我的日志分析之道：简单的Web日志分析脚本
铁匠发表于 2017-02-15 06:23:23
前言
长话短说，事情的起因是这样的，由于工作原因需要分析网站日志，服务器是windows，iis日志，在网上找了找，github找了找，居然没找到，看来只有自己动手丰衣足食。
那么分析方法我大致可分为三种：
1. 基于时间：将请求url按时间段分类，那么我们根据每个时间段的url数量及攻击数量就可以大致判断出哪个时间段有apt类型攻击，哪个时间段是扫描器行为；
2. 基于攻击ip：正常的攻击肯定会有请求被记录（当然你要是有0day当我没说，正常的探测总会有吧=。=！），然后每个ip去分析；
3. 基于访问请求的状态码，也大致可以判断出行为。
规则可以基于开源waf规则，分析扫描器写正则也可以，开源waf地址是
https://github.com/loveshell/ngx_lua_waf/tree/master/wafconf。
扫描器正则https://github.com/smarttang/w3a_SOCD的database里面有详细地址
https://github.com/smarttang/w3a_SOC/tree/master/db_sql。
Sql语句里面有想把它做的功能全一些，但是学python学习时间也不是很长，写出来的代码也没有pythonic，会慢慢写的。目前分三个模块，一个日志归类模块命名为url.py，攻击分析模块attac.py， ip地理位置查询模块ipfind.py，还有一个主函数。
日志归类模块url.py
<span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">re
</span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">os
</span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">sys
</span><span lang="en-us" xml:lang="en-us">from </span><span lang="en-us" xml:lang="en-us">datetime </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">datetime
dt = datetime.now()
date = </span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(dt.date())
loglist = [] </span><span lang="en-us" xml:lang="en-us">#
</span><span lang="en-us" xml:lang="en-us">iplist = [] </span><span lang="en-us" xml:lang="en-us"># ip</span>统计<span lang="en-us" xml:lang="en-us">
</span><span lang="en-us" xml:lang="en-us">urllist = [] </span><span lang="en-us" xml:lang="en-us"># url</span>统计列表<span lang="en-us" xml:lang="en-us">
</span><span lang="en-us" xml:lang="en-us">needlist = [] </span><span lang="en-us" xml:lang="en-us"># </span>需要统计的<span lang="en-us" xml:lang="en-us">
</span><span lang="en-us" xml:lang="en-us">errorlist = [] </span><span lang="en-us" xml:lang="en-us"># </span>格式错误的列表<span lang="en-us" xml:lang="en-us">
</span><span lang="en-us" xml:lang="en-us">ipdict,urldict = {},{}
rizhi = </span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">input</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>请输入要分析的日志文件名<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">))
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">find_log():
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>开始解析日志<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">with </span><span lang="en-us" xml:lang="en-us">open</span><span lang="en-us" xml:lang="en-us">(rizhi,</span><span lang="en-us" xml:lang="en-us">'r'</span><span lang="en-us" xml:lang="en-us">,</span><span lang="en-us" xml:lang="en-us">encoding</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">'UTF-8'</span><span lang="en-us" xml:lang="en-us">,</span><span lang="en-us" xml:lang="en-us">errors</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">'ignore'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">as </span><span lang="en-us" xml:lang="en-us">f:
</span><span lang="en-us" xml:lang="en-us">#loglist = f.readlines()
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">f.readlines(): </span><span lang="en-us" xml:lang="en-us">#
</span><span lang="en-us" xml:lang="en-us">if </span><span lang="en-us" xml:lang="en-us">i[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">] != </span><span lang="en-us" xml:lang="en-us">'#'</span><span lang="en-us" xml:lang="en-us">:
b = re.split(</span><span lang="en-us" xml:lang="en-us">' '</span><span lang="en-us" xml:lang="en-us">,i)
iplist.append(b[</span><span lang="en-us" xml:lang="en-us">10</span><span lang="en-us" xml:lang="en-us">])
urllist.append(b[</span><span lang="en-us" xml:lang="en-us">6</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">try</span><span lang="en-us" xml:lang="en-us">:
needlist.append([b[</span><span lang="en-us" xml:lang="en-us">10</span><span lang="en-us" xml:lang="en-us">],b[</span><span lang="en-us" xml:lang="en-us">1</span><span lang="en-us" xml:lang="en-us">],b[</span><span lang="en-us" xml:lang="en-us">5</span><span lang="en-us" xml:lang="en-us">],b[</span><span lang="en-us" xml:lang="en-us">6</span><span lang="en-us" xml:lang="en-us">],b[</span><span lang="en-us" xml:lang="en-us">15</span><span lang="en-us" xml:lang="en-us">]])
