clydeiii/cybersecurity · Datasets at Hugging Face

text stringlengths 4 429
same time, Snake exposed a flexibility to conduct its operations by engaging these noticeably different architectures.
In the first model, the network communications are carried out from the userland - i.e. the area of the computer system where
application software executes. In another model, the network communications are handled by a kernel mode driver - i.e. the area
where lower level system code such as device drivers run. The choice of what architecture should be used may depend on a specific
target
s environment, allowing the Snake operators to choose the most suitable architecture to be deployed.
In both architectures there is a kernel mode driver installed and a usermode DLL injected by the driver into the system processes.
In both architectures, there is both 32-bit and 64-bit code involved. In order to distinguish between these architectures, we will call
them the usermode-centric and the kernel-centric architectures respectively.
The remainder of this report gives a detailed explanation of how the two Snake architectures embed themselves in the target
system and communicate with the outside world. We have also provided a set of technical indicators in the Appendix to enable
organisations and the security research community to identify compromises.
ERROR: type should be string, got " https://www.gdata.de/rdk/dl-en-rp-Uroburos"
BAE Systems Applied Intelligence: Snake Rootkit Report 2014 5
SNAKE SAMPLES
In total we have collected over 100 unique files related to this espionage toolkit. Many of these were submitted to online malware
analysis websites by victims and investigators over several years. In many cases the source country information of the submission
is available. These allow us to visualise the distribution of countries where this malware has been seen:
#Samples Submission Year
Source country 2010 2011 2012 2013 2014 Total
Ukraine 1 3 6 8 14 32
Lithuania 9 2 11
Great Britain 4 4
Belgium 2 2
Georgia 2 2
United States 1 1 2
Romania 1 1
Hungary 1 1
Italy 1 1
Total 1 4 7 24 20 56
Whilst this view is likely to only be the tip of the iceberg, it does give us an initial insight into the profile of targets for the Snake
operations.
Other useful visualisations of the operations come from the compile timestamps. Below is shown a table with a count of the
number of files in our sample set from recent years. Two samples compiled in late January 2014 show that this activity is ongoing.
#Samples by compile month
Year 01 02 03 04 05 06 07 08 09 10 11 12 Total
2006 1 3 4
2007 1 1 1 3
2008 2 1 2 1 2 8
2009 1 1 1 3 2 2 10
2010 1 1 1 1 1 2 7
2011 1 4 1 3 1 3 13
2012 2 1 1 1 2 7 14
2013 1 13 5 2 5 4 3 2 1 2 1 39
2014 2 2
Total 8 15 9 4 9 13 6 2 1 11 10 12 100
Plotting the day of the week in which the samples were compiled shows a now familiar pattern for analysts of modern cyber-attacks.
The creators of the malware operate a working week, just like any other professional. The single sample in our set which was
compiled on a Saturday is an outlier, but doesn
t alter the conclusion. Similarly, plotting the hour of the day in which the samples
were compiled reveals another human pattern
the working day. This has been adjusted to UTC+4 to show a possible fit to the
operators
local time.
#Samples compiled per day of the week #Samples compiled by hour of the day
(adjusted to UTC+4)
Mon Tue Wed Thu Fri Sat Sun
BAE Systems Applied Intelligence: Snake Rootkit Report 2014 6
USERMODE-CENTRIC ARCHITECTURE
The usermode-centric architecture of Snake is known to have been used from 2011 till 2014, with the most recent sample compiled
on January 28, 2014.
With this architecture, the Snake driver is mainly used to load the DLL module into the usermode processes, and then use that
module for the communications.
One of the analysed samples exposed multiple debug messages and source control check-in logs. It is not clear why those
messages were allowed in the deployed driver - possibly an operational security lapse. However, they give some insight into the
internal structure of the source code.
For example, the analysed driver gave away the following source file names:

d:\proj\cn\fa64\common\loadlib\common/loadlib_helpers.c

d:\proj\cn\fa64\common\loadlib\win/loadlib.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/libunhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\ntsystem/libhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/hook_helpers.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/libhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/idthook.c

.\rk_ntsystem.c

..\common\helpers\interface_s.c

..\k2\fa_registry.c

..\k2\syshook.c
The source control check-in log examples, showing the names of the developers to be
vlad
and
gilg

$Id: snake_config.c 5204 2007-01-04 10:28:19Z vlad $

$Id: mime64.c 12892 2010-06-24 14:31:59Z vlad $

$Id: event.c 14097 2010-11-01 14:46:27Z gilg $

same time, Snake exposed a flexibility to conduct its operations by engaging these noticeably different architectures.

