Monday, October 26, 2015

RAI UNO: TG1 Cyber Security Speciale

Giving my contribute to the Italian public mainstream channel RAI UNO. Talking about Cyber Security, Malware and Targeted Attacks with Barbara Carfagna.

Click on the Preview (above) to watch the video from the official website.

Monday, October 19, 2015

SandBoxes personal evaluations

Understanding the "sandbox" technology is a fundamental step in Malware prevention. While it is obvious the new evasion techniques such as (but not limited to); Malware Encryption, Malware Packing, Metamorphism and Polimorfism are able to evade romantic defensive technologies such as (but not limited to) AntiVirus, Intrusion Detection and Prevention Systems, URL Filtering and Proxy, is it not obvious enough that Malware can evade sandbox too.

This post is not about evasion techniques, I've been talking and writing a lot on evasion techniques, but is about better knowing sandbox technologies.

Each SandBox implements one or more specific detection strategies which makes SandBox an unique environment. It's hard to find two SandBoxes implementing the same strategies.  My point here is to understand the technologies and later on figuring out the different strategies behind them.

If you are wondering why I decided to start from "implementation" (describing how specific sandboxes work) rather then from "foundations" (describing what are the most common ways to detect Malwares ) the answer is pretty simple: "I believe is much more interesting a bottom-up approach: starting from real technologies to reach out more general strategies". I know it's questionable.

Let me start from Anubis. ANalysis of Unknown BInarieS is one of the "oldest" Sandboxes becoming one of the most known online analysis system. Anubis decided to implements its own running device on an emulated environment consisting of a Windows XP operating system running as the guest in Qemu. The analysis is performed by monitoring the invocation of Windows API functions, as well as system service calls to the Windows Native API. Additionally, the parameters passed to these functions are examined and tracked.

CWSandbox. Is another quite famous Sandbox system. It executes the sample in a "patched" way in order to discover what it does. It executes the sample into memory by adding a function call built to monitoring the API and SysCall before and after the execution. It uses an "instrumentation" procedure to patch the bynaries.

Norman SandBox. The Norman sandbox is a dynamic malware analysis solution which executes the sample in a small controlled virtual environment which simulate Windows operating system. The simulation involve multiple OS levels (such as API, System Calls, libraries, etc) as well as the Local area Network and the external Internet connectivity. Norman supports memory protection emulation and Multi-Threading supports in order to better emulate Windows OS. It has been used to mostly detect Net Worms since it makes eavy usage of DNS, Connection resolultions etc.  It monitors the auto-start extensibility points (ASEPs) often used by Malware to achieve persistance.

Joebox. One of the first SandBox system built to live in real hardware and not on Virtual/Emulated environment. It is based on a client/server architecture in which the client runs the Suspicious sample by hooking (user mode) API, Syscall invocations, Export Address Table, and System Service Descriptor Table). It enables a Kernel driver to cloaking binary patching in order to "inject calls-debugging code". It uses AutoIT to emulate users interaction with the machine during the analysis phase.

LastLine. Developed by Anubis's creators, Lastline implements multiple techniques including Intrusion Detection System, DNS analysis and Sandbox. It does suspicious file analysis by performing a multi path technique on a fully emulated environment. While Norman implemets an OS emulation Lastline performs a CPU and Memory emulation being hidden from the entire OS (not care if Kernel mode or  user mode at this stage, since being behind the whole OS). 

ReVirt. Mainly based on virtualization environment (VBox) where a local engine logs alld the grabbed data. A centralized collector processes and analyze the produced logs. It includes a system called BackTracker that helps system administrators understand (and thereby recover from) an intrusion, by automatically identifying potential sequences of steps that occurred in an intrusion.

FireEye. It is one of the most known player of Cyber Security Solitions, it provides many products in addition to a sandbox, but we want to focus on FireEye SandBox. It is based on proprietary virtualization. They do not run samples in known VM-Managers such as: ESX, Hypervisor and VM-Ware, but they built their own VM engine in order to "avoid" the "VM-Aware" Malware.

Buster. Buster is a toolkit applied to Sanboxie running on a VirtualMachine as well as on a real hardware. Buster Sandbox Analyzer is a tool that has been designed to analyze the behaviour of processes and the changes made to system and then evaluate if they are malware suspicious. The changes made to system can be of several types: file system changes, registry changes and port changes. It does not include any anti evasion technique to mitigate the "VM-Aware" Malware.

WildFire. Powered by PaloAlto, wildfire claims to be a fully os level emulation system able to detect  and report Malware within 15 minutes from the discovery. It basically emulates the Operation System cheating the Malware and observing syscalls, API calls, File Mutex, RegKeys, and so on .