</span><span lang="en-us" xml:lang="en-us">except</span><span lang="en-us" xml:lang="en-us">:
errorlist.append(i)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>日志解析完毕<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">count(iplist,urllist): </span><span lang="en-us" xml:lang="en-us">#</span>统计<span lang="en-us" xml:lang="en-us">ip url</span>访问量函数<span lang="en-us" xml:lang="en-us">
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>开始分析<span lang="en-us" xml:lang="en-us">url</span>与<span lang="en-us" xml:lang="en-us">ip</span>访问量<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">global </span><span lang="en-us" xml:lang="en-us">ipdict,urldict
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(iplist):
ipdict[i] = iplist.count(i)
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(urllist):
urldict[i] = urllist.count(i)
ipdict = </span><span lang="en-us" xml:lang="en-us">sorted</span><span lang="en-us" xml:lang="en-us">(ipdict.items(),</span><span lang="en-us" xml:lang="en-us">key</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">lambda </span><span lang="en-us" xml:lang="en-us">d: d[</span><span lang="en-us" xml:lang="en-us">1</span><span lang="en-us" xml:lang="en-us">], </span><span lang="en-us" xml:lang="en-us">reverse</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">True</span><span lang="en-us" xml:lang="en-us">)
urldict = </span><span lang="en-us" xml:lang="en-us">sorted</span><span lang="en-us" xml:lang="en-us">(urldict.items(),</span><span lang="en-us" xml:lang="en-us">key</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">lambda </span><span lang="en-us" xml:lang="en-us">d: d[</span><span lang="en-us" xml:lang="en-us">1</span><span lang="en-us" xml:lang="en-us">], </span><span lang="en-us" xml:lang="en-us">reverse</span><span lang="en-us" xml:lang="en-us">=</span><span lang="en-us" xml:lang="en-us">True</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">type</span><span lang="en-us" xml:lang="en-us">(urldict))
iplist = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(ipdict)
urllist = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(urldict)
ipdict,urldict = {},{}
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>url</span>与<span lang="en-us" xml:lang="en-us">ip</span>分析完毕<span lang="en-us" xml:lang="en-us">.......'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">return </span><span lang="en-us" xml:lang="en-us">[iplist,urllist]
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">save_count():
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>正在保存分析结果<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
ipname = </span><span lang="en-us" xml:lang="en-us">'ip-'</span><span lang="en-us" xml:lang="en-us">+date+</span><span lang="en-us" xml:lang="en-us">'.txt'
</span><span lang="en-us" xml:lang="en-us">urlname = </span><span lang="en-us" xml:lang="en-us">'url-'</span><span lang="en-us" xml:lang="en-us">+date+</span><span lang="en-us" xml:lang="en-us">'.txt'
</span><span lang="en-us" xml:lang="en-us">with </span><span lang="en-us" xml:lang="en-us">open</span><span lang="en-us" xml:lang="en-us">(ipname,</span><span lang="en-us" xml:lang="en-us">'w'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">as </span><span lang="en-us" xml:lang="en-us">f:
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">iplist:
f.write(</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(i))+</span><span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">\n</span><span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">with </span><span lang="en-us" xml:lang="en-us">open</span><span lang="en-us" xml:lang="en-us">(urlname,</span><span lang="en-us" xml:lang="en-us">'w'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">as </span><span lang="en-us" xml:lang="en-us">f:
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">urllist:
