Anti Virus Scan

  • Subscribe to our RSS feed.
  • Twitter
  • StumbleUpon
  • Reddit
  • Facebook
  • Digg
Showing posts with label prefetch. Show all posts
Showing posts with label prefetch. Show all posts

Wednesday, 5 December 2012

NTOSBOOT Prefetch File

Posted on 19:47 by Unknown
Knowing the programs that executed on a system can answer numerous questions. The answers can help on a range of cases from acceptable use policy violations to investigations to intrusions to malware. One program execution artifact is prefetch files which have been well documented including in the article Decoding Prefetch Files for Forensic Purposes and on the Prefetch Wiki page. The purpose of this post is to explore the NTOSBOOT prefetch file and how it is relevant for malware investigations.

What is the NTOSBOOT Prefetch File


The Windows operating system enables prefetching to make system boots or applications startups faster. Windows Internals, Part 2: Covering Windows 2008 Server R2 and Windows 7 states “the prefetcher tries to speed the boot process and application startup by monitoring the data and code accessed by boot and application startups and using that information at the beginning of a subsequent boot or application startup to read in the code and data”. Prefetch files are what store the information required for faster startups.

The NTOSBOOT prefetch file stores the information related to the boot process. The information in this prefetch file reflects the files accessed during system startup. Specifically, the files accessed, according to the Windows Internals, “from system start through the 30 seconds following the start of the user’s shell (typically Explorer) or, failing that, up through 60 seconds following Windows service initialization or through 120 seconds, whichever comes first”.

Relevance for Malware Investigations


The NTOSBOOT prefetch file stores information about the files accessed during the boot process and in some cases 30 seconds following a user logging onto a system. Remember the Rootkit Paradox, malware wants to hide but it must run. If a piece of malware runs during the boot process then there will be a reference for the malware in the NTOSBOOT prefetch file. To illustrate this point I examined a NTOSBOOT prefetch file to show how it contains traces of malware execution. (side note: to view the file handles in prefetch files one could use Strings or Harlan’s pref.pl)

The NTOSBOOT prefetch file in question came from a system I infected last year by opening the attachment in a NYS Department of Motor Vehicles Uniform Traffic Ticket email. Looking through the file handles for any unusual file paths (files in temporary folders, unusual file names, etc) I came across three suspicious files as shown below.




The first file 17F7FFF4.COM was suspicious since the file was located in the All Users temporary folder. The second file KB961710.exe was suspicious because Windows updates are not stored in the application data folders. The last file SVCHOST.EXE was suspicious since not only was it located in a temp folder but the svchost.exe file is normally located in the Windows\System32 folder. I located these files by solely reviewing the file handles referenced in the NTOSBOOT prefetch file and the screenshot below confirms my suspicions.


Summary


Program execution artifacts are vital when examining a system infected with malware. Despite malware’s best efforts to hide on a system at some point for them to do their jobs they have to run. When malware eventual runs there will be traces of its execution left on a system which can be used to find its hiding spot. The NTOSBOOT prefetch file is one such artifact and the file handle information will reveal what’s hiding on a system.
Read More
Posted in malware, malware analysis, prefetch | No comments

Wednesday, 11 July 2012

Combining Techniques

Posted on 14:59 by Unknown
“You do intrusion and malware investigations, we do CP and fraud cases” is a phrase I saw Harlan mention a few times on his blog. To me the phrase is more about how different the casework is; about how different the techniques are for each type of case. Having worked both fraud and malware cases I prefer to focus on what each technique has to offer as opposed to their differences. How parts of one technique can be beneficial to different types of cases. How learning just a little bit about a different technique can pay big dividends by improving your knowledge, skills, and process that you can use on your current cases. To illustrate I wanted to contrast techniques for malware and fraud cases to show how they help one another.

Program Execution


Malware eventually has to run and when it does execute then there may be traces left on a system. Understanding program execution and where these artifacts are located is a valuable technique for malware cases. Examining artifacts containing program execution information is a quick way to find suspicious programs. One such artifact is prefetch files. To show their significance I’m parsing them with Harlan’s updated pref.pl script. Typically I start examining prefetch files by first looking at what executables ran on a system and where they executed from. I saw the following suspicious programs looking at the output from the command “pref.exe -e -d Prefetch-Folder”.