In the first model, the network communications are carried out from the userland - i.e. the area of the computer system where

application software executes. In another model, the network communications are handled by a kernel mode driver - i.e. the area

where lower level system code such as device drivers run. The choice of what architecture should be used may depend on a specific

target

s environment, allowing the Snake operators to choose the most suitable architecture to be deployed.

In both architectures there is a kernel mode driver installed and a usermode DLL injected by the driver into the system processes.

In both architectures, there is both 32-bit and 64-bit code involved. In order to distinguish between these architectures, we will call

them the usermode-centric and the kernel-centric architectures respectively.

The remainder of this report gives a detailed explanation of how the two Snake architectures embed themselves in the target

system and communicate with the outside world. We have also provided a set of technical indicators in the Appendix to enable

organisations and the security research community to identify compromises.

ERROR: type should be string, got " https://www.gdata.de/rdk/dl-en-rp-Uroburos"

BAE Systems Applied Intelligence: Snake Rootkit Report 2014 5

SNAKE SAMPLES

In total we have collected over 100 unique files related to this espionage toolkit. Many of these were submitted to online malware

analysis websites by victims and investigators over several years. In many cases the source country information of the submission

is available. These allow us to visualise the distribution of countries where this malware has been seen:

#Samples Submission Year

Source country 2010 2011 2012 2013 2014 Total

Ukraine 1 3 6 8 14 32

Lithuania 9 2 11

Great Britain 4 4

Belgium 2 2

Georgia 2 2

United States 1 1 2

Romania 1 1

Hungary 1 1

Italy 1 1

Total 1 4 7 24 20 56

Whilst this view is likely to only be the tip of the iceberg, it does give us an initial insight into the profile of targets for the Snake

operations.

Other useful visualisations of the operations come from the compile timestamps. Below is shown a table with a count of the

number of files in our sample set from recent years. Two samples compiled in late January 2014 show that this activity is ongoing.

#Samples by compile month

Year 01 02 03 04 05 06 07 08 09 10 11 12 Total

2006 1 3 4

2007 1 1 1 3

2008 2 1 2 1 2 8

2009 1 1 1 3 2 2 10

2010 1 1 1 1 1 2 7

2011 1 4 1 3 1 3 13

2012 2 1 1 1 2 7 14

2013 1 13 5 2 5 4 3 2 1 2 1 39

2014 2 2

Total 8 15 9 4 9 13 6 2 1 11 10 12 100

Plotting the day of the week in which the samples were compiled shows a now familiar pattern for analysts of modern cyber-attacks.

The creators of the malware operate a working week, just like any other professional. The single sample in our set which was

compiled on a Saturday is an outlier, but doesn

t alter the conclusion. Similarly, plotting the hour of the day in which the samples

were compiled reveals another human pattern

the working day. This has been adjusted to UTC+4 to show a possible fit to the

operators

local time.

#Samples compiled per day of the week #Samples compiled by hour of the day

(adjusted to UTC+4)

Mon Tue Wed Thu Fri Sat Sun

BAE Systems Applied Intelligence: Snake Rootkit Report 2014 6

USERMODE-CENTRIC ARCHITECTURE

The usermode-centric architecture of Snake is known to have been used from 2011 till 2014, with the most recent sample compiled

on January 28, 2014.

With this architecture, the Snake driver is mainly used to load the DLL module into the usermode processes, and then use that

module for the communications.

One of the analysed samples exposed multiple debug messages and source control check-in logs. It is not clear why those

messages were allowed in the deployed driver - possibly an operational security lapse. However, they give some insight into the

internal structure of the source code.

For example, the analysed driver gave away the following source file names:

d:\proj\cn\fa64\common\loadlib\common/loadlib_helpers.c

d:\proj\cn\fa64\common\loadlib\win/loadlib.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/libunhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\ntsystem/libhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/hook_helpers.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/libhook.c

d:\proj\cn\fa64\uroboros\rk_common\libhook\common/idthook.c

.\rk_ntsystem.c

..\common\helpers\interface_s.c

..\k2\fa_registry.c

..\k2\syshook.c

The source control check-in log examples, showing the names of the developers to be

vlad

and

gilg

$Id: snake_config.c 5204 2007-01-04 10:28:19Z vlad $

$Id: mime64.c 12892 2010-06-24 14:31:59Z vlad $

$Id: event.c 14097 2010-11-01 14:46:27Z gilg $