clydeiii/cybersecurity · Datasets at Hugging Face

text stringlengths 4 429
x
.tex
e
e
e
e
e
e
e
e
0
10 011 011 01
10 011 001 01
EXECUTIVE
SUMMARY
BAE Systems Applied Intelligence: Snake Rootkit Report 2014 2
OVERVIEW
One of the questions which comes up in the months after big security whitepaper disclosures is:
where are they now? In other words, what happened to the operators, tools, and infrastructure
which was revealed in the reports, blog-posts, and press interviews.
Did they continue on as before, did they re-build the disclosed infrastructure and tools, did they go
away and get jobs in another line of work?
In some cases, the disclosure had little, if any impact on the operation. For example, after the
McAfee ShadyRAT report in 2011, there was absolutely no change in the attacks from the group
behind this. However, when Mandiant released their APT1 report in 2013, there was a noticeable
reduction in activity from the group
and much of the tools and infrastructure has not been seen
since.
In the September 2010 issue of Foreign Affairs magazine1, former US Deputy Secretary of Defense
William J. Lynn discussed a cyber-attack which happened two years previously on the DoD
classified computer networks. Lynn described how a foreign intelligence agency planted malicious
code on the networks with the aim of transferring data to servers under their control.
The article included the now oft-quoted phrase
digital beachhead
to describe what was
undoubtedly a significant compromise of US military systems. Further reports in the press2 kept the
story alive in 2011, but since then this threat has received remarkably little attention.
However, the operation behind the attacks has continued with little modification to the tools and
techniques, in spite of the widespread attention a few years ago. They use highly sophisticated
malware tools to maintain persistent access to their targets. These tools can be used for covert
communications in a number of different modes, some of which present significant challenges for
traditional security technologies to detect.
There are some threats which come and go, whilst there are others which are permanent
features of the landscape. In this paper, we describe the tools and techniques of one of the most
sophisticated and persistent threats we track. We hope this will help victims identify intrusions and
understand their need to improve defences. Cyber security is a collaborative effort
the operation
described in this paper again raises the bar for the security community in their efforts to keep up
with the attackers in cyber-space
http://www.foreignaffairs.com/articles/66552/william-j-lynn-iii/defending-a-new-domain
http://www.reuters.com/article/2011/06/17/us-usa-cybersecurity-worm-idUSTRE75F5TB20110617
BAE Systems Applied Intelligence: Snake Rootkit Report 2014 3
TECHNICAL
DESCRIPTION
BAE Systems Applied Intelligence: Snake Rootkit Report 2014 4
BACKGROUND
When antivirus back-end classification platforms cannot identify a malware family for an analysed malicious sample, they assign
generic names, such as
Trojan Horse
or
Agent
. The variant letters are also assigned automatically, by using hexavigesimal
(or Base26) notation. That is, the variant letters are auto-assigned starting from
, followed with
, and so on until
. Next
comes
and so on, until
. After that, the variant letters start from
and so on, until
Back in 2008 an unknown malicious file was discovered and auto-classified as
Agent.BTZ
, meaning it was registered as unknown
malicious sample #1,898 in an anti-virus classification system. It wasn
t given an actual name, only a generic one.
Meanwhile, internally the authors behind this malware were using their own naming systems - with specific titles for their file
components and projects such as
snake
uroburos
sengoku
, and
snark
used to denote variants of their framework.
A recent report from German security company GData3 described a sample from the
uroburos
variant of this framework. Their
report revealed the complex nature of this malware family, and showed that the operation behind
Agent.BTZ
has continued. As
a result of this disclosure, we are also releasing our own technical analysis of the threat, including a timeline of known samples,
known Command-and-Control (C&C) servers, and other indicators to aid investigators in discovering attacks.
Reverse engineering of recent malware samples shows these to be much more advanced variants of Agent.BTZ, though still sharing
many similarities and encryption methods with the original. Further investigation allowed us to locate related samples compiled
between 2006 and 2014, and spanning across several distinctive generations. The first section of this report gives an overview of
the samples collected, where they were reported and the timelines derived from their analysis.
Snake
s architecture turned out to be quite interesting. We have identified two distinct variants, both highly flexible but with two
different techniques for establishing and maintaining a presence on the target system. In general, its operation relies on kernel
mode drivers, making it a rootkit. It is designed to covertly install a backdoor on a compromised system, hide the presence of its
components, provide a communication mechanism with its C&C servers, and enable an effective data exfiltration mechanism. At the

x

.tex

e

0

10 011 011 01

10 011 001 01

EXECUTIVE

SUMMARY

BAE Systems Applied Intelligence: Snake Rootkit Report 2014 2

OVERVIEW

One of the questions which comes up in the months after big security whitepaper disclosures is:

where are they now? In other words, what happened to the operators, tools, and infrastructure

which was revealed in the reports, blog-posts, and press interviews.