TMP77E.EXE-02781D7C.pf Last run: Fri Mar 12 16:29:05 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\TMP77E.EXE

TMPDC7.EXE-2240CBB3.pf Last run: Fri Mar 12 16:29:07 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\TMPDC7.EXE

UPDATE.EXE-0825DC41.pf Last run: Fri Mar 12 16:28:57 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\DESKTOP\UPDATE.EXE

ASD3.TMP.EXE-26CA54B1.pf Last run: Fri Mar 12 16:34:49 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASD3.TMP.EXE

ASD4.TMP.EXE-2740C04A.pf Last run: Fri Mar 12 16:34:50 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASD4.TMP.EXE

DRGUARD.EXE-23A7FB3B.pf Last run: Fri Mar 12 16:35:26 2010 (2)
\DEVICE\HARDDISKVOLUME1\PROGRAM FILES\DR. GUARD\DRGUARD.EXE

ASD2.TMP.EXE-2653E918.pf Last run: Fri Mar 12 16:34:27 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASD2.TMP.EXE

ASR64_LDM.EXE-3944C1CE.pf Last run: Fri Mar 12 16:29:06 2010 (1)
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASR64_LDM.EXE

These programs stood out for a few different reasons. First I noticed the path they executed from was a temporary folder in a user’s profile. Unusual file paths are one way to spot malware on a system. The other thing I noticed is that some of the programs only executed once. This behavior resembles how downloaders and droppers work. Their sole purpose is to execute once to either download additional malware or install malware. The last dead giveaway is they all executed within a few minutes of each other. The first sweep across the prefetch files netted some interesting programs that appear to be malicious. The next thing to look at is the individual prefetch files to see what file handles were open when the program ran. The TMP77E.EXE-02781D7C.pf prefetch file showed something interesting as shown below (the command used was “pref.pl -p -i -f TMP77E.EXE-02781D7C.pf”).

EXE Name : TMP77E.EXE
Volume Path : \DEVICE\HARDDISKVOLUME1
Volume Creation Date: Fri Nov 2 08:56:57 2007 Z
Volume Serial Number: 6456-B1FD

\DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\NTDLL.DLL
\DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\KERNEL32.DLL
\DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\UNICODE.NLS
*****snippet*****


EXEs found:

\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\TMP77E.EXE
\DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\NET.EXE
\DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\SC.EXE
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASR64_LDM.EXE

DAT files found:

\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\LOCAL SETTINGS\TEMPORARY INTERNET FILES\CONTENT.IE5\INDEX.DAT
\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\COOKIES\INDEX.DAT
\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\LOCAL SETTINGS\HISTORY\HISTORY.IE5\INDEX.DAT

Temp paths found:

\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\TMP77E.EXE
\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\LOCAL SETTINGS\TEMPORARY INTERNET FILES\CONTENT.IE5\INDEX.DAT
\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\LOCAL SETTINGS\TEMPORARY INTERNET FILES\CONTENT.IE5\M20M2OXX\READDATAGATEWAY[1].HTM
\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\ASR64_LDM.EXE

The pref.pl file handle portion to the output was trimmed to make it easier to read but I left in the intelligence provided by the script through its filters. The filters highlight file handles containing exes, dats, and temporary folders. The exe filter shows in additional to a handle to the TMP77E.EXE file there were handles to SC.EXE, NET.EXE, and ASR64_LDM.EXE. SC.EXE is a Windows program for managing services including creating new services while NET.EXE is a Windows program for doing various tasks including starting services. ASR64_LDM.EXE was another suspicious program that ran on the system after TMP77E.EXE. The file handles inside each prefetch file of interest provided additional information which is useful during a malware case.