HookAnalyzer. Well, actually it's hard to define this tool a SandBox, but let me put it in this environment. It basically works by hooking the executable getting dynimic as well as stati conformation on it. It runs on physical hardware and do not monitor any level of virtualization/emulation.

Cuckoo SandBox. A simple sandbox based on Virtualization within many embedded euristics to detect "VM-aware" malware. It is implemented according to Client-Server Infrastructural paradigm. On client (VirtualMachine running potentially almost any OS ) a daemon runs on system-level. It (named cuckoomon) hooks API and Syscall and sends back to the VM manager (cuckoo result-server) the entire result set. A post processing engine elaborates the cuckoomon results inflating a DB. Eventually a webinterface shows results and hits euristics.

Out there many other SandBoxe solutions are available for the security community, for example: SandBlust (By Checkpoint), Jotti, PayloadSecurity and ThreatGrid (Cisco) are some of the most known ones. Each SandBox owns particularity in botht technical implementations and detection techniques. Even on my hart I definitely know what is the actual best solution (for the time being) considering the Malware Zoo I am observing, I wont to give a gudjement since I  known that technologies follows markets, so there is not an absolute "best in class". For such a reason consider my contribution on providing the following considerations based on logics and real experiences expressed by the following chart.

SandBox Strategies VS Comparison Properties (click to enlarge)

On the time line: Virtualization, Virtualization + Anti-detection Heuristics, Emulation OS, Hardware Emulation, Hardware Agent Base and Hardware Based Agentless are the most used strategies in nowadays SandBoxes. On the leftside hand evaluation parameters: 
  • Evasion Grade: how easy it could be, for a Malware, to detect the SandBox system. Higher this value easier is the evasion.
  • Analysis Depth: how deep is the SandBox Analysis due to the owned data. Much more "raw" data is involved much more deep analysis is possible, much high is the complexity.
  • Solution Complexity: the complexity of the SandBox solution. Higher is the complexity higher is the bug rate (not feedback chain in the evasion technique has been considered so far).
  • Realization: considering my current experiences on Malware detection and evasion technologies, the state of the art I see on the implementation phase grouped by detection strategy.
As you might observe there is really not a best solution right now. SandBoxes strategies with low detection rate are the one with the highest complexity and the lower realization rate. SandBoxes stategies with the highest detection rate could be adopt heuristics to try to decrease it (once the sample has been positive analyzed, ergo time consuming) but own the lowest complexity rate and highest realization rate.

Depending on the involved "business" you might need to decide the right sandbox in the right time frame. This activity includes (but not limited to) organization analysis, threat detection, asset risks, past attacks and underground economy knowledge. I hope this post will help you out to decide the right sandbox for your business.

Monday, September 21, 2015 New "Speed" and New Samples Available now.

Hello everybody, today is about speed improvements and new malware samples in If you followed the genesys you might remeber the early stage development where took between 8 to 10 minutes to visualize statistics over 43k Malware Analysis. Today it runs much better alost 15 seconds to visualize 76.2K Malware Analysis (ok, I know.. it really depends on Network speed and Computation power... but tested on the same machine you might experience a hug performance gap).

Let me just remind you what is about:
"The continued growth in number and in complexity of malware is a well established fact. Malwares are no longer simple pieces of code that rely on unsuspecting users to spread and thrive. They can change, adapt and hide themselves from analysts, using very sophisticated techniques. Static analysis is complex and time consuming, and it could be difficult to deduce every possible malicious behaviour, yet it is often very effective because it hinders the capability of malware to detect the analysis environment.  The purpose of is to provide valuable assistance to the phase of static analysis, supporting analysts in their exploration of code features, by letting them make more focused, statistically motivated and structured decisions."
We are facing a "Big Data" problem. Thousands of samples produce Hundred Thousands of results, which end up to be Giga Bytes of well structured Text. And.. yes, I want to make general tatistics so far (general !== from "time frame defined") so I am not interested on filtering data (well..I know I will end up putting a time filter on the main page.. but not today!). My main goal is to answer in the quickest way to such a questions: " What are the most used packers ?" or "What are the most used evasion techniques?" or again "What are the most used API or Anti-Debbugging Techniques?" and so on and so forth. Obviusly I want to give such statistics by using a simple and intuitive web interface. You might wonder why those questions are so important for me !? Well, because they really drive my decisions during a romantic Malware analysis.

