Port Scanner 7

Sendmail is certainly the buggiest daemon on earth; it has the highest number of known exploits amongst all the daemons. So this probably should get us through. Let us telnet to Port 25 and find out whether an exploitable version of Sendmail is running.




C:\windows> telnet xxx.bol.net.in 25



220 xxx.bol.net.in ESMTP Sendmail 8.9.1 (1.1.20.3/27Jun00-0346PM) Thu, 29 Jun 2000 14:18:12 0530 (IST)



When you telnet to Port 25, then the first thing that you come across would be a something like the above welcome daemon banner. A daemon banner is a Hacker’s best friend. It reveals important information about the host, which proves to be invaluable in breaking into it. It basically tells you which daemon or service is running on that port and also the version of that particular service. Like for example, in this case, the Sendmail daemon banner tells us that ESMTP Sendmail 8.9.1 is running and it also gives us other information about the host at which this service is running.



Anyway, getting back to the topic, this banner reveals a big vulnerability existing in the host computer. It tells us that xxx.bol.net.in is running an old, vulnerable version of Sendmail. The latest version is Sendmail 8.9.4 (correct me if I am wrong.), so this particular version of Sendmail wouldn’t be without any bugs.



So then what you do is visit PacketStorm or search at your favorite Hacking stuff related search engine for a C program which demonstrates how to exploit version 8.9.4 of Sendmail. Now, all this might sound a bit too simple, well it certainly isn’t, read on for more info.



Now, there are a couple of things that you need to keep in mind while getting this done. Say, you have found out that the victim runs Sendmail 8.9.4, now you cannot simply break in by running any exploit for this version. By that what, I mean to say is that, an exploit, which is coded to be executed on a Linux platform, will not work if you try to compile and run it on a Windows platform. So basically before you execute the ‘kewl’ exploit program that you downloaded, you should find out which platform it is meant for and if you are not running that platform, then you will need to get your gray cells working.

Jai Shree Raam

Port Scaning Part 6

you could get into some serious trouble. [Well actually not much, only say your account might be disabled. However, it could be worse.]




Ok, you are in, now let us get the FTP client to tell us which commands are available by typing the help command.



ftp> help

Commands may be abbreviated. Commands are:



! delete literal prompt

? debug ls put

append dir mdelete pwd

ascii disconnect mdir quit

bell get mget quote

binary glob mkdir recv

bye hash mls remotehelp

cd help mput rename

close lcd open rmdir



Uhmmm.,.., none of the above commands seem to be or sound to be of use to us. So the ‘help’ command did not reveal any useful commands. However, you see the above list of commands are commands which are offered by the FTP client and almost more often than not, the FTP daemon offers a wider array of commands. To get a complete list of commands offered by the FTP daemon, use the ‘remote help’ command:



Ftp> remote help

214-The following commands are recognized (* =>'s unimplemented).

USER PORT STOR MSAM* RNTO NLST MKD CDUP

PASS PASV APPE MRSQ* ABOR SITE XMKD XCUP

ACCT TYPE MLFL* MRCP* DELE SYST RMD STOU

SMNT* STRU MAIL* ALLO CWD STAT XRMD SIZE

REIN* MODE MSND* REST XCWD HELP PWD MDTM

QUIT RETR MSOM* RNFR LIST NOOP XPWD

214 End of help



Note: To get a single line description of each command, type help followed by a space and the command of which you want a description.



One thing to remember here is that to execute any command from the remote FTP commands list you need to make use of the ‘literal’ keyword. What I mean by that is that all remote FTP commands have to be preceded by the word ‘literal’. For example, say you want to execute the remote FTP command: ‘stat’, then you would type:



ftp> literal stat



***************

HACKING TRUTH: According to FTP help, the literal command is described as:

ftp> help literal

literal send arbitrary ftp command

***************



Anyway, amongst the remote FTP commands, the commands of interest to us are-: ‘stat’ and ‘syst’. Let us see what they return when executed-:



ftp>literal stat

211- ftp2.xxx.bol.net.in FTP server status:

Version 5.60

Connected to 203.xx.251.198 (203.xx.251.198)

Logged in anonymously

TYPE: ASCII, FORM: Nonprint; STRUcture: File; transfer MODE: Stream

211- No data connection

211 End of status



Note: The IP address is of xxx.bol.net.in and not your machine.



ftp> literal syst

215 UNIX Type: L8 Version: BSD-198911



Voila, we get the Operating System name running on ftp2.xxx.bol.net.in. At last some useful information.



Finger and HTTP both failed, what do we do now? Let us turn to the den of the Buggiest daemon on Earth i.e. Sendmail: Port 25, the SMTP port.








Jai Shree Raam

Port Scaning Part 5

We are immediately greeted by the FTP daemon banner, which tells us that this is the FTP server where, people using MTNL’s (My ISP) Internet Services, can upload their site. Now, normally FTP daemon banners are more informative than this one. They usually do give away the name of the Operating System running and also the FTP daemon running. Well, actually it is the login prompts of the daemon banner which gives us the Operating System running on it. Normally, a typical daemon banner would have the following Login prompt:




220 xxx2.bol.net.in FTP server (Digital UNIX Version 5.60) ready.