Program execution is vital for malware cases and I saw how the same technique can apply to fraud cases. On fraud cases a typical activity is to identify and locate financial data. At times this can be done by running keyword searches but most of the time (at least for me) the actual financial data is unknown. What I mean by this is a system is provided and it’s up to you to determine what data is financial. This is where the program execution technique comes into play. The programs that ran on the system can be reviewed to provide leads about what kind of financial data may be present on the system. Using the first sweep across the prefetch files I located these interesting programs (command used was “pref.exe -e -d Prefetch-Folder”). Note: the system being looked at is not from a fraud case but it still demonstrates how the data appears.

WINWORD.EXE-C91725A1.pf Last run: Tue Jul 10 16:42:26 2012 (42)
\DEVICE\HARDDISKVOLUME2\PROGRAM FILES\MICROSOFT OFFICE\OFFICE12\WINWORD.EXE

ACROBAT_SL.EXE-DC4293F2.pf Last run: Fri Jun 22 18:14:12 2012 (1)
\DEVICE\HARDDISKVOLUME2\PROGRAM FILES\ADOBE\ACROBAT 9.0\ACROBAT\ACROBAT_SL.EXE

EXCEL.EXE-C6BEF51C.pf Last run: Tue Jul 10 16:30:18 2012 (22)
\DEVICE\HARDDISKVOLUME2\PROGRAM FILES\MICROSOFT OFFICE\OFFICE12\EXCEL.EXE

POWERPNT.EXE-1404AEAA.pf Last run: Thu Jun 21 20:14:52 2012 (22)
\DEVICE\HARDDISKVOLUME2\PROGRAM FILES\MICROSOFT OFFICE\OFFICE12\POWERPNT.EXE

When I look at program execution for fraud cases I look for financial applications, applications that can create financial data, and programs associated with data spoliation. The system didn’t have any financial or data spoliation programs but there were office productivity applications capable of creating financial documents such as invoices, receipts, proposals, etc. These programs were Microsoft Office and Adobe Acrobat and this means the data created on the system is most likely Word, Excel, Powerpoint, or PDFs. The number of executions for each program is also interesting. I look to see what applications are heavily used since it’s a strong indication about what program the subject uses. Notice Adobe only ran once while Word was ran 42 times. Looking at the file handles inside individual prefetch file also contains information relevant to a fraud case. Below are a few sanitized handles I found in the WINWORD.EXE-C91725A1.pf prefetch file (the command used was “pref.pl -p -i -f WINWORD.EXE-C91725A1.pf”).

EXE Name : WINWORD.EXE
Volume Path : \DEVICE\HARDDISKVOLUME2
Volume Creation Date: Fri Aug 26 18:13:26 2011 Z
Volume Serial Number: E4DD-S23A

\DEVICE\HARDDISKVOLUME2\WINDOWS\SYSTEM32\NTDLL.DLL
\DEVICE\HARDDISKVOLUME2\WINDOWS\SYSTEM32\KERNEL32.DLL
\DEVICE\HARDDISKVOLUME2\WINDOWS\SYSTEM32\APISETSCHEMA.DLL
*****snippet*****
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\SANTIZED.DOCX
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\~$SANTIZED.DOCX
\DEVICE\HARDDISKVOLUME2\USERS\USERNAME\APPDATA\LOCAL\TEMP\MSO4F30.TMP
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\SANTIZED 2.DOCX
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\~$SANTIZED 2.DOCX
\DEVICE\HARDDISKVOLUME2\USERS\USERNAME\APPDATA\LOCAL\TEMP\MSO7CF3.TMP
\DEVICE\HARDDISKVOLUME2\USERS\USERNAME\APPDATA\LOCAL\TEMP\20517251.OD
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\SANTIZED 3.DOCX
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\~$SANTIZED 3.DOCX
\DEVICE\HARDDISKVOLUME4\$MFT
\DEVICE\HARDDISKVOLUME2\USERS\USERNAME\APPDATA\LOCAL\MICROSOFT\WINDOWS\TEMPORARY INTERNET FILES\CONTENT.MSO\SANTIZED.JPEG
\DEVICE\HARDDISKVOLUME4\FORENSICS\RESEARCH\Folder\SANTIZED 3.DOCX:ZONE.IDENTIFIER
*****snippet*****

The file handles shows documents stored in the folder FORENSICS\RESEARCH\Folder\ on a different volume were accessed with Word. I think this is significant because not only does it provide filenames to look for but it also shows another storage location the subject may have used. Where ever there is storage accessible to the subject then there’s a chance that is where they are storing some financial data. Also, notice in the output how the last line shows one of the documents was downloaded from the Internet (zone identified alternate data stream).