The following image shows the today stats on detail

In order to provide a fast and reilable web visualization user interface I've tried several algorithms and several frameworks but my best choice (so far)  has to approached the problem using the Javascript "Web Workers" (HTML5). total samples.

From W3C School :
A web worker is a JavaScript that runs in the background, independently of other scripts, without affecting the performance of the page. You can continue to do whatever you want: clicking, selecting things, etc., while the web worker runs in the background.

 The new and simple algorithm (which is not the best I can create and it is not remarkable in any point but it made a huge improvement) which made possible the huge visualization improvement from the last two versions is available here.  The following image shows the principal code function responsible to build the output, before passing it to google graphs.

Simple Visualization Algorithm
 As you might agree with me the entire code should be protected (which is not protected on undefinition, null pointers, etc..) and even improved in speed introducing multiple web workers. If you like to be involved in that project just drop me an email, any suggestion is welcomed as well. Enjoy the new results !

Thursday, September 3, 2015

Shifu: A new interesting Banking Trojan

Hello everybody, today I'd like to share some infos on "Shifu" a new incredibly interesting banking trojan. At this point you might think:
"Why are you writing about Shifu among many other new threats (even more discussed)  out there ? "
Well... Shifu is a new banking trojan which actually attacks Japanese banks mostly,  it's actually well geo-localized and probably it will end up on a specific amount of organizations, but what fascinates me is the way it implements many features by copying what have done so far some of the "best in class" known Malware. Shifu implements the following features:
  • Domain Generation Algorithm (DGA): Shifu uses the Shiz Trojan’s DGA. The exposed algorithm itself is easy to find online, and the developers behind Shifu have elected to use it for the generation of random domain names for covert botnet communications. 
  • Theft From Bank Apps: Theft of passwords, authentication token files, user certificate keys and sensitive data from Java applets is one of Shifu’s principal mechanisms. This type of modus operandi is familiar from Corcow’s and Shiz’s codes. Both Trojans used these mechanisms to target the banking applications of Russia- and Ukraine-based banks. Shifu, too, targets Russian banks as part of its target list in addition to Japanese banks.
  •  Anti-Sec: Shifu’s string obfuscation and anti-research techniques were taken from Zeus VM (in its Chtonik/Maple variation), including anti-VM and the disabling of security tools and sandboxes. 
  • Stealth: Part of Shifu’s stealth techniques are unique to the Gozi/ISFB Trojan, and Shifu uses Gozi’s exact same command execution scheme to hide itself in the Windows file system.
  • Config: The Shifu Trojan is operated with a configuration file written in XML format — not a common format for Trojans, and similar to the Dridex Trojan’s configuration (Dridex is a Bugat offspring). 
  • Wipe System Restore: Shifu wipes the local System Restore point on infected machines in a similar way to the Conficker worm, which was popular in 2009. 
  • Commuication protocol: Shifu implements an SSL communication layer based on a Self-signed certificate. The implemented module reminds analysts to the one used on Dyre Trojan campains in Late 2015.
Another interesting feature is about Point Of Sales. To make matters worse, Shifu searches for specific POS memory strings (and processes). If it finds a POS trace it starts a "stealing credit card numbers" procedure.

Last but not least Shifu makes sure none else will own the attacked system. Once it gets installed on the victim machine is starts an "AV" procedure (forgive me, is not actually an AV procedure, but it makes the idea) which locates "suspicious" files and  denies their installation. According to IBM Security Intelligence's report (here) the Malware is likely developed by a Russian group.

Let's get dirty hands on it performing basics Reverse Engineering actions to see what are the real countermeasures it adopts.  From the IBM Report (linked abouve) you may find the Malware signature (NmE5ZDRhMzIzOTg3NDg5YzhlOGI1NTc2ZjY3YjJjOTQ) which can be used into common online SandBox systems to look for samples. As you might observe the sample I've got implemets some anti-debugging techniques as well as some basic SandBox evasion techniques (for more information please have a look to malwarestats):

GetLastError, IsDebuggerPresent, GetVolumeInformations, etc..
 An interesting sequences of API calls were found: GetProcessAddress  (Retrieve the address of of an exported function or variable from the specified dynamic-link library) -- VirtualProtect (stack) (Changes the protection on a region of committed pages in the virtual address space of the calling process.) -- VirtualAlloc (Reserves, commits, or changes the state of a region of pages in the virtual address space of the calling process. Memory allocated by this function is automatically initialized to zero.) -- Sleep (Suspends the execution of the current thread until the time-out interval elapses.) -- VirtualAlloc -- 

Another interesting pattern found during the simple static analysis performed phase (showed on the following image) is the dynamically loaded Library pattern (previous downloaded).  As you may observe on row 2861 the system points out to a specific location and call LoadLibraryA to load it into memory.