User (bol.net.in :( none)):



Notice the System name in the brackets on the first line. However, normally almost all FTP daemons are better configured (that is the case in the example target system: xxx.bol.net.in) and their login prompt is somewhat like the below:



220 ftp2.xxx.bol.net.in FTP server ready.

User (mail2.bol.net.in :( none)):



See, no Operating System name. However, with the help of some kewl commands, such systems too can be reveal the OS running on them. However, before we go on, there is one thing that you have to be clear about. Now, we had FTP’ed to xxx.bol.net.in, so you normally expect to connect to Port 21 of xxx.bol.net.in, however that is not true. (Atleast in this case.) If you look at the daemon banner again, then you would notice that the last line says:



220 ftp2.xxx.bol.net.in FTP server ready.



Now how did that happen? Well, is Port 21 not open on xxx.bol.net.in ? Well, no and yes. What actually happens is that, Port 21 of xxx.bol.net.in is open and a daemon there is listening for connections. As soon as a connection is established, it transfers the control or connected the visitor to ftp2.xxx.bol.net.in, which is on the same network as xxx.bol.net.in. Now this, ftp.xxx.bol.net.in system is solely a FTP machine. It has no other services running. So whatever information, we gather from such a FTP port is not of xxx.bol.net.in but of ftp2.bol.net.in. Get it?



Anyway, when you get the login prompt, then login anonymously with the anonymous as the Username and a false email address as the password.



220 ftp2.xxx.bol.net.in FTP server ready.

User (ftp2.xxx.bol.net.in:(none)): anonymous

331 Guest login ok, send your complete e-mail address as password.

Password: xxx@linux.net

230 User anonymous logged in. Access restrictions apply.



Even if you have an account at the FTP server into which you plan to break in, it is always better not to use that pair of Username and Password. Logging in anonymously has many advantages. Say if you did cause some harm to the target system and if you use your (Nonanonymous) Username and Password pair, then if you were not able to edit the server logs

Jai Shree Raam

Port Scaning Part 4

I get the NOT FOUND error message; this probably means that this system does not support CGI-Scripts. If the CGI-Bin directory had been blocked from public access, then we would probably have gotten the Forbidden Error Message.

However, finding out that our target system does have the CGI-Bin directory cannot be said to be disappointing as the known CGI exploits are almost primitive and finding out new exploits should be kept out of this manual.



OK, so Port 80 and Port 79 are ruled out, they neither have any vulnerability nor do they give any information about the target system. [Well actually the HTTP port does give us some valuable information, but we will come to that later.]



Anyway, so let us try Port 21 or the FTP port. Now, there are two ways of connecting to Port 21 of a host, the first one is to telnet to Port 21 and other one is to use the MS-DOS FTP client. You could choose any of the two for this section, however, I kind of like the command line FTP client, although many people say it is lame. Anyway, so I launch up a FTP connection to xxx.bol.net.in.



C:\windows>ftp xxx.bol.net.in



Connected to xxx.bol.net.in.

220-

220-#*************************************************************

220-# Welcome to MTNL's ftp site

220-#*************************************************************

220-#

220-# You can upload your own homepages at this site!!!

220-#

220-# Just login with your username and upload the HTML pages.

220-# (You can use your favourite HTML editor as well)

220-#

220-# World will see it at http://web2.mtnl.net.in/~yourusername/

220-#

220-# So get going......UNLEASH YOUR CREATIVITY !!!!

220-#

220-#*************************************************************

220-

220 ftp2.xxx.bol.net.in FTP server ready.

Jai Shree Raam

Vertual port scaning Part 3

Some ISP's are quite aware of Hacking Activities and are one step ahead. They may be running some excellent software, which will keep hackers away. Ether Peek is an excellent example of sniffing software, which can easily trace users who are port scanning. Nuke Nabber a Windows freeware claims to be able to block Port Scans. I have not tested it so I can't say for sure. Then there is another fun program known as


Port Dumper, which can fake daemon (services) like Telnet, Finger etc. There is also some software, which will show a weird list of open ports. What I mean by that is, if you port scan a host running such software, then it will keep showing random open ports, and you port Scanning Software will go crazy.



Anyway, so once you get a list of open ports, start analyzing the weak points or the services which might help us to get more information about the target system which would prove invaluable to the breaking in process. Try to exploit the commands or the options available on each open port to either find a vulnerability, which could be exploited, or some kind of information on the target system. That is pretty much the only kind of things that we would be looking for. Now, let me explain how I try to find out such things with the list of open ports (of my ISP) and services running on them.



Note: Before proceeding, refer to the table of open ports which we got earlier(of host xxx.bol.net.in) in the manual and yes, I am starting from Port 79 as if I start from Port 21, then the manual will become very very short.