Did they continue on as before, did they re-build the disclosed infrastructure and tools, did they go

away and get jobs in another line of work?

In some cases, the disclosure had little, if any impact on the operation. For example, after the

McAfee ShadyRAT report in 2011, there was absolutely no change in the attacks from the group

behind this. However, when Mandiant released their APT1 report in 2013, there was a noticeable

reduction in activity from the group

and much of the tools and infrastructure has not been seen

since.

In the September 2010 issue of Foreign Affairs magazine1, former US Deputy Secretary of Defense

William J. Lynn discussed a cyber-attack which happened two years previously on the DoD

classified computer networks. Lynn described how a foreign intelligence agency planted malicious

code on the networks with the aim of transferring data to servers under their control.

The article included the now oft-quoted phrase

digital beachhead

to describe what was

undoubtedly a significant compromise of US military systems. Further reports in the press2 kept the

story alive in 2011, but since then this threat has received remarkably little attention.

However, the operation behind the attacks has continued with little modification to the tools and

techniques, in spite of the widespread attention a few years ago. They use highly sophisticated

malware tools to maintain persistent access to their targets. These tools can be used for covert

communications in a number of different modes, some of which present significant challenges for

traditional security technologies to detect.

There are some threats which come and go, whilst there are others which are permanent

features of the landscape. In this paper, we describe the tools and techniques of one of the most

sophisticated and persistent threats we track. We hope this will help victims identify intrusions and

understand their need to improve defences. Cyber security is a collaborative effort

the operation

described in this paper again raises the bar for the security community in their efforts to keep up

with the attackers in cyber-space

http://www.foreignaffairs.com/articles/66552/william-j-lynn-iii/defending-a-new-domain

http://www.reuters.com/article/2011/06/17/us-usa-cybersecurity-worm-idUSTRE75F5TB20110617

BAE Systems Applied Intelligence: Snake Rootkit Report 2014 3

TECHNICAL

DESCRIPTION

BAE Systems Applied Intelligence: Snake Rootkit Report 2014 4

BACKGROUND

When antivirus back-end classification platforms cannot identify a malware family for an analysed malicious sample, they assign

generic names, such as

Trojan Horse

or

Agent

. The variant letters are also assigned automatically, by using hexavigesimal

(or Base26) notation. That is, the variant letters are auto-assigned starting from

, followed with

, and so on until

. Next

comes

and so on, until

. After that, the variant letters start from

and so on, until

Back in 2008 an unknown malicious file was discovered and auto-classified as

Agent.BTZ

, meaning it was registered as unknown

malicious sample #1,898 in an anti-virus classification system. It wasn

t given an actual name, only a generic one.

Meanwhile, internally the authors behind this malware were using their own naming systems - with specific titles for their file

components and projects such as

snake

uroburos

sengoku

, and

snark

used to denote variants of their framework.

A recent report from German security company GData3 described a sample from the

uroburos

variant of this framework. Their

report revealed the complex nature of this malware family, and showed that the operation behind

Agent.BTZ

has continued. As

a result of this disclosure, we are also releasing our own technical analysis of the threat, including a timeline of known samples,

known Command-and-Control (C&C) servers, and other indicators to aid investigators in discovering attacks.

Reverse engineering of recent malware samples shows these to be much more advanced variants of Agent.BTZ, though still sharing

many similarities and encryption methods with the original. Further investigation allowed us to locate related samples compiled

between 2006 and 2014, and spanning across several distinctive generations. The first section of this report gives an overview of

the samples collected, where they were reported and the timelines derived from their analysis.

Snake

s architecture turned out to be quite interesting. We have identified two distinct variants, both highly flexible but with two

different techniques for establishing and maintaining a presence on the target system. In general, its operation relies on kernel

mode drivers, making it a rootkit. It is designed to covertly install a backdoor on a compromised system, hide the presence of its

components, provide a communication mechanism with its C&C servers, and enable an effective data exfiltration mechanism. At the