浅谈筛选日志中的IP地址信息

作为运维人员,经常会需要会对日志中的某些重要信息进行筛选,比如说ip等参数。

案例一:筛选出IP地址信息

日志信息如下:

[root@C67-X64-A1 hanghang]# cat test.txt 
Jul 13 08:13:09 localhost sshd[14678]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:09 localhost sshd[14679]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=222.73.173.143 user=root
Jul 13 08:13:11 localhost sshd[14691]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 user=admin
Jul 13 08:13:11 localhost sshd[14692]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=222.73.173.143 
Jul 13 08:13:14 localhost sshd[14707]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:14 localhost sshd[14711]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=222.73.173.143 user=root
Jul 13 08:13:17 localhost sshd[14722]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:17 localhost sshd[14724]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=222.73.173.143 
Jul 13 08:13:20 localhost sshd[14739]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 user=root
Jul 13 08:13:23 localhost sshd[14753]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 user=root
Jul 13 08:13:26 localhost sshd[14767]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:29 localhost sshd[14781]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:32 localhost sshd[14795]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:35 localhost sshd[14809]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:38 localhost sshd[14823]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:41 localhost sshd[14837]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 user=apache
Jul 13 08:13:44 localhost sshd[14851]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:47 localhost sshd[14865]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:49 localhost sshd[14876]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172 
Jul 13 08:13:53 localhost sshd[14895]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=61.152.95.172

方法1:利用awk命令进行筛选

[root@C67-X64-A1 hanghang]# awk -F "rhost=" '{print $NF}' test.txt |awk '{print $1'}|sort -r|uniq
61.152.95.172
222.73.173.143

方法2:利用grep的扩展命令egrep进行筛选

[root@C67-X64-A1 hanghang]# egrep -o '[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}' test.txt |sort -r|uniq
61.152.95.172
222.73.173.143

方法3:利用sed命令进行筛选

[root@C67-X64-A1 hanghang]# sed -nr 's/.*[^0-9](([0-9]+\.){3}[0-9]+).*/\1/p' test.txt |sort -r|uniq
61.152.95.172
222.73.173.143
[root@C67-X64-A1 hanghang]# sed -nr 's/(^|.*[^0-9])(([0-9]+\.){3}[0-9]+).*/\2/p' test.txt |sort -r|uniq
61.152.95.172
222.73.173.143

案例二:根据需求对日志信息进行筛选

需求:

最近需要处理下网站日志:

例如

A 1.1.1.1      用户访问  有index.html  和a.jpg   的日志

B 20.20.20.20  用户访问  有index.html  的日志 没其他文件记录的日志

现在需要提取B的IP    不需要A的IP 

日志信息如下:

[root@C67-X64-A1 hanghang]# cat files 
1.1.1.1 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
10.10.10.10 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
10.10.10.10  - - [19/Jul/2013:15:01:39 +0800] "GET /logo.jpg  HTTP/1.1
10.10.10.10  - - [19/Jul/2013:15:01:39 +0800] "GET /a.js  HTTP/1.1
3.3.3.3 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
20.20.20.20 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
20.20.20.20  - - [19/Jul/2013:15:01:39 +0800] "GET /logo.jpg  HTTP/1.1
20.20.20.20  - - [19/Jul/2013:15:01:39 +0800] "GET /a.js  HTTP/1.1
30.30.30.30 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
30.30.30.30  - - [19/Jul/2013:15:01:39 +0800] "GET /logo.jpg  HTTP/1.1
30.30.30.30  - - [19/Jul/2013:15:01:39 +0800] "GET /a.js  HTTP/1.1
4.4.4.4 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
5.5.5.5 - - [19/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
1.1.1.1 - - [20/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
2.2.2.2 - - [21/Jul/2013:15:01:39 +0800] "GET /index.html  HTTP/1.1
3.3.3.3 - - [21/Jul/2013:15:01:55 +0800] "GET /index.html  HTTP/1.1
4.4.4.4 - - [21/Jul/2013:16:01:55 +0800] "GET /index.html  HTTP/1.1
5.5.5.5 - - [21/Jul/2013:17:02:55 +0800] "GET /index.html  HTTP/1.1

Shell脚本实现:

