Syndicate

Syndicate content

Flattr


Flattr this

If you like this, you can use flattr. ;)

Imprint

About
eMail: wishinet at gmail . com
PGP ID: 0xCCCA5E74

Jabber: wishi@jabber.ccc.de

Memory forensics explained

txttxt

Not just the disks!

CFAE6575-BF32-4BD4-8F4A-D1C02C06E385.jpg
it turns out rock climbing sometimes is easier than diving.

Generations of forensic experts just used data from the hard-disk. They dived down deeply into the filesystems to dig for all kinds of incident. But what's with the surface?

It turns out if you've got a chance to get hands on RAM nowadays, you should take it. - Even in pentesting: here's why and how.

Why?

Yesterday I started to listen to Rob Lee's quality webcast about Windows memory analysis and forensics - a available summit from SANS's forensics and incident response.

I'm not directly in the forensics field, but in incident response and penetration testing. It makes a lot of sense to pull the passwords out of memory instead of bruteforceing them; or cracking the hashes via rainbow-tables. Memory forensics allows faster and more elegant solutions: for tracking all kinds of changes, for getting the authentication (password or hash), or to reconstruct the initial situation when the infection happened.

As long as nobody pulled the plug... you're in.

How?



I very much recommend the webcast here... I'll just list headwords and links to follow.

You can find most of the tools in the SIFT VMware applicance, which is a Fedora Linux. It has everything needed in the following directories:

/forensics
Location of the files used for the Autopsy Toolset

/usr/local/src
Source files for Autopsy, The Sleuth Kit, and other tools

/usr/local/bin
Location of the forensic pre-compiled binaries

/images
Location of the images that were seized from your compromised 
system

/mnt/hack
Location of the mount points for the file system images

Furthermore there's a cheat-sheet. The hot stuff is to use mdd or something else on the particular Windows box, and to dump the memory. Than you're able to use the volatility framework - and some of the modules to get the bait.

/usr/local/src/volatility-SIFT Workstation 1.3

The usage instructions are copied from Rob's slides:

volatility command –f /path/to/windows_xp_memory.img 
[Supported commands] 
connscan 		Scan for connection objects 
files        	  	Print list of open files for each process 
hibinfo 		Convert hibernation file to linear raw image 
procdump 		Dump a process to an executable sample 
pslist 			Print list of running processes 
regobjkeys 		Print list of open regkeys 
 
for each process 
sockets       		Print list of open sockets 
sockscan 		Scan for socket objects 
Command Help 
# cd /usr/local/src/volatility 
# python  volatility  command -–help

I guess this is self-explaining. Here's something I just learned yesterday: if you get your hands on a compromised box there's a plugin command "hivedump" in the volatility framework, that dumps the registry to a CSV. That's awesome!

Of course SIFT contains lots of other tools - not EnCase or FTK Imager because these aren't free and just Windows-based.

Perspective



I see a good future point for this in IR. In case of forensics I must admit that maybe there'll be too much own overhead in the acquisition files? - But... that depends. In any case you collected the incident-data.

Links to the Windows tools




Have fun,
wishi

Post new comment

The content of this field is kept private and will not be shown publicly.
CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.

Save the nature. Don't print this!


I provide textual exports for every blog entry. However let's save the nature together. The nature is everything around us. Every being should be respected. Save the nature - don't print too much.


Die Umgehung dieser Ausdrucksperre ist nach § 95a UrhG verboten!
Inhaltlich Verantwortlicher gemäß § 10 Absatz 3 MDStV: Marius Ciepluch - Anschrift via eMail. Die eMail Adresse entnehmen sie dem Impresseum dieser englischsprachigen Seite.
Aus Datenschutzgründen habe ich weder offiziellen noch behördlichen Schriftverkehr via eMail. Dazu ist die postalische, beim Dienstleister hinterlegte, Anschrift zu verwenden.

Datenerfassung

Es werden keine personenbezogenen Daten erfasst. Logdaten werden anonymisiert.