Dynamically Loaded DLL
Dynamic Analysis clearly shows Sample's RAT features by spawning a shell (on my machine PID: 1388 within Parent PID: 788 owning to the executed Sample ) and executing commands. Unfortunately the evasion techniques detected the SandBox execution. The following image shows the check of Python presence, which often is one of the detection mechanisms (How many common users have Python on their Windows Machines ? Not much, really).

Python Detection

After a simple de-obfuscation round (Visual C Packer was detected) the analyst could appreciate the command line parser. Probably the one used to communicate through Command and Control (not much further analysis has been performed)

Command Line Parser
Network wise the sample embeds the following addresses:
  • ( Noisy maker
  • ( Much more interesting because geolocalized in China and the domain has changed at least two servers during the last year.
A simple nmap scan on it shows up-and-running a nginx server on both ports 80 and 443, used to comunicate to Malware and a ssh daemon active on standard port and and an interesting port 53 TCP opened. Statically analized behaviour presents the following TimeLine (click on it to enlarge):

Behaviour Time Line
Not really a significant one but the cmd.exe spawned feels like an hero. Concluding my post I wanted to impress on my pages this significant piece of Malware which embeds many different techniques borrowed from many older Malware underlining a new Malware writers skill sets, able to make harder and harder piece of code as their wish (just by adding feature from different Malwares).

Tuesday, August 11, 2015

Exploit Kits on August 2015

Often people, including students and security professionals asks me about Exploit kits (EK). EKs play a foundamental role in todays malware propagation because developed to deliver content through vulnerabilities. Aims of the EK is to exploit a target client machine through well known or sometimes "less known" vulnerabilities which usually target browsers, Java Runtime Environment, Adobe products and commonly used applications including (but not limited to): Media Players, Visualisation utilities, Microsoft Office documents and so on. A key characteristic of an exploit kit is the ease with which it can be used even by attackers who are not IT or security experts. The attacker doesn’t need to know how to create exploits to benefit from infecting systems. Further, an exploit pack typically provides a user-friendly web interface that helps the attacker track the infection campaign. Some exploit kits offer capabilities for remotely controlling the exploited system, allowing the attacker to create an Internet crimeware platform for further malicious activities.

The following table (from contagiodump ) keeps trace of most of the known exploit kits out there within relatives exploited vulnerabilities.

Click to Enlarge, credits to Contagio Data

As you might appreciate from the Sally's work many vulnerabilities are covered by most of the exploit kits but not all, so depending on the administration console (which almost every EK gives to attackers) and, most important, on the target system, the attacker could choose between several EKs. While several exploits kits are available nowadays only a subset of them are mostly used. As described in this post from from MalwareBytes the most used EKs are represented in the following picture.

Exploit Kits from MalwareBytes analysis.

Now you would probably know how the EK infection process works, well a nice work made by TrendMicro explains in a simple view the 4 stage infection chain.

4 stage EKs infection chain by TrendMicro

Contact is the beginning of infection, where an attacker attempts to make people access the link of an exploit kit server. Contact is often done through spammed email, wherein recipients are tricked into clicking a link through social engineering lures. 

Traffic redirection system refers to the capacity with which the exploit kit operator can screen through victims based on certain condition sets. This is done through a traffic direct system, such as SutraTDS or KeitaroTDS, for aggregating and filtering redirect traffic before accessing the exploit kit server.

Once users are successfully tricked into clicking the link of an exploit kit server in the contact stage and filtered in the redirect stage, they will be directed to the exploit kit’s landing page. The landing page is responsible for profiling client environment and in determining which vulnerabilities should be used in the ensuing attack.

According to TrendMicro research (except for SweetOrange)  I do observe the following EK in almost the same score position in my current Cyber Attack detections

Most used Exploit Kits
As Malware does, ExploitKits are in continuous development conditions and day by day we observe different variants and improved evasion techniques as well as exploits integrations. Be aware that  those kits made really simple (well, I didn't say easy) Malware propagation so watch out your apps !

Monday, June 22, 2015

Static Analysis Malware Statistics

During the past month I've been dedicated some of my free time in building a Malware static analysis pipeline. Goal of this work is to give to Malware analists usefull statistics on what evasion techniques current Malware are implementing. If you are interested on Malware evasion techniques please have a look to my previous post on that topic ( here ). As my readers know one of my favorite Cyber Security topic is Malware and thier creation, if you are new about it, I suggest you to take a look to the following "blog posts": 

The following image shows the as appears nowaday. Besides the "romantic algebraic sums" (of the analyzed samples),  the number of xor encrypted detections, the Malicious DLL found over the total amount of detections and the average file size, more graphs showing out  more "evasion techniques" are represented.