#!/bin/bash
#author molewan
for i in `grep -v "/index.html"  files  | awk '{print $1}' | uniq`;do
    echo "| grep -v "$i" " >> tmp_title
done
M=`cat tmp_title | tr "\n" " " | sed 's#^#cat files | sort -r | uniq#'`
echo $M | bash | awk '{print $1}'
rm -rf tmp_title

Python脚本实现:

假设日志信息是放在文件log.dat里面的:

#! /usr/bin/env python                                                                
import re                     
Dip_reso = {}  
pattern = re.compile('(\d+\.\d+\.\d+\.\d+).*GET /(.*) .*')             
f = open('log.dat')
               
for line in f: 
    resource = re.match(pattern, line)                                 
    key = resource.group(1)                                            
    value = resource.group(2)                                          
    if key in Dip_reso:
        if value not in Dip_reso[key]:                                                
            Dip_reso[key].append(value)    
        else:
            continue           
    else:      
        Dip_reso[key] = []                                             
        Dip_reso[key].append(value)                                    
f.close()      
               
for k in Dip_reso:                                                     
    if len(Dip_reso[k]) == 1 and  cmp(Dip_reso[k][0], 'index.html') == 1:
        print k

                

#如果你要搜集数据,可以这样

# ip_data = [ip for ip in Dip_reso if len(Dip_reso[ip]) == 1 and cmp(Dip_reso[ip][0], 'index.html') == 1]

这样,ip_data就是所有的ip了。

原创文章,作者:Net21-冰冻vs西瓜,如若转载,请注明出处:http://www.178linux.com/24749

(1)
Net21-冰冻vs西瓜Net21-冰冻vs西瓜
上一篇 2016-07-22 10:09
下一篇 2016-07-22 10:10

相关推荐

  • Windows 10 VS. Linux

    原文出处: Linux中国 – zpl1025   前阵子 Windows 10 好像占据了绝大部分头条,甚至在一些Linux圈里也是一样。最具代表性的是 betanews.com 的 Brian Fagioli 说 Windows 10 已经为 Linux 桌面系统敲响了丧钟,Microsoft 如今宣布将为忠实的…

    系统运维 2015-03-24
  • 轻松实现源码打包安装[原创]

    通常我们在Linux/Unix下安装一平台时往往需要十几甚至更多安装包,这些源码包来源于网络、本地硬盘、移动设备。有时碰到网络不畅通或下载地址失效会带来很多麻烦,一个好方法便是将常用的软件包下载到本地硬盘存放。问题是久而久之连自己都不晓得哪些包才是适用的。现用makeself来实现自解压倒安装倒是一个很好的解决方案,下面以制作Func客户端安装包为例。 一、…

    Linux干货 2015-03-27
  • Linux网络管理之路由和team网络组

    如何实现把linux当路由器来使用,实现不同网段之间的通信访问,这就是今天所要实现的路由配置   环境准备:4台虚拟机,两台中间两台作为路由器来使用(每天主机两块网卡),其他两台作为终端主机,目的是让不同网段的两台主机之间互相访问(这里在vmware里面做的实验,所以网卡的类型都设置为桥接)   拓扑图:   步骤一:将A、D两…

    系统运维 2016-09-10
  • Shell 变量类型及用法

    一、      什么是变量          变量源于数学,在计算机语言中能储存计算机结果或能表示值的抽象概念,变量可以由变量名访问,在指令语言中,变量通常是可变的。Linux是一个多用户的操作系统。每个用户登录系统后,都会有一个…

    Linux干货 2016-08-12
  • redis主从复制(4)— client buffer

    1、 client buffer的设计 redis server以单进程的方式处理接收到的请求,而redis完成请求有些工作比较慢,比如网络IO和磁盘IO等比较慢的操作。redis为了提高处理客户端请求的响应时间,做了很多优化。比如网络io和磁盘io是异步完成、使用后台进程完成bgsave和bgrewriteaof工作,在server端为客户提供读buffe…

    Linux干货 2016-03-28
  • Hadoop Hive与Hbase整合+thrift

    1.  简介 Hive是基于Hadoop的一个数据仓库工具,可以将结构化的数据文件映射为一张数据库表,并提供完整的sql查询功能,可以将sql语句转换为MapReduce任务进行运行。 其优点是学习成本低,可以通过类SQL语句快速实现简单的MapReduce统计,不必开发专门的MapReduce应用,十分适合数据仓库的统计分析。 Hive与HBas…

    Linux干货 2015-04-13