User Activity


The program execution showed how fraud cases benefited from a technique used in malware cases. Now let’s turn the tables to see how malware cases can benefit from fraud. As I mentioned before most of the times I have to find where financial data is located whether if it’s on the system or in network shares. The best approach I found was to look at artifacts associated with user activity; specifically file, folder, and network share access. My reasoning is if someone is being looked at for committing a fraud and they are suspected of using the computer to commit the fraud then they will be accessing financial data from the computer to carry out the fraud. Basically, I let their user activity show me where the financial data is located and this approach works regardless if the data is in a hidden folder or stored on a network. There are numerous artifacts containing file, folder, and network share access and one of them is link files. To show their significance I’m parsing them with TZWorks LNK Parser Utility. When I examine link files I parse both the Recent and Office/Recent folders. This results in some duplicates but it catches link files found in one folder and not the other. I’ve seen people delete everything in the Recent folder while not realizing the Office\Recent folder exists. I saw some interesting target files, folders, and network shares by running the command “dir C:\Users\Username\AppData\Roaming\Microsoft\Windows\Recent\*.lnk /b /s | lp -pipe -csv > fraud_recent.txt”.

{CLSID_MyComputer}\E:\Forensics\Research\Folder\sanatized.doc
{CLSID_MyComputer}\E:\Forensics\Research\Folder\sanitized 2.doc
{CLSID_MyComputer}\E:\Forensics\Research\Folder\sanitized 3.doc
{CLSID_MyComputer}\C:\Atad\material\sanitized 1.pdf
{CLSID_MyComputer}\F:\Book1.xls
\\192.168.200.55\ share\TR3Secure

The output has been trimmed (only shows target file column) and sanitized since it’s from one of my systems. The link files show files and folders on removable media and a network share has been accessed in addition to a folder not inside the user profile. I’ve used this same technique on fraud cases to figure out where financial data was stored. One time it was some obscure folder on the system while the next time it was a share on a server located on the network.

Tracking user activity is a great way to locate financial data on fraud cases and I saw how this same technique can apply to malware cases. On malware cases it can help answer the question how did the computer become infected. Looking at the user activity around the time of the initial infection can help shed light on what attack vector was used to compromise the system. Did the user access a network share, malicious website, removable media, email attachment, or peer to peer application? The user activity provides indications about what the account was doing that contributed to the infection. On the malware infected system there were only two link files in the Recent folder shown below is the target create time and target name (command used was “dir "F:\Malware_Recent\*.lnk" /b /s | lp -pipe -csv > malware_recent.txt”).

3/12/2010 16:17:04.640 {CLSID_MyComputer}\C:\downloads
3/12/2010 16:18:59.609 {CLSID_MyComputer}\C:\downloads\link.txt

These link files show the user account accessed the downloads folder and the link text file just before the suspicious programs started executing on the system. Looking at this user activity jogged my memory about how the infection occurred. I was researching a link from a SPAM email and I purposely clicked the link from a system. I just never got around to actually examining the system. However, even though the system was infected on purpose examining the user activity on the malware cases I worked has helped answer the question how did the system become infected.

Closing Thoughts


DFIR has a lot of different techniques to deal with the casework we face. Too many times we tend to focus on the differences; the different tools, different processes, and different meanings of artifacts. Focusing on the differences distracts from seeing what the techniques have to offer. What parts of the techniques can strengthen our processes and make us better regardless of what case we are up against. If I didn’t focus on what the techniques had to offer then I would have missed an opportunity. A chance to develop a better DFIR process by combining malware and fraud techniques; a process that I think is far better than if each technique stood on their own.
Read More
Posted in fraud, malware analysis, prefetch, program execution, shortcut files | No comments