f.write(</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(i))+</span><span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">\n</span><span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>分析结果保存完毕<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
find_log()
[iplist,urllist] = count(iplist,urllist)
save_count()</span>iis日志和apache日志觉得都差不多，就是切割时候改一下就行了。
Iis日志大概是这样的，用pythonreadlines然后切割出来就好了。
这个url.py我加了个功能把ip访问量及url访问量排序输出出来所以有点慢，=.=没办法野路子哪里会什么算法。将地址，时间，ip，状态码都扔进一个列表里就行了。
攻击分析模块attack.py
<span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">os
</span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">sys
</span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">url
sqllist,xsslist,senlist = [],[],[]
otherurl,xssip,sqlip,senip = [],[],[],[]
feifa = []
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">find_attack(needlist):
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>开始检测攻击<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
sql = </span><span lang="en-us" xml:lang="en-us">r'product.php|preg_\w+|execute|echo|print|print_r|var_dump|(fp)open|^eval$|file_get_contents|include|require|require_once|shell_exec|phpinfo|system|passthru|\(?:define|base64_decode\(|group\s+by.+\(|%20or%20|%20and%20|sleep|delay|nvarchar|exec|union|^select$|version|insert|information_schema|chr\(|concat|%bf|sleep\((\s*)(\d*)(\s*)\)|current|having|database'
</span><span lang="en-us" xml:lang="en-us">xss = </span><span lang="en-us" xml:lang="en-us">r'alert|^script$|<|>|%3E|%3c|&#x3E|\u003c|\u003e|&#x'
</span><span lang="en-us" xml:lang="en-us">sen = </span><span lang="en-us" xml:lang="en-us">r'\.{2,}|%2e{2,}|%252e{2,}|%uff0e{2,}0x2e{2,}|\./|\{FILE\}|%00+|json|\.shtml|\.pl|\.sh|\.do|\.action|zabbix|phpinfo|/var/|/opt/|/local/|/etc|/apache/|\.log|invest\b|\.xml|apple-touch-icon-152x152|\.zip|\.rar|\.asp\b|\.php|\.bak|\.tar\.gz|\bphpmyadmin\b|admin|\.exe|\.7z|\.zip|\battachments\b|\bupimg\b|uploadfiles|templets|template|data\b|forumdata|includes|cache|jmxinvokerservlet|vhost|bbs|host|wwwroot|\bsite\b|root|hytop|flashfxp|bak|old|mdb|sql|backup|^java$|class'
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">needlist:
</span><span lang="en-us" xml:lang="en-us">if </span><span lang="en-us" xml:lang="en-us">i[</span><span lang="en-us" xml:lang="en-us">2</span><span lang="en-us" xml:lang="en-us">] == </span><span lang="en-us" xml:lang="en-us">'POST' </span><span lang="en-us" xml:lang="en-us">or </span><span lang="en-us" xml:lang="en-us">i[</span><span lang="en-us" xml:lang="en-us">2</span><span lang="en-us" xml:lang="en-us">] == </span><span lang="en-us" xml:lang="en-us">'HEAD' </span><span lang="en-us" xml:lang="en-us">or </span><span lang="en-us" xml:lang="en-us">i[</span><span lang="en-us" xml:lang="en-us">2</span><span lang="en-us" xml:lang="en-us">] == </span><span lang="en-us" xml:lang="en-us">'GET'</span><span lang="en-us" xml:lang="en-us">:
response = re.findall(sql,i[</span><span lang="en-us" xml:lang="en-us">3</span><span lang="en-us" xml:lang="en-us">],re.I)
</span><span lang="en-us" xml:lang="en-us">if </span><span lang="en-us" xml:lang="en-us">response == []:
responsexss = re.findall(xss,i[</span><span lang="en-us" xml:lang="en-us">3</span><span lang="en-us" xml:lang="en-us">],re.I)
</span><span lang="en-us" xml:lang="en-us">if </span><span lang="en-us" xml:lang="en-us">responsexss == []:
responsesen = re.findall(sen,i[</span><span lang="en-us" xml:lang="en-us">3</span><span lang="en-us" xml:lang="en-us">],re.I)
</span><span lang="en-us" xml:lang="en-us">if </span><span lang="en-us" xml:lang="en-us">responsesen == []:
otherurl.append(i)
</span><span lang="en-us" xml:lang="en-us">else</span><span lang="en-us" xml:lang="en-us">:
senlist.append(i)
senip.append(i[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(responsesen)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>检测出敏感目录扫描<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(i)
</span><span lang="en-us" xml:lang="en-us">else</span><span lang="en-us" xml:lang="en-us">:
xsslist.append(i)
