecialEverywhere
KoreLogicRulesAppendNumNumNum_AddSpecialEverywhere
KoreLogicRulesL33t.
您可以下载这些规则，并将其添加到您的john.conf文件中：
http://paste2.org/p/2048517
http://paste2.org/p/2048512
现在，让我们看到john the Ripper的一些例子…
使用–rules=Single：
./john –format=raw-md5-opencl–wordlist=../../Wordlists/all.lst –rules:Single ~/hashes/md5_1
哈希破解结果存储在默认文件john.pot中。您可以使用cat，more，head和tail检查它的内容。
root@m3g9tr0n:~/Tools/Password_Cracking/magnum-jumbo-OpenCL/run$tail -n 9 john.pot
$MD5$0fad81e7a61b47d387dde893fcf8e88a:anacarolinagu
$MD5$0f82fc9a81f5db07eb9289767390fd2b:fabulousfoodsu
$MD5$0e22933267b2e7df062703c4e5842029:fabuloustravelu
$MD5$0d40086a54fefe993c9816d1441672ac:modularhomeu
$MD5$0ed8181fc4d18e260dd8e36633124bfd:greenshoppingu
$MD5$0d6e8da4017ec5c384ac5536087da44d:lawofattractionu
$MD5$0eb916d3c6a66a32cedd4acc6edb1dbb:hotreportu
$MD5$0e241f99b5c13d56686ec618ab54d5fa:flightsandholidaysu
$MD5$0f3c99478362aae389d2cbf716394269:stthomasmoresu
To generate a wordlist from the john.pot file,you can use the following command.
要从john.pot文件生成从一个字典，可以使用下面的命令。
cut -d: -f 2- john.pot | sort -u >cracked.dic
生成的字典可以与上述规则相结合来破解哈希值。
当我使用oclHashcat-plus破解MD5哈希值时，我观察到，一些生成的密码被拒绝。这是因为oclHashcat-plus有一个关于字符串长度的限制。出于这个原因，我管道hashcat的结果输出到john the Ripper，并且在john the Ripper中更好的使用hashcat规则。
./hashcat-cli64.bin –stdout~/Wordlists/d3ad0ne.dic -r rules/best64.rule | ./john –format=raw-md5-opencl–stdin ~/hashes/md5_1
尝试所有字典与上述规则相结合后，是时候使用john the Ripper进行暴力破解破解。不幸的是，john the Ripper暴力破解***时不使用掩码***以产生密码。我们有基于包含在破解的密码来创建我们自己的字符集john.pot。
./john –make-charset=eNtr0pY.chr
Loaded 7948325 plaintexts
Generating charsets… 1 2 3 4 5 6 78 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DONE
Generating cracking order… DONE
Successfully written charset file:eNtr0pY.chr (95 characters)
你们中许多人会想知道“DONE 31”……这仅仅是因为我编译john the Ripper用31个字符长度。默认情况下，john the Ripper被编译为8个字符长度的支持，所以最好通过修改头文件params.h中的以下行来改变，它位于john the Ripper的scr文件夹中。
#define CHARSET_MIN ‘ ‘
#define CHARSET_MAX 0x7E
#define CHARSET_SIZE (CHARSET_MAX – CHARSET_MIN+ 1)
#define CHARSET_LENGTH 8 //更改为31或者你期待的值
最后，在john.conf中你必须包括您所创建的字符集，以如本例中规定：
# Incremental modes
[Incremental:eNtr0pY]
File = $JOHN/eNtr0pY.chr
MinLen = 0
MaxLen = 31
CharCount = 95
Now it is time to use bruteforce attackswith our own charstet!
现在是时候用我们自己的字符串来进行暴力破解***！
./john –format=raw-md5-opencl–incremental=eNtr0pY ~/hashes/md5_1
如果你查看john.conf，你会看到一些暴力破解***模式分为外部。这些都是双重的，带键盘（使用从键盘字符产生的相邻组合），KnownForce，日期时间，重复，顺序，亚群和DumbForce为疯狂的密码格式。
./john –format=raw-md5-opencl–external=DumbForce ~/hashes/md5_1
在这里提供的dict2hash.pl脚本帮助下，我们也喜欢破解双重MD5哈希值。
perl dict2hash.pl < rockyou.txt | ./john–format=raw-md5-opencl –stdin ~/md5_1
在这里，您可以看到与john the Ripper破获MD5值的一些示例：
http://pastebin.com/uaGFXkBA
http://pastebin.com/LT5Gda9L
Personally, I believe a password like”$MD5$0b26a0faf1344d6e772bf55628e10e29:n34=mn { .clipboard $me }” isimpossible to crack with bruteforce attacks.
