Friday, September 12, 2014

El Jefe - Alerts

In the never ending cat and mouse game between offense and defense, the defenders generally represent the mice. The objective of vulnerability management is to flip the script on the proverbial felines and set functional mouse traps for them. We can do this using machine learning to predict a potential attack vector or, if you feel like being an oxymoron, by deploying a fast and effective clean up.

Polemic methodologies such as "0day feeds" are ineffective as they protect against samples, not classes. While useful to demonstrate actionable risk to management, catching a single instance of a vulnerability implementation (AKA exploit) , 0day or otherwise, does very little to effectively protect you at an enterprise scale against similar vulnerabilities. Shooting fish in a barrel generally does not equate to a sane defensive posture.

El Jefe Alerts is our approach to a more comprehensive methodology of vulnerability management.

With El Jefe Alerts you are able to use, share and extend scenario driven vulnerability heuristics for entire vulnerability classes that trigger well defined response events.

A simple example:  Suppose every time iexplorer.exe parents an executable anywhere in your enterprise I want to retrieve the executable, run it through a sandbox and then receive an event summary via email. We can do that with El Jefe Alerts.

Or suppose that every time iexplorer.exe's virtual size gets bigger than 2 GB we would like to receive an urgent SMS, as this is a likely indicator of a heap spray in progress. We can do that with El Jefe Alerts.

What if I want to be notified whenever someone in the Enterprise launches a certain executable outside of work hours? Not a problem for El Jefe Alerts.

Let's walk through a simple example of how to set up an El Jefe Alert.

If you've attended DEFCON in the last three years, you probably noticed a distinct pattern in any presentation that involves Windows exploitation. Namely that a lot of attacker demos end up executing calc.exe! It seems they consider math a very important step of their post-exploitation strategy. Lets ruin their 4th grade algebra aspirations with the following:
1) The first step is to create a Filter, which is the module responsible for the heuristic that will trigger the El Jefe response action. Filters are django models which you can find in the alerts/ file.

ExecutionFilter will be triggered when a binary filepath is on our blacklist

You create the model and you implement the heuristic in the "filter" function. In this case, we want to trigger an action if and when an executed binary is listed in our blacklist.

2) The second step is to create an El Jefe Action script. The Action script defines the El Jefe response action to a filter trigger event. You can define multiple actions per filter. In our example script, we create a simple file that just prints some basic information about the event, and we place it in the xmlserver/actions/ folder.
A simple action that will print the username, ip and binary executed

3) The final step is to put everything together. Once we've implemented our Filters and Alerts, they will automatically show up in the El Jefe interface.

You can now go to the "Alerts" section in the navigation bar, and select "Add". You select your Filter and associate one or more Actions and then select the executables you want to keep track of.
Adding a working filter
4) Voila! You are now Busticati proof!
Gotcha! You should never do your math homework on a compromised machine!

In real life, this simple Alert is useful when  you have recognized a particular pattern with an attacker - for example, many of them use WMIC.exe because that is how metasploit's meterpreter and some other trojans do their WMI activity (for enumerating all sorts of important system information). Normal users almost never use WMIC.exe so this is an interesting heuristic to apply across your network. For the record, INNUENDO does not shell out to WMIC - it has WMI support built into it.

Tuesday, June 24, 2014

El Jefe v2.1 Release

It has been a couple of weeks since our last communication but the team is working hard to make El Jefe your first and best choice for threat analysis.

Today's release, while technically a minor release, has many new features.

Cuckoo configuration is not only fully integrated into our WebUI, but we also modified the implementation to allow you to use remote Virtual Machines! El Jefe now lets you configure where and how you want to run the sample inside the sandbox, allowing you to properly use it to look at more complex malware problems.
Configure a remote Sandbox Virtual Machine

For example, assume that you are ahead of the curve and you are using El Jefe to monitor your C-level executives. Each of these will probably have their own software set-up, operating system, etc. Once you find a potential threat (manually, or using the built in El Jefe heuristics), you will want to understand the penetration behavior of the sample in the exact same environment it was run in on the executive's computer. That's why you set up a different Virtual Machine to mimic each executive's personal environment and you can have the whole process run through El Jefe.