It has Port 79 open or in other words, has finger running, however, almost all Finger daemon are configured to not return much information about Users, however, let us try some common Finger exploits which can sometimes very very rarely get you root.



finger root

finger system

finger



These exploits are very very old and do not work almost 99 times out of 100. So the Finger port is ruled out.



Now let us move on, in the list of open ports, the HTTP port or Port 80 is also open, this means that this target system probably maintains a web site. So let me launch my favorite browser (Internet Explorer, if you are interested.) and see what they have on their site. Well, actually we are not even remotely interested in what they have on their site, but what we are interested in is to see, whether they have the CGI-BIN directory open to public or not, an dif yes if any of the common CGI exploits, which get you root, work or not.



So I type in the following in the URL box of my browser:



http://xxx.bol.net.in/cgi-bin









Jai Shree Raam

vertual port scaning part 2

Anyway, let me assume that you have got hold of a good ‘impossible to detect’ Port Scanner, now scan the target system for all open ports and record the open lists:




Note: In this manual, I have taken up my ISP as an example target system. It would be foo-barred throughout as xxx.bol.net.in



In my case, I found that the following ports were open:



Port Number Service



21 FTP

23 Telnet

25 SMTP

53 DNS

79 Finger

80 HTTP

110 POP

111 Not Useful

389 Not Useful

512 rlogin



Note: Only a few Port Scanners give you both the open Ports and the services running on them. Most Port Scanners only return the list of Open Ports. This is fine too; as once you get the list of open ports then you can find out the corresponding services running on them, referring to the RFC 1700. It contains the complete list of Port Numbers and the corresponding popularly running services.



Now port scanning takes advantage the 3-stage TCP handshake to determine what ports are open on the

Remote computer. To learn more about the TCP\IP protocol read the networking manuals that I distribute on

My mailing list.



Tools like SATAN and lots of them more allow you to find out the list of open ports, the daemon or the service running at each open port and also the service's vulnerability at the click of a button. You can't call yourself a hacker if you need some Software, which first of all is not written by you to do something as lame as a port scan. Well yes I do agree that looking for open ports manually on a server would take a long time. But what I am suggesting is that you use a Port Scanning tool, which just gives you a list of open ports without the list of services and the vulnerabilities. I assure you, if you try and explore an open port of a remote server manually, you will be able to learn more about the remote system and also it will give you a taste of what hacking actually is. If you use a port scanner, which gives you all details at the click of a button to impress your friends, let me assure you none of them will be impressed, as I am sure anyone can use SATAN and other such scanners.



Another thing you need to be careful about before port scanning your ISP is that most port scanners are very easily detected and can easily be traced and you have no excuse if you are caught doing a port scan on a host., it a sure sign of Hacker Activity. There are many stealth scanners like Nmap, which claim to be untraceable. But the truth is that they are very much traceable and they are quite inaccurate as they send only a single packet to check if a port is open or not. And if the host is running the right kind of Sniffer software maybe Etherpeek then the Port scan can be easily detected and the IP of the user logged. Anyway some ISP's are really afraid of Hacking activities and even at the slightest hint of some suspicious hacking activity something like Port scanning, they can disable your account. So just be careful.

************

Evil Hacking Trick: Well try to keep an eye on TCP port 12345, and UDP port 31337 these are the default

ports for the popular trojans NetBus and BO, respectively

*************







Jai Shree Raam

Virtual Port Scaning Part 1

HACKING TRUTH: There are two types of ports. There are hardware ports, which are the slots existing behind the CPU cabinet of your system, into which you plug-in or connect your hardware to. For Example, COM1, COM2, Parallel Port etc. However, we are not interested in such ports. We are concerned with the other type of ports, which are the virtual or the software ports. Such a virtual port is basically a virtual pipe through which information goes in and out. And all open ports have a service or daemon running on it. A service or a daemon is nothing but the software running on these ports, which provide a certain service to the users who connect to it. For Example, Port 25 is always open on a server handling mails, as it is port where the Sendmail service is running by default.


**********************



So basically the first step in your quest to breaking into a system is to get as much information on it, as you can. Try to get, the list of open ports, the list of services running on the respective open ports and whole lots of other kind of information to which I will come later.



Anyway, so firstly, get a good Port Scanner, preferably stealth and then do a port scan on the target host. Now one thing that you must remember while doing a port scan is the fact that there are various so called ‘stealth’ port scanners around which claim to be undetectable, however most of them are detectable. So instead of using such’ false claims’ port scanners, I suggest you code one on your own.



But why do I need to use a stealth Port Scanner and how can I code my own Port Scanner? Well, the reason as to why you need a stealth port Scanner is that many system administrators log all port scans and records the IP and other information on such attempts; this makes you susceptible to getting caught. In my opinion the best Port Scanners around are those, which send SYN/FIN packets from a spoofed host, making logging useless. Such a port Scanner would be coded in C, but will not run in Windows. This was just an idea, now it is up to you to code it yourself.







Jai Shree Raam