Note: All the abovementioned techniques canbe used with oclHashcat-plus by defining -m 100 and with John the Ripper bydefining –format=raw-sha1-opencl for SHA1 cracking with OpenCL!
就个人而言，我相信密码像“ $ MD5 $ 0b26a0faf1344d6e772bf55628e10e29：N34 = {MN $.clipboard$me} ”是不可能使用暴力破解***来破解。
注：所有上述技术可以在oclHashcat-plus定义参数“-m 100”使用，john the Ripper定义“–format=raw-sha1-opencl”使用OpenCL破解SHA1
密码分析
最后，值得一看的分析使用pipal（密码分析者）从破解结果中收集的样本。
root@m3g9tr0n:~/pipal$ ruby1.9.1 pipal.rb \
-o eNtr0pY_1~/Wordlists/Ultimate/Part1/eNtr0pY_5.dic
Total entries = 759103
Total unique entries = 758299
Top10 passwords
niezgadniesz123 = 3 (0.0%)
ubqu = 3 (0.0%)
amonys = 3 (0.0%)
centralitie = 3 (0.0%)
bobydu = 3 (0.0%)
hanghuynh = 3 (0.0%)
hmadyousi = 3 (0.0%)
matthewperman = 3 (0.0%)
shadowninja2 = 3 (0.0%)
lhz4 = 3 (0.0%)
Top10 基本密码
august = 219 (0.03%)
july = 205 (0.03%)
april = 199 (0.03%)
june = 195 (0.03%)
march = 165 (0.02%)
alex = 161 (0.02%)
love = 132 (0.02%)
chris = 130 (0.02%)
daniel = 128 (0.02%)
dragon = 122 (0.02%)
密码长度(以长度进行排序)
1 = 13 (0.0%)
2 = 103 (0.01%)
3 = 1332 (0.18%)
4 = 16781 (2.21%)
5 = 19831 (2.61%)
6 = 95800 (12.62%)
7 = 202414 (26.66%)
8 = 158562 (20.89%)
9 = 103855 (13.68%)
10 = 75652 (9.97%)
11 = 46023 (6.06%)
12 = 24997 (3.29%)
13 = 8423 (1.11%)
14 = 3772 (0.5%)
15 = 1560 (0.21%)
密码长度(以计数排序)
7 = 202414 (26.66%)
8 = 158562 (20.89%)
9 = 103855 (13.68%)
6 = 95800 (12.62%)
10 = 75652 (9.97%)
11 = 46023 (6.06%)
12 = 24997 (3.29%)
5 = 19831 (2.61%)
4 = 16781 (2.21%)
13 = 8423 (1.11%)
14 = 3772 (0.5%)
15 = 1560 (0.21%)
3 = 1332 (0.18%)
2 = 103 (0.01%)
1 = 13 (0.0%)
|
|
|
||
||
||
||
|||
||||
||||
|||||
|||||
||||||
||||||
|||||||||
|||||||||||||||||
00000000001111111
01234567890123456
Oneto six characters = 133854 (17.63%)
One to eight characters = 494828 (65.19%)
More than eight characters = 264275(34.81%)
Onlylowercase alpha = 154996 (20.42%)
Only uppercase alpha = 14072 (1.85%)
Only alpha = 169068 (22.27%)
Only numeric = 119581 (15.75%)
First capital last symbol = 6088 (0.8%)
First capital last number = 73611 (9.7%)
Months
january = 109 (0.01%)
february = 45 (0.01%)
march = 247 (0.03%)
april = 251 (0.03%)
may = 850 (0.11%)
june = 246 (0.03%)
july = 223 (0.03%)
august = 300 (0.04%)
september = 80 (0.01%)
october = 134 (0.02%)
november = 113 (0.01%)
december = 115 (0.02%)
Days
monday = 59 (0.01%)
tuesday = 20 (0.0%)
wednesday = 7 (0.0%)
thursday = 38 (0.01%)
friday = 46 (0.01%)
saturday = 7 (0.0%)
sunday = 70 (0.01%)
Months (Abreviated)
jan = 1482 (0.2%)
feb = 249 (0.03%)
mar = 8397 (1.11%)
apr = 692 (0.09%)
may = 850 (0.11%)
jun = 889 (0.12%)
