text
stringlengths 4
429
|
---|
repeated against a changing |
target space till enough |
computers are compromised. |
Attachment types used |
Half of all the attachments |
analyzed used an extension that |
would run directly when double |
clicked. This old method is still |
the most common scheme used. |
Of all attachments analyzed, |
38 percent were .exe and 12 |
percent were .src files. In total |
only 6 percent used double |
extensions like .pdf.exe to fool |
the user. It should also be noted Figure 4: Average number of mail accounts targeted per company |
that 23 percent were Microsoft |
Word documents using some exploit to execute custom code on the computer. |
There were also some more exotic |
extensions used like AutoIt scripts |
(.au3) and ZX-Edit files (.zed), but |
these are the exception rather |
than the rule. It might be that the |
attackers tried to bypass some email |
filtering software by experimenting |
with different attachment types. |
Sometimes even older exploits like |
the Microsoft DirectX DirectShow |
Length Record Remote Code |
Execution Vulnerability (CVE- |
2009-1539) in .mp4 files are still |
occasionally used. This indicates |
that either not all attackers have |
the knowhow to use newer exploits |
that are publicly available or they |
speculate that the target has not |
patched all computers. Some of |
the attackers do not seem to be too |
sophisticated. For example they used |
www.[COMPANY-NAME].com.exe |
an attachment name, clearly missing |
that the |
.com |
at the end would be |
sufficient to run it and the additional |
.exe |
was not needed. |
Figure 5: Extensions used in targeted attack emails |
Page 23 |
Targeted Attacks Against the Energy Sector |
B. Visualization with TRIAGE |
To identify a series of targeted attacks that are likely performed by the same individuals, we have used a novel attack |
attribution methodology named TRIAGE. |
Developed by Symantec Research Labs, |
TRIAGE is data mining software that |
relies on |
multi-criteria decision analysis and intelligent data fusion algorithms to reliably link different attacks to the |
same source. This framework has been |
developed in order |
to automate cyberintelligence tasks and reduce the time |
needed to get insights into organized cybercrime activities. By enabling rapid analysis of large security data sets, |
Symantec analysts can then quickly and more efficiently attribute various waves of cyberattacks to a specific attack |
campaign likely run by the same individuals. |
Figure 6: Graph view of attack wave against company targeted in the New Year |
s campaign |
The TRIAGE framework was recently enhanced with novel visualizations |
thanks to VIS-SENSE, a European research |
project aiming at developing visual analytics technologies for network security applications. |
Page 24 |
Targeted Attacks Against the Energy Sector |
Figure 7: Visualization graph of the Greek oil campaign |
Since its original conception, |
TRIAGE has been |
successfully used to analyze the behavior of |
cybercriminals involved in various types of Internet |
attack activities, such as rogue antivirus websites [1], |
spam botnets operations [2], scam campaigns [3] and |
targeted attacks performed via spear phishing emails |
[4,5]. |
C. Phases of targeted attacks |
As with any other targeted attacks, attacks against |
the energy sector often follow the same pattern. It |
can be broken down in different phases of attack. It |
should be noted that we have seen attackers modify |
their behavior and exceptions from the norm and this |
is possible especially if the target company has special |
circumstances or security measures in place. |
Figure 8: Typical phases of targeted attacks |
Page 25 |
Targeted Attacks Against the Energy Sector |
Reconnaissance phase |
During this phase the attacker tries to learn as much as possible about the targeted organization. |
Information sources often include social networks, job posting sites and press releases. This enables the attacker |
to learn the contact details of possible target individuals as well as context that can be used in social engineering |
scenarios. The attacker will often create a list of implemented security software used at the targeted company |
from whatever information is available. These investigations often start completely passively without any direct |
contact with the company, since there are many data sources publicly available. Subsequently the attacker can use |
more interaction if needed. Some attackers go through all the effort of creating a fake social media account and |
befriending key employees. After a period of small talk, to create a false sense of security, such a connection can |
then be used to pass on an infected document or find out about some key information. Depending on the targeted |
location, physical reconnaissance and eavesdropping may also be used. |