One of the most interesting information I wanted to give was about the used evasive techniques to detect the virtualized environment the sample might be in. These information have been collected and represented in the "Used  Evasion Technique" graph. 

As a today (please refer to the "blog post" date) the most common Virtual Environment evasion technique is the VMCheck.dll (Red Pill) followed by QEMU CPUID Trick and VirtualBox Detection.


The second most important information given is about Packers. Whate ater the most used packer Malware implements to evade signature detection? The following pie chart shows represents the most used packers among others.


Active analysts (and IDA Pros) will agree to me when I say that one of the most time consuming avtivity is to debug a given sample. Figuring out what is the most used Anti-Debugging technique, could be time saving especially when the analyst is at the beginning of his analysis. The following graph shows my statistics on 21k malware (confirmed malware and not just sample).

 More stats will be available on the web site:, please have a look ! 

How To Contribute:
Day by day I'll add more and more samples but actually the pushing pipeline is not available online and is not available for free submiting. If you wish to contribute (and please do!) you should share with me your malware (GoogleDrive, DropBox, MegaTransfer, etc... might help the sharing process) I'll add them to my simple importing pipeline and I'll put your name on contributor page.

 The data is hosted for free on who accepted to get me a free license for that project.

Thank you !

Sunday, May 10, 2015

Volatility on Darkcomet

Let's assume you've got a friend who asked you to have a look to his computer because he feels like something wrong is happening. What would you do? 

Option 1: "I have no idea about how to investigate on 'computer stuff', please contact your reseller "
Option 2: "Ok, Let me access to your computer, I will see what I can do"

I's raining a lot and my friend was pretty serious about it so I decided to choose the option number 2... :O

I've been starting by downloading DumpIT by MoonSols which is a "single click" Windows memory dumping tool. After a few command line answers I've got a fully dumped memory in one file. I downloaded it on my MAC and started the volatility analysis hunting the "something wrong".  By running imageinfo, volatility analyses the memory layout getting back the memory profile used by identify the analyzed machine.

Volatility imageinfo
Understanding what are the processes running on the analyzed machine is a foundamental step to grab the eventually "unwanted software". The following image shows the volatility psxview. Few processes are suspicious to me but the most weird is the one named runddl32.exe. It 's suspicious (at least to me) because the name mispelling and because it tries to evade "deskthrd" detection (not common at all). Psxview is a nice volatility plugin which compares the following different proces' searches in order figure out hiding techniques. The implemented process searche techniques follows:
  •  PsActiveProcessHead linked list 
  •  EPROCESS pool scanning
  •  ETHREAD pool scanning (then it references the owning EPROCESS) 
  •  PspCidTable 
  •  Csrss.exe handle table 
  •  Csrss.exe internal linked list

Volatility Psxview
Let's have a deep look into runddl32.exe by running a dlllist on such a pid. Dlllist returns the memory location and the location path of each used DLL. This information is useful to recognazie malicious patterns in file locations. Malicious files are used to be located into TEMP directories due to dir rights. QED (See the following image) !

Volatility dlllist
Volatility dumpfiles helps researchers to dump pieces of memory and saving them into files. The following image shows how I used dumpfiles to obtain the physical supicious files. Having them means to be able to perform static analysis (It wont run... no dyno) on the samples figuring out what they do and if they might be the cause of the "weird behavior".

Volatility dumpfiles

Just few steps into static analysis to discover the sample is actually doing something very bad such as: keylogging, selfupdate, drop and download,  shllcoding etc etc...

AntiDebug functions
Looking into the sample's memory page -- for sure -- something strange is happening ! Page EXECUTE_READWRITE is found. VAD Tree (ref: here) is used to check for injections with a super positive result! We can know assert tha the PC was infected.

VAD Tree search on volatility
Let me try to search the file on Virustotal to se if I get more on it.... Here it goes, VirusTotal identifies the sample as Darkcomet... a simple opensource Remote Aadministration Tool (RAT).

VirusTotal DarkComet

Weird things were happening to my friend's PC and he was right. Actually Darkcomet is only one of the suspicuous file indentified on the psxview, for example I saw a notepad.exe child of explorer.exe and an cmd.exe child of explorer.exe as well. It was a nice hunting saturday night !