Select the right Sandbox Virtual Machine to run your sample

We want to be as open as possible and give you the freedom to interface El Jefe in new and powerful ways that we have not even thought of. That's why we built a plugin to allow you to use COSEINC's  CAMAL instead of Cuckoo to get even more accurate results from your sandbox. (Kudos to Thomas and the COSEINC team for their fast response and an excellent product!)
Integration with CAMAL

El Jefe clients are now fully configured via the El Jefe UI, which will certainly make your life easier when deploying across your user-base.
Set-up your client completely from the WebUI

El Jefe clients now support proxies - which is obviously important for those of you in large enterprises.

People seem to love the Event filter, so we've started adding more features to it. You can now trigger an email warning and send suspicious binaries directly to sandbox analysis. This is just an example, of course. Our new API is easy enough to learn that you can create your own event filter in less than 20 minutes - which is important if you are in the middle of an Incident. Fast reaction times when facing custom malware and an advanced penetration team are often the only hope of your IR team for containing an ongoing compromise.

In case you didn't notice, El Jefe is released by Immunity under the GPL v3 license. But, just in case, we now let you know in every file. Source code to El Jefe (for both the client and the server) is provided on the website and will soon be moved to a GitHub repository.

We also put together an installer for RHEL (widely used in enterprises) along with the Ubuntu one to make everything install automatically and we have a beautifully crafted PDF with a step by step guide on how to install El Jefe.

Whether you're currently responding to an incident or you think you may someday want to respond to an incident, we think installing and getting to know El Jefe is a great first step for you. It's Free, and it works, and it's only getting better.

Of course, we welcome any of your feedback - send it to and we'll be happy to respond!

Download it here
sha1sum: b8ee361ecf67e76ec0888e570153f76b15dfcea5  eljefe2.1.release.tar.gz

Thursday, May 8, 2014

Connecting El Jefe 2.0 with the Cuckoo malware sandbox

One of the great new features in ElJefe May release is the integration of the Cuckoo malware analysis system as part of our interface. Cuckoo runs the malware executable in a sandboxed environment inside a virtual machine and produces amazing data, which we display in the El Jefe interface for you. However, configuring it is not a trivial task. That's why we put together a little blogpost to make our users happier. Always thinking about our El Jefe users, is our motto!

We are going to setup the host machine and the guest machine. The last one is where we are going to run the malicious files.
The following commands will install the necessary files:
  $ sudo apt-get install python python-sqlalchemy python-bson python-pip libcap2-bin
  $ sudo pip install sqlalchemy bson Django
You also need to install mongodb. You can download the necessary files from
When Cuckoo is analyzing a submitted files there are some modules and libraries we can install to get better and more complete analysis and reports. These modules are optional but their installation is highly recommended.
We are using setcap to give privileges to tcpdump to run as root without having to run Cuckoo as root.
  $ sudo setcap cap_net_raw,cap_net_admin=eip /usr/sbin/tcpdump
We can check the results with:

  $ getcap /usr/sbin/tcpdump
  /usr/sbin/tcpdump = cap_net_admin,cap_net_raw+eip

You will need to create a new user, you can do that with the following command:
  $ sudo adduser cuckoo
You can download Cuckoo from You will need to copy the Cuckoo folder into the El Jefe root folder. You should see a directory listing like this:
client cuckoo dependencies installer webapp

You will also need to uncomment some lines of code to make cuckoo work:
webapp/ lines 105 to 124
webapp/xmlserver/ lines from 55 to 58.
webapp/home/ lines 44, 45, 1055, 1316 to 1334.
webapp/analysis/ lines 20 to 25.
webapp/templates/base_.html remove the {% comment %} and {% endcomment %} tags on lines 121 and 142.

Now it's time to setup the virtual machine where we are going to run and analyze our binaries. Cuckoo supports VirtualBox, KVM and VMWare for virtualization. We choose VMWare as our virtualization software, so the following steps will show you how to configure Cuckoo to work with VMWare. If you wish to use other virtualization software, follow the guidelines in the following URL:

First, we'll need to create a new virtual machine. The preferred OS is Windows XP, but you can use the OS of your preference. Obviously for future versions of El Jefe we will support automatically choosing the right target VM based on the El Jefe client's OS.

We need to install Python on the virtual machine and PIL ( if we want Cuckoo to be able to take screenshots of the binary as it runs.