jul = 1051 (0.14%)
aug = 785 (0.1%)
sept = 215 (0.03%)
oct = 512 (0.07%)
nov = 821 (0.11%)
dec = 874 (0.12%)
Days (Abreviated)
mon = 4319 (0.57%)
tues = 28 (0.0%)
wed = 217 (0.03%)
thurs = 44 (0.01%)
fri = 758 (0.1%)
sat = 769 (0.1%)
sun = 1018 (0.13%)
Includes years
1975 = 411 (0.05%)
1976 = 388 (0.05%)
1977 = 446 (0.06%)
1978 = 432 (0.06%)
1979 = 441 (0.06%)
1980 = 541 (0.07%)
1981 = 453 (0.06%)
1982 = 519 (0.07%)
1983 = 533 (0.07%)
1984 = 603 (0.08%)
1985 = 585 (0.08%)
1986 = 616 (0.08%)
1987 = 710 (0.09%)
1988 = 641 (0.08%)
1989 = 941 (0.12%)
1990 = 931 (0.12%)
1991 = 995 (0.13%)
1992 = 935 (0.12%)
1993 = 905 (0.12%)
1994 = 907 (0.12%)
1995 = 4021 (0.53%)
1996 = 858 (0.11%)
1997 = 486 (0.06%)
1998 = 443 (0.06%)
1999 = 416 (0.05%)
2000 = 1024 (0.13%)
2001 = 643 (0.08%)
2002 = 586 (0.08%)
2003 = 1132 (0.15%)
2004 = 1254 (0.17%)
2005 = 796 (0.1%)
2006 = 818 (0.11%)
2007 = 1442 (0.19%)
2008 = 1019 (0.13%)
2009 = 742 (0.1%)
2010 = 767 (0.1%)
2011 = 516 (0.07%)
2012 = 925 (0.12%)
2013 = 165 (0.02%)
2014 = 142 (0.02%)
2015 = 146 (0.02%)
2016 = 118 (0.02%)
2017 = 139 (0.02%)
2018 = 131 (0.02%)
2019 = 172 (0.02%)
2020 = 179 (0.02%)
Years (Top 10)
1995 = 4021 (0.53%)
2007 = 1442 (0.19%)
2004 = 1254 (0.17%)
2003 = 1132 (0.15%)
2000 = 1024 (0.13%)
2008 = 1019 (0.13%)
1991 = 995 (0.13%)
1989 = 941 (0.12%)
1992 = 935 (0.12%)
1990 = 931 (0.12%)
Colours
black = 485 (0.06%)
blue = 549 (0.07%)
brown = 184 (0.02%)
gray = 89 (0.01%)
green = 348 (0.05%)
orange = 125 (0.02%)
pink = 262 (0.03%)
purple = 73 (0.01%)
red = 2974 (0.39%)
white = 179 (0.02%)
yellow = 85 (0.01%)
violet = 63 (0.01%)
indigo = 22 (0.0%)
Single digit on the end = 92080 (12.13%)
Two digits on the end = 87587 (11.54%)
Three digits on the end = 103715 (13.66%)
Last number
0 = 45407 (5.98%)
1 = 64764 (8.53%)
2 = 52570 (6.93%)
3 = 52890 (6.97%)
4 = 43719 (5.76%)
5 = 55185 (7.27%)
6 = 42826 (5.64%)
7 = 46169 (6.08%)
8 = 42475 (5.6%)
9 = 44930 (5.92%)
|
|
| ||
||||
|||| | | |
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
||||||||||
0123456789
Last digit
1 = 64764 (8.53%)
5 = 55185 (7.27%)
3 = 52890 (6.97%)
2 = 52570 (6.93%)
7 = 46169 (6.08%)
0 = 45407 (5.98%)
9 = 44930 (5.92%)
4 = 43719 (5.76%)
6 = 42826 (5.64%)
8 = 42475 (5.6%)
Last 2 digits (Top 10)
95 = 14675 (1.93%)
23 = 12192 (1.61%)
12 = 9230 (1.22%)
11 = 8214 (1.08%)
01 = 7606 (1.0%)
00 = 7131 (0.94%)
07 = 6295 (0.83%)
10 = 6182 (0.81%)
21 = 5881 (0.77%)
99 = 5868 (0.77%)
Last 3 digits (Top 10)
123 = 6857 (0.9%)
995 = 4122 (0.54%)
971 = 2916 (0.38%)
972 = 2850 (0.38%)
007 = 2514 (0.33%)
000 = 1868 (0.25%)
234 = 1725 (0.23%)