Monday, 19 March 2012

Second Look at Prefetch Files

Posted on 18:04 by Unknown
The one thing I like about sharing is when someone opens your eyes about additional information in an artifact you frequently encounter. Harlan has been posting about prefetch files and the information he shared changed how I look at this artifact. Harlan’s first post Prefetch Analysis, Revisited discussed how the artifact contains strings -such as file names and full paths to modules that were either used or accessed by the executable. He also discussed how the data can not only provide information about what occurred on the system but it could be used in data reduction techniques. One data reduction referenced was searching on the file paths for words such as temp. Harlan’s second post was Prefetch Analysis, Revisited...Again... and he expanded on what information is inside prefetch files. He broke down what was inside a prefetch from one of my test systems where I ran Metasploit against a Java vulnerability. His analysis provided more context to what I found on the system and validated some of my findings by showing Java did in fact access the logs I identified. Needless to say, his two posts opened my files to additional information inside prefetch files. Additional information I didn’t see the first the first time through but now I’m taking a second look to see what I find and to test out how one of Harlan's data reduction techniques would have made things easier for me.

Validating Findings

I did a lot of posts about Java exploit artifacts but Harlan did an outstanding job breaking down what was inside one of those Java prefetch files. I still have images from other exploit artifact testing so I took a look at prefetch files from an Adobe exploit and Windows Help Center exploit. The Internet Explorer prefetch files in both images didn’t contain any references to the attack artifacts but the exploited applications’ prefetch files did.

The CVE-2010-2883 (PDF Cooltype) vulnerability is present in the cooltype.dll affecting certain Adobe Reader and Acrobat versions. My previous analysis identified the following: the system had a vulnerable Adobe reader version, a PDF exploit appeared on the system, the PDF exploit is accessed, and Adobe Reader executed. The strings in the ACRORD32.EXE-3A1F13AE.pf prefetch file helped to validate the attack because it shows that Adobe Reader did in fact access the cooltype.dll as shown below.

\DEVICE\HARDDISKVOLUME1\PROGRAM FILES\ADOBE\READER 9.0\READER\COOLTYPE.DLL

The prefetch file from the Windows Help Center URL Validation vulnerability system showed something similar to the cooltype.dll exploit. The Seclists Full disclosure author mentioned that Windows Media Player could be used in an attack against the Help Center vulnerability. The strings in the HELPCTR.EXE-3862B6F5.pf prefetch file showed the application did access a Windows Media Player folder during the exploit.

\DEVICE\HARDDISKVOLUME1\DOCUMENTS AND SETTINGS\ADMINISTRATOR\LOCAL SETTINGS\APPLICATION DATA\MICROSOFT\MEDIA PLAYER\

Finding Malware Faster

Prefetch files provided more information about the exploit artifacts left on a system. By itself this is valuable enough but another point Harlan mentioned was using the strings inside prefetch files for data reduction. One data reduction technique is to filter on files' paths. To demonstrate the technique and how effective it is at locating malware I ran strings across the prefetch folder in the image from the post Examining IRS Notification Letter SPAM. (note, strings is not the best tool to analyze prefetch files and I’m only using the tool to illustrate how data is reduced) I first ran the following command which resulted in 7,905 lines.

strings.exe –o irs-spam-email\prefetch\*.pf

I wanted to reduce the data by only showing the lines containing the word temp to see if anything launched from a temp folder. To accomplish this I ran grep against the strings output which reduced my data to 84 lines (the grep -w switch matches on whole word and –i ignores case).

strings.exe –o irs-spam-email\prefetch\*.pf | grep –w –i temp

The number of lines went from 7,905 down to 84 which made it fairly easy for me to spot the following interesting lines.