Now it's time to configure the virtual machine's networking. We are going to disable the Windows Firewall first, and then create a virtual host only network for the guest and the host.

We are going to use the network, the guest configuration will be:
IP Address:
Cuckoo supports DNS resolution, so you can use as your DNS server. In our experience, we get better analysis results by using a public DNS server.

And the host configuration (for the vmnet adapter) will be:
Ip Address:
You can choose whatever network and addreses you like.

Now we need to configure packet forwarding on the host machine to give the guest machine Internet access. We can do this with the following commands ( replacing eth0 with your outgoing interface and vboxnet0 with your virtual interface).
iptables -A FORWARD -o eth0 -i vboxnet0 -s -m conntrack --ctstate NEW -j ACCEPT
iptables -A FORWARD -m conntrack --ctstate ESTABLISHED,RELATED -j ACCEPT
iptables -A POSTROUTING -t nat -j MASQUERADE
sysctl -w net.ipv4.ip_forward=1
This concludes the networking setup. Now we need to install the cuckoo agent on the guest, by copying, located in cuckoo/agent in the El Jefe root folder, to the Windows Startup folder.

Now it's time to make a snapshot. Before executing a malware binary the snapshot is reverted and then the binary is executed and analyzed.

Next, we will see how to setup the configuration files on Cuckoo.

Change the interface to the one you are using.

Set your db connection on the [database] section and the result server IP address on the
[resultserver] section (we are using

Set the mongodb settings on the [mongodb] section.

If you are going to use only one VM, you will need to modify the stuff inside the [cuckoo1] section only, otherwise you will have to create one additional section for every extra VM you wish to use.
Inside these sections you will have to set the VM path, snapshot name and IP address of the machine (we are using

We are done setting Cuckoo for El Jefe. Before starting El Jefe remember to start the mongodb and postgresql services. You will also need to start the service from cuckoo/ in the El Jefe root folder.

So in conclusion, you should now have Cuckoo set up for yourself, integrated into the world's most powerful open source host monitoring system, and ready for your incident response team to use. We welcome any comments or questions, and of course are ready to help you if you have problems.

David Arch

Friday, April 11, 2014

Revamping El Jefe

El Jefe 2.0 - Process Chain Visualization and Heuristics

One improvement in modern Anti-Virus is the move away from signatures to heuristics and program behavior analysis on any one process. This turns out to be a decent protection at first, but as attackers evolve it will rapidly become less effective. El Jefe offers a radically different leap-frog in the level of analysis done with its protective heuristics by looking at the entire chain of process creation, rather than each process creation event alone.

Walking through the events on multiple stations

Intuitively, a human being will ask questions based on process chain anomalies:
  • Why is IExplorer.exe popping up commands like crazy? 
  • Why does Adobe PDF process spawn another process as LOCAL/SYSTEM?
  • Why did this user suddenly start using commands like wmic, which only a system administrator normally would use?
A big part of doing analysis is looking at pages and pages of text looking for patterns and ways to correlate stuff to find anomalies. This takes a tremendous amount of work, and has the risk of becoming a stultifying routine for the analyst. In other words, many companies are drowning in their own Big Security Data. 
As part of our Digital Executive Protection program, we spend hours looking at our client's high process creation data looking for potential signs of attack. One way we found to make this more effective is to add alternative ways to inspect and visualize this data.
The D3 JavaScript library provides us with  great resource for doing that. The first visualization tool we built show us easily the amount of usage of any given process. Instead of going the traditional way with big colorful circles for the most used processes, we invert it so with one fast view you can look at the processes that are hardly ever used - the exceptions which are most likely to be worth looking at further.
Process Usage: An easy way to identify process used only a small amount of time

The second graph we are experimenting with examines the relationship between all the processes executed on the system. This provides us with an easy way to move around different processes that have been run on any system and understand how they were triggered, by who and which activities were done.

Event Relationsip: Move around events and find the relationship with each other.

You might notice events are a key feature in El Jefe. One thing we want to explore is the correlation of an event's properties within multiple instances of that event. For example, when people use IE to host their trojans that IE has a very different memory and thread-count from a normal IE. This can be visualized in El Jefe and will stand out even though IE itself is not malicious.