xssip.append(i[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(responsexss)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>检测出<span lang="en-us" xml:lang="en-us">xss</span>攻击<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(i)
</span><span lang="en-us" xml:lang="en-us">else</span><span lang="en-us" xml:lang="en-us">:
sqllist.append(i)
sqlip.append(i[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(responsexss)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>检测出<span lang="en-us" xml:lang="en-us">sql</span>攻击<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(i)
</span><span lang="en-us" xml:lang="en-us">else</span><span lang="en-us" xml:lang="en-us">:
feifa.append(i[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>非法请求<span lang="en-us" xml:lang="en-us">:'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(feifa))+</span><span lang="en-us" xml:lang="en-us">'</span>次<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(feifa))))+</span><span lang="en-us" xml:lang="en-us">'</span>个<span lang="en-us" xml:lang="en-us">ip'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>攻击检测完毕<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">return </span><span lang="en-us" xml:lang="en-us">[xssip,sqlip,senip,sqllist,xsslist,senlist,otherurl]
</span>这个就简单多了，基于正则分析的正则不是很完善，还有好多是根据自己公司情况来定，大牛轻喷，检索完毕返回<span lang="en-us" xml:lang="en-us">ip</span>及<span lang="en-us" xml:lang="en-us">url</span>。
IP地理位置查询模块ipfind.py
Ipfind.py是查找ip地理位置的
<span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">re
</span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">urllib.request
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">url_open(ip):
url = </span><span lang="en-us" xml:lang="en-us">'http://www.ip138.com/ips138.asp?ip='</span><span lang="en-us" xml:lang="en-us">+ip
response = urllib.request.urlopen(url)
html = response.read().decode(</span><span lang="en-us" xml:lang="en-us">'gb2312'</span><span lang="en-us" xml:lang="en-us">)
</span><span lang="en-us" xml:lang="en-us">return </span><span lang="en-us" xml:lang="en-us">html
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">find_ip(html):
a = </span><span lang="en-us" xml:lang="en-us">r'</span>本站数据<span lang="en-us" xml:lang="en-us">.{20,}</li>'
</span><span lang="en-us" xml:lang="en-us">p = re.compile(a,re.I)
response = re.findall(p,html)
</span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">response:
b = i
response = re.split(</span><span lang="en-us" xml:lang="en-us">r'</li><li>'</span><span lang="en-us" xml:lang="en-us">,b)
ipaddrs = </span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(response[</span><span lang="en-us" xml:lang="en-us">0</span><span lang="en-us" xml:lang="en-us">][</span><span lang="en-us" xml:lang="en-us">5</span><span lang="en-us" xml:lang="en-us">:])+</span><span lang="en-us" xml:lang="en-us">','</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(response[</span><span lang="en-us" xml:lang="en-us">1</span><span lang="en-us" xml:lang="en-us">][</span><span lang="en-us" xml:lang="en-us">6</span><span lang="en-us" xml:lang="en-us">:])+</span><span lang="en-us" xml:lang="en-us">','</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(response[</span><span lang="en-us" xml:lang="en-us">2</span><span lang="en-us" xml:lang="en-us">][</span><span lang="en-us" xml:lang="en-us">6</span><span lang="en-us" xml:lang="en-us">:-</span><span lang="en-us" xml:lang="en-us">5</span><span lang="en-us" xml:lang="en-us">])
</span><span lang="en-us" xml:lang="en-us">return </span><span lang="en-us" xml:lang="en-us">ipaddrs
</span><span lang="en-us" xml:lang="en-us">def </span><span lang="en-us" xml:lang="en-us">find_ipaddrs(ip):
html = url_open(ip)
ipaddrs = find_ip(html)
</span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(ip+</span><span lang="en-us" xml:lang="en-us">' : '</span><span lang="en-us" xml:lang="en-us">+ipaddrs)
</span>这个简单我是直接像爬虫那样写的，用<span lang="en-us" xml:lang="en-us">ip138</span>的网址<span lang="en-us" xml:lang="en-us">(</span>接口没有找到，百度注册了好几次都不成功，有<span lang="en-us" xml:lang="en-us">api</span>的可以用<span lang="en-us" xml:lang="en-us">api)</span>。