\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\TEMPORARY DIRECTORY 1 FOR IRS%20DOCUMENT[1].ZIP\IRS DOCUMENT.EXE

\DEVICE\HARDDISKVOLUME1\DOCUME~1\ADMINI~1\LOCALS~1\TEMP\PUSK3.EXE

Using one filtering technique enabled me to quickly spot interesting executables in addition to the possibly finding the initial infection vector (a malicious zip file). This information was obtained by running only one command against the files inside a prefetch folder. In hindsight, my original analysis on the prefetch files was fairly limited (executable paths, runcounts, and filenames) but going forward I'll look at this artifact and the information they contain in a different light.
Read More
Posted in adobe, exploits, hcp, java, malware analysis, prefetch, program execution | No comments
Older Posts Home
Subscribe to: Posts (Atom)

Popular Posts

  • Ripping Volume Shadow Copies Sneak Peek
    I was hesitant to do a sneak peak about a different approach to examine Volume Shadow Copies (VSCs). I personally don’t like sneak peeks and...
  • Extracting ZeroAccess from NTFS Extended Attributes
    This past week I was reading a paper about the ZeroAccess Trojan when a section about a clever data hiding technique caught my eye. The pape...
  • Metasploit The Penetration Testers Guide Book Review
    A penetration test is a method to locate weaknesses in an organization’s network by simulating how an attacker may circumvent the security c...
  • Reviewing Timelines with Excel
    Generating timelines used to be a manual method for me until I started to use Harlan Carvey's timeline tools and Kristinn Gudjonsson...
  • You Are Not Admin with UAC
    There is a tendency to focus on what is different when we are faced with newer operating systems. What are the security changes and how does...
  • Linkz for Incident Response
    Due to changes with my employer last Spring my new responsibilities include all things involving incident response. I won’t go into details ...
  • Triaging Malware Incidents
    Triage is the assessment of a security event to determine if there is a security incident, its priority, and the need for escalation. As it ...
  • Linkz 4 Free Infosec and IT Training
    In this day and age budgets are shrinking, training funds are dwindling, and the threats we face continue to increase each day. It's not...
  • Tr3Secure Data Collection Script Reloaded
    There are a few movies I saw in my childhood that had an impact on me. One of those movies was Back to the Future. To this day I still have ...
  • Plugins: soft_run user_run
    The next two RegRipper plugins I wanted to highlight are: soft_run and user_run. Some may have been familiar with what these plugins did an...

Categories

  • adobe
  • advice
  • antivirus
  • attack vectors
  • autoplay
  • autorun
  • book review
  • categories
  • chain of evidence
  • conferences
  • delivery artifacts
  • detection
  • digital forensics search
  • drive-by
  • education
  • EEDI
  • email
  • encase
  • examination steps
  • exploits
  • feedreader
  • fraud
  • Google
  • hcp
  • investigation process
  • java
  • jumplists
  • kinect
  • links
  • malvertizing
  • malware
  • malware analysis
  • memory analysis
  • metadata
  • microsoft office
  • NTFS
  • perl
  • pfic
  • prefetch
  • program execution
  • readiness
  • registry
  • regripper
  • scams
  • script
  • search poisoning
  • sharing
  • shortcut files
  • spam
  • testing
  • timeline
  • timestomping
  • tools
  • tr3secure
  • training
  • triage
  • uac
  • volume shadow copies
  • xbox

Blog Archive

  • ▼  2013 (20)
    • ▼  November (1)
      • Linkz for Incident Response
    • ►  October (2)
    • ►  September (4)
    • ►  July (2)
    • ►  May (3)
    • ►  April (2)
    • ►  March (3)
    • ►  February (1)
    • ►  January (2)
  • ►  2012 (38)
    • ►  December (3)
    • ►  November (3)
    • ►  October (3)
    • ►  September (1)
    • ►  August (3)
    • ►  July (3)
    • ►  June (3)
    • ►  May (3)
    • ►  April (3)
    • ►  March (5)
    • ►  February (6)
    • ►  January (2)
  • ►  2011 (41)
    • ►  December (4)
    • ►  November (3)
    • ►  October (3)
    • ►  September (4)
    • ►  August (4)
    • ►  July (3)
    • ►  June (5)
    • ►  May (3)
    • ►  April (3)
    • ►  March (3)
    • ►  February (2)
    • ►  January (4)
  • ►  2010 (17)
    • ►  December (2)
    • ►  November (3)
    • ►  October (3)
    • ►  September (3)
    • ►  August (6)
Powered by Blogger.

About Me

Unknown
View my complete profile