The Art of Memory Forensics Book Review

Sunday, December 28, 2014 Posted by Corey Harrell
Christmas is in the rear view mirror and you may be left wondering about the gift you didn't find under the tree. The gift loaded with DFIR goodness to bring you into the new year. A gift you can use to improve your knowledge and skills. A gift to help you get to the next level in digital forensics, incident response, or malware analysis. You didn't get any DFIR goodness so it's a great opportunity to reward yourself with a gift of your own choice; one to accomplish what the previous sentences alluded to. If you fit this description then the book The Art of Memory Forensics is what you should be looking for. Even if the description doesn't fit and you don't already own this book then you should seriously check it out. This post is my review of the book The Art of Memory Forensics.

Three in One


The book addresses memory forensics on the following three operating systems: Windows, Linux, and Mac. This makes it an outstanding book since it addresses the most commonly faced operating systems. Furthermore, the content not only addresses memory forensic techniques but goes into detail about operating system internals. The majority of the systems I encounter are Windows systems so my focus was on the Windows portion of the book. (I did skim the other sections but I took my time in the Windows section.) The content went deep into the various Windows data structures and function calls. This makes the book an outstanding reference to better understand operating system internals. I easily envision myself using this book as a reference for years to come.

Not Just Words but Hands-on


One thing I tend to look for in a technical security book is how easy is it for the reader to take the content/techniques then apply it elsewhere. This is another area where the Art of Memory Forensics shines. The book's website provides additional materials that accompany the book. The items include lab questions, lab answers, and memory images for each chapter. This allows the reader an opportunity to do the hands-on labs to re-enforce that chapter's content. It's a great way to learn since you are actually performing memory forensics on an image after reading about it. Furthermore, to explain concepts the book uses - for the most part- memory images freely available on the Internet. As you read the book you can follow along by performing the same activities on the same memory images. At times the authors don't explicitly say what memory image they are using but the name of the memory images is pretty revealing. For example, in the Detecting Registry Persistence section (Kindle version page 4626) the Volatility handles plug-in is ran against a memory image named "laqma.mem". On the SampleMemoryImages webpage you can see a Laqma memory image is available and this is the one used in the book. This occurs frequently in the book as well as the authors specifically mentioning the memory image they are using.

The one area I thought that could make this book even better would be for the authors to explicitly state the memory image being used in the examples. This would make it easier for others (especially people who are not aware about the available memory images) to follow along in the book doing the same examples.

Memory Forensics in Toolbox


The last point I wanted to touch on about why I think so highly of this book is memory forensics is a process we need to have in our toolbox. This is true regardless if the work involves incident response or malware analysis. In incident response, there are times when you need to examine the volatile data on a system to obtain an answer. For example, a system is making network connections to a known malicious domain, which is setting off alerts. To tie the network connections to an actual process on the system requires memory forensics. The need for this in incident response is even more so with the recent increase of an exploit kit leveraging fileless malware. In malware analysis, there are times when memory forensics can provide additional information about a sample under examination. Does it open a socket, make network connections, inject code, hook functions, etc.. Memory forensics is now a process we need available in our toolbox and this book can help put it there.

All in All


If you are looking for a gift loaded with DFIR goodness, looking to improve your knowledge/skills, or looking for help to get to the next level in DFIR then this book is for you. The Art of Memory Forensics is a hefty book loaded with excellent content. It's an outstanding book and for those who don't already own it should seriously consider making it their next DFIR purchase. Just make sure to get your money's worth by grabbing the labs, memory images, and then putting hands to the keyboard as you read along.
  1. > Furthermore, the content not only addresses memory forensic techniques but goes into detail about operating system internals.

    You are better of buying the separate system internals books since this book contains a lot of errors in those sections. See: http://downloads.artofmemoryforensics.com/errata.txt

    Note that far more have been reported.

  2. My graduate level TCP/IP text book has 5 pages of corrections posted on its errata page. Its expected.

  3. @Joachim,

    I do use other resources for Windows internals. Specifically, the Windows Internals 6th edition books, Windows System Programming book, and the msdn.microsoft.com website. Also, the Rootkit Arsenal book did a nice job covering Windows functions as well. Finally, when in doubt I tend to use Process Monitor to help narrow down what may be occurring. I found all of these to be excellent resources and the best I found so far. The thing I liked about the Art of Memory Forensics book is it put it into DFIR context. By context I mean addressed the data structure, Windows function, and how that impacted what you are seeing in the tool's output.

    I was aware about the Errata Page and did read it but thanks for posting it for those who may not have seen it.

  4. > I do use other resources for Windows internals.
    Good, cross reference checking is a necessary evil (so to speak).

    > The thing I liked about the Art of Memory Forensics book is it put it into DFIR context.

    IMO the authors put it in a malware analysis context, very little in the context of actual digital forensics, but feel free to point me to a section that does.

    So IMO the book does very little to none worth to forensic in its title, there is very little discussion about the context of law and cross validation of claims. The authors even mess up the meaning of MACB indicators in the SleuthKit mactime output. This might sound harsh but IMO is one of the worst errors you can make in a book about forensics since the manual can be found here: http://wiki.sleuthkit.org/index.php?title=Mactime_output and clearly states what it should be.

    So the book is highly unfortunate titled if they had named it "The Art of memory analysis" or "The Volatility lost manual" that would have been more appropriate. And I would have less reason to be critical ;)

    > I was aware about the Errata Page and did read it but thanks for posting it for those who may not have seen it.

    Noprob, as indicated don't rely too much on the information being fully correct.

  5. Thanks for the support Corey!

  6. @Andrew when are you going to update the errata with the additional issues pointed out?

Post a Comment