666 = 1465 (0.19%)
777 = 1389 (0.18%)
004 = 1347 (0.18%)
Last 4 digits (Top 10)
1995 = 3886 (0.51%)
1234 = 1379 (0.18%)
2007 = 1325 (0.17%)
2004 = 1121 (0.15%)
2003 = 1016 (0.13%)
2008 = 869 (0.11%)
2000 = 846 (0.11%)
1991 = 819 (0.11%)
2012 = 809 (0.11%)
1990 = 789 (0.1%)
Last 5 digits (Top 10)
12345 = 743 (0.1%)
23456 = 652 (0.09%)
54321 = 189 (0.02%)
23123 = 140 (0.02%)
56789 = 127 (0.02%)
34567 = 102 (0.01%)
11111 = 99 (0.01%)
45678 = 75 (0.01%)
00000 = 73 (0.01%)
88888 = 68 (0.01%)
US Area Codes
971 = Oregon: Metropolitan Portland,
Salem/Keizer area,
incl Cricket Wireless (OR)
972 = Texas: Dallas Metro (TX)
234 = NE Ohio: Canton, Akron (OH)
Character sets
loweralphanum: 330937 (43.6%)
loweralpha: 154996 (20.42%)
numeric: 119581 (15.75%)
mixedalphanum: 41121 (5.42%)
upperalphanum: 41078 (5.41%)
mixedalpha: 28464 (3.75%)
upperalpha: 14072 (1.85%)
loweralphaspecial: 10222 (1.35%)
loweralphaspecialnum: 5735 (0.76%)
mixedalphaspecial: 4724 (0.62%)
upperalphaspecial: 2939 (0.39%)
mixedalphaspecialnum: 2247 (0.3%)
specialnum: 648 (0.09%)
upperalphaspecialnum: 374 (0.05%)
special: 47 (0.01%)
Character set ordering
stringdigit: 349534 (46.05%)
allstring: 197532 (26.02%)
alldigit: 119581 (15.75%)
digitstring: 28873 (3.8%)
othermask: 18649 (2.46%)
stringdigitstring: 14577 (1.92%)
stringspecial: 10441 (1.38%)
digitstringdigit: 9981 (1.31%)
stringspecialstring: 5469 (0.72%)
stringspecialdigit: 3075 (0.41%)
specialstring: 834 (0.11%)
specialstringspecial: 510 (0.07%)
allspecial: 47 (0.01%)
Hashcat masks (Top 10)
?d?d?d?d?d?d?d: 85053 (11.2%)
?l?l?l?l?l?l: 38400 (5.06%)
?l?l?l?l?l?l?l?l: 36217 (4.77%)
?l?l?l?l?l?l?l: 35468 (4.67%)
?l?l?l?l?l?l?d?d?d: 24051 (3.17%)
?l?l?l?l?l?l?d?d: 18591 (2.45%)
?l?l?l?l?l?d?d?d: 18047 (2.38%)
?d?d?d?d?d?d: 16048 (2.11%)
?l?l?l?l?l?l?l?l?l: 14236 (1.88%)
?l?l?l?l?d?d?d: 13802 (1.82%)
结论
这是一个非常耗费时间和艰苦的工作，因为我不拥有最快的显卡。整个破解过程历时约5个月来完成，因为我必须完成我的CCNP认证学习。
从中学到的教训是，一个好的智能词典结合方便的规则，使用Hashcat或John the Ripper甚至可以破解强密码。基于以上情况，管理员应该使用更强的哈希算法（加盐）来存储你的密码，并在在定期的基础上更改自己的密码。
谢谢阅读
你可以在推特上找到我, @m3g9tr0n.
下载
你可以下载破解的哈希结果值
721.9 MB -m3g9tr0n_122Million_Passwords_WordLists.zip
所提供的KoreLogic torrent文件包含了各种独特的密码哈希值。出于这个原因，你可能会在这些字典中发现重复的密码，作为一个单独的密码，可以使用各种算法来计算散列值！这意味着1.22亿独特的哈希值（MD5，SHA1，双MD5等）被破解，结果中有8亿3600万个独特的密码。
您也可以下载“都在一个”版本，干净和分类：
270.2 MB -m3g9tr0n_Passwords_WordList_CLEANED.zip
export LC_ALL=’C’ && cat * | sort |uniq > eNtr0pY_ALL_sort_uniq.dic
参考：
hashcat Wiki [ http://hashcat.net/wiki/ ]
John the Ripper Wiki [http://openwall.info/wiki/john ]
原创文章，作者：DZDIG，如若转载，请注明出处：https://www.beidandianzhu.com/g/4332.html
