In my older post, I claimed that replacing command line buffer is the shortest anti-forensics code as it takes 25 bytes only.
Sorry, I lied.
There are many ways to come up with a much shorter anti-forensics code. I will describe three ideas here. All of them are super obvious.
#1
Just use CopyFile function and copy Notepad.exe or any other file always present on the system over the file that is meant to be deleted (before its deletion). You may need some extra bytes for the file names, but well, it’s just one API call that it takes to implement a simple secure deletion.
#2
The second idea came (indirectly) from Harlan Carvey. He has been recently writing about his ongoing work to improve Prefetch file parsing script. While reading the MSDN article he referred to, I stumbled upon a description that I have read many times before, but never really thought of it from the anti-forensic angle.
In order to know what it should prefetch, the Windows XP Cache Manager monitors the page faults, both those that require that data be read from disk (hard faults) and those that simply require that data already in memory be added to a process’s working set (soft faults), that occur during the boot process and application startup. By default it traces through the first two minutes of the boot process, 60 seconds following the time when all Win32 services have finished initializing, or 30 seconds following the start of the user’s shell (typically Microsoft Internet Explorer), whichever of these three events occurs first. The Cache Manager also monitors the first 10 seconds of application startup.
Read the last sentence one more time:
The Cache Manager also monitors the first 10 seconds of application startup.
That is.
To reduce a number of forensically useful artifacts stored inside a Prefetch file, all the application has to do is… wait for 10 seconds. This can be easily achieved by running a function Sleep or NtDelayExecution at the entry point. Of course, there are extra bytes needed to import the function, etc., but since malware is often using one of these API anyway, a few bytes is what it takes.
I ran a quick test and it works as described:
The same code executed without any delay produced this Prefetch file:
General information : Filename : PREFTEST.EXE Run count : 1 Run time : Fri Mar 16 17:29:33 2012 (UTC) Prefetch file information: Reported size : 2468 Real size : 2468 Last access time : Fri Mar 16 17:29:48 2012 (UTC) Modification time : Fri Mar 16 17:29:43 2012 (UTC) Creation time : Fri Mar 16 17:29:48 2012 (UTC) Filepaths block (1240) : \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\NTDLL.DLL \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\KERNEL32.DLL \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\UNICODE.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\LOCALE.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\SORTTBLS.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\PREFETCH\PREFTEST.EXE \DEVICE\HARDDISKVOLUME1\WINDOWS\PREFETCH\FILE.TXT Volume information block (1976) : Volume Path : \DEVICE\HARDDISKVOLUME1 Serial Number : 1216058920 Creation Time : Sun Jan 6 13:31:42 2008 (UTC) \DEVICE\HARDDISKVOLUME1\ \DEVICE\HARDDISKVOLUME1\WINDOWS\ \DEVICE\HARDDISKVOLUME1\WINDOWS\PREFETCH\ \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\
and after adding at least 10 seconds delay at its entry point we get the following result:
General information : Filename : PREFTEST.EXE Run count : 1 Run time : Fri Mar 16 17:30:30 2012 (UTC) Prefetch file information: Reported size : 2304 Real size : 2304 Last access time : Fri Mar 16 17:30:46 2012 (UTC) Modification time : Fri Mar 16 17:30:40 2012 (UTC) Creation time : Fri Mar 16 17:30:46 2012 (UTC) Filepaths block (1184) : \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\NTDLL.DLL \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\KERNEL32.DLL \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\UNICODE.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\LOCALE.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\SORTTBLS.NLS \DEVICE\HARDDISKVOLUME1\WINDOWS\PREFETCH\PREFTEST.EXE Volume information block (1824) : Volume Path : \DEVICE\HARDDISKVOLUME1 Serial Number : 1216058920 Creation Time : Sun Jan 6 13:31:42 2008 (UTC) \DEVICE\HARDDISKVOLUME1\ \DEVICE\HARDDISKVOLUME1\WINDOWS\ \DEVICE\HARDDISKVOLUME1\WINDOWS\PREFETCH\ \DEVICE\HARDDISKVOLUME1\WINDOWS\SYSTEM32\
So, if you come across an empty Prefetch file in the future, don’t be surprised 🙂
I used this script by Jean François Gingras to parse Prefetch files.
And this is a source of a masm32 code I used in a test:
.586 .MODEL FLAT,STDCALL
include windows.inc include kernel32.inc includelib kernel32.lib include user32.inc includelib user32.lib .data file db 'file.txt',0 .data? numOfBytesRead dd ? buffer db 256 dup(?) .code Start: ; invoke Sleep,15*1000 invoke CreateFileA, OFFSET file, \ GENERIC_READ, \ FILE_SHARE_READ or FILE_SHARE_WRITE, \ NULL, \ OPEN_EXISTING, \ FILE_ATTRIBUTE_NORMAL, \ NULL invoke ReadFile, eax, \ OFFSET buffer, \ SIZEOF buffer, \ OFFSET numOfBytesRead, \ NULL ; invoke Sleep,15*1000 invoke ExitProcess,-1 END Start
All you have to do is uncomment one of the lines ‘; invoke Sleep,15*1000‘ – first one, if you want to keep the Prefetch file clean, and second one if you want to see ‘file.txt’ inside the Prefetch file associated with the executable. Note, for the .pf file to be created program needs to run for at least 10 seconds and actually read the file.
This sort of tricks may sound like a good idea, but as many forensic investigators already know – and as pointed out by Harlan who commented on one of my previous posts here – that (quoting) “the absence of an artifact where you expect to find one is itself an artifact”. Indeed. Not having any file listed inside the Prefetch file doesn’t change the fact that its sole presence indicates the executable _has_ been ran and some useful forensic data can still be fetched from the *.pf files even if they are kind of empty.
This leads me to the third idea that is the most obvious.
#3
Malware can either delete the prefetch file, or use some way to hide it e.g. using PrefetchADS technique I described in one of my forensic riddles. This also can be achieved with one API call.
So, there you have it. Anti-forensics, counter-forensics, or forensics prevention can be implemented in a simple and very effective way. And even if they can’t fool an experienced investigator, they surely can take lots of fun from investigation, increase time spent on chasing after ghosts and remove all the juice from the final report. After all, it’s the ‘smoking gun’ that we all want to show, not a circumstantial evidence that leaves many questions unanswered.