主函数
主函数main.py
<span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">re </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">os </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">sys </span><span lang="en-us" xml:lang="en-us">from </span><span lang="en-us" xml:lang="en-us">datetime </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">datetime </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">url </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">attack </span><span lang="en-us" xml:lang="en-us">import </span><span lang="en-us" xml:lang="en-us">ipfind needlist = url.needlist sqllist,xsslist,senlist = [],[],[] otherurl,iplist = [],[] [xssip,sqlip,senip,sqllist,xsslist,senlist,otherurl]=attack.find_attack(needlist) xssip = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(xssip)) sqlip = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(sqlip)) senip = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(senip)) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>检测出<span lang="en-us" xml:lang="en-us">xss</span>攻击<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(xsslist))+</span><span lang="en-us" xml:lang="en-us">'</span>次<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">'</span>共计<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(xssip))+</span><span lang="en-us" xml:lang="en-us">'</span>个<span lang="en-us" xml:lang="en-us">ip'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(xssip) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>检测出<span lang="en-us" xml:lang="en-us">sql</span>攻击<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(sqllist))+</span><span lang="en-us" xml:lang="en-us">'</span>次<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">'</span>共计<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(sqlip))+</span><span lang="en-us" xml:lang="en-us">'</span>个<span lang="en-us" xml:lang="en-us">ip'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(sqlip) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'>>>>>>></span>检测出敏感目录扫描<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(senlist))+</span><span lang="en-us" xml:lang="en-us">'</span>次<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">'</span>共计<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">+</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(senip))+</span><span lang="en-us" xml:lang="en-us">'</span>个<span lang="en-us" xml:lang="en-us">ip'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(senip) iplist = </span><span lang="en-us" xml:lang="en-us">list</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">set</span><span lang="en-us" xml:lang="en-us">(xssip+sqlip+senip)) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">len</span><span lang="en-us" xml:lang="en-us">(iplist)) </span><span lang="en-us" xml:lang="en-us">print</span><span lang="en-us" xml:lang="en-us">(</span><span lang="en-us" xml:lang="en-us">'</span>开始分析<span lang="en-us" xml:lang="en-us">ip</span>地理位置<span lang="en-us" xml:lang="en-us">'</span><span lang="en-us" xml:lang="en-us">) </span><span lang="en-us" xml:lang="en-us">for </span><span lang="en-us" xml:lang="en-us">i </span><span lang="en-us" xml:lang="en-us">in </span><span lang="en-us" xml:lang="en-us">iplist: ipfind.find_ipaddrs(</span><span lang="en-us" xml:lang="en-us">str</span><span lang="en-us" xml:lang="en-us">(i))</span>
要分析什么就把需要分析的和main.py放在一个目录下就行了
总结
脚本大概说了一遍，说说不足及怎么分析吧。
现实不足：看了差不多有3个月日志了吧，先说一个最严重的问题，post请求data看不见，本身日志就看不到data，何况等到https普及了日志什么样也不知道，要是有能力最好做成和waf联动那样的。还有就是未知威胁从waf来看基于正则，基于关键词有很多都是能绕过的，但是有攻击行为没删日志的话，肯定是会留下攻击痕迹的，这样可以从检测出来的ip来看具体攻击的url，而未知威胁则不同了，就好比一个0day，攻击waf没用了，日志分析看不出来了，那么只能依靠应急响应以及服务器的报警了。
还有好多攻击类型没有加入到里面，后期打算把判断攻击类型写成函数，拿if,else判断，类型少还可以，类型多了感觉很容易乱，还有user-agent的收集与判断（虽然大多数扫描器都能改user-agent）。
具体分析：我都是用脚本跑一遍，然后按ip来看会比较方便些，而这里缺少机器识别，我单独写了一个简易的机器识别的东西，其实要实现很简单，把全部日志按时间，url，ip扔进一个列表里统计一下相同时间相同ip的就可以了。我写的是识别短信轰炸的，后期还会渐渐的完善，如果有能力就把它结合django来弄成图形化，毕竟脚本始终是脚本，终究听着不好听。
效果如下
具体规则有待完善。