Analyzing triggered events over time 
The second big feature we will introduce in this new version is the integration with the famous Cuckoo sandbox. We love the work that team has done over the years and it matches up perfectly with the procedure our analysts were using for El Jefe. We connect to Cuckoo within El Jefe, so now every time you find a suspicious binary, you can seamlessly ask El Jefe to grab the binary from the target machine and run it on the isolated Cuckoo sandbox. The result, for those unfamiliar with Cuckoo, is a beautiful report on what the binary has done, including files dropped, registries touched and a PCAP with everything the binary has sent over the network.
Event Inspection
Based on these new cool features we think El Jefe is a good addition to a company's security stance - given that it is both free, extensible, and more effective than traditional AV tools.

Keep tuned, because this new release is around the corner!

Monday, November 4, 2013

Exploiting CVE-2013-3881: A Win32k NULL Page Vulnerability

Microsoft Security Bulletin MS13-081 announced an elevation of privilege vulnerability []. Several days later Endgame published [] some further details on the vulnerability in question but did not provide full exploitation details. In this post we will discuss how to successfully exploit CVE-2013-3881.

The Vulnerability

The vulnerability resides in xxxTrackPopupMenuEx, this function is responsible for displaying shortcut menus and tracking user selections. During this process it will try to get a reference to the GlobalMenuState object via a call to xxxMNAllocMenuState, if the object is in use, for example: when another pop-up menu is already active, this function will try to create a new instance.

If xxxMNAllocMenuState fails it will return False but it will also set the pGlobalMenuState thread global variable to NULL. The caller verifies the return value, and in case of failure it will try to do some cleanup in order to fail gracefully.

During this cleanup the xxxEndMenuState procedure is called. This function's main responsibility is to free and unlock all the resources acquired and saved for the MenuState object, but it does not check that the pGlobalMenuState variable is not NULL before using it. As a result a bunch of kernel operations are performed on a kernel object whose address is zero and thus potentially controlled from userland memory on platforms that allow it.

Triggering the vulnerability is relatively easy by just creating and displaying two popup instances and exhausting GDI objects for the current session, as explained by Endgame. However, actually getting code execution is not trivial.


Usually a NULL dereference vulnerability in the kernel can be exploited by mapping memory at address zero in userland memory (when allowed by the OS), creating a fake object inside of this null page and then triggering the vulnerability in the kernel from the current process context of your exploit which has the null page mapped with attacker controlled data. With some luck we get a function pointer of some sort called from our controlled object data and we achieve code execution with Kernel privileges (e.g. this was the case of MS11-054). As such, NULL dereference vulnerabilities have for many years provided a simple and straightforward route to kernel exploitation and privilege escalation in scenarios where you are allowed to map at zero.

Unfortunately  in the case of CVE-2013-3881 life is not that simple, even on platforms that allow the null page to be allocated.

When xxxTrackPopupMenuEx calls xxxMNAllocMenuState and fails, it will directly jump to destroy the (non-existant) MenuState object, and after some function calls, it will inevitably try to free the memory. This means that it does not matter if we create a perfectly valid object at region zero. At some point before xxxEndMenuState returns, a call to ExFreePoolWithTag(0x0, tag) will be made. This call will produce a system crash as it tries to access the pool headers which are normally located just before the poolAddress which in this case is at address 0. Thus the kernel tries to fetch at 0-minus something which is unallocated and/or uncontrolled memory and we trigger a BSOD.

This means the only viable exploitation option is to try and get code execution before this Free occurs.

Situational Awareness

At this point we try to understand the entire behavior of xxxEndMenuState, and all of the structures and objects being manipulated before we trigger any fatal crash. The main structure we have to deal with is the one that is being read from address zero, which is referenced from the pGlobalMenuState variable:

+0x000 pGlobalPopupMenu : Ptr32 tagPOPUPMENU
+0x004 fMenuStarted : Pos 0, 1 Bit
+0x004 fIsSysMenu : Pos 1, 1 Bit
+0x004 fInsideMenuLoop : Pos 2, 1 Bit
+0x004 fButtonDown : Pos 3, 1 Bit
+0x004 fInEndMenu : Pos 4, 1 Bit
+0x004 fUnderline : Pos 5, 1 Bit
+0x004 fButtonAlwaysDown : Pos 6, 1 Bit
+0x004 fDragging : Pos 7, 1 Bit
+0x004 fModelessMenu : Pos 8, 1 Bit
+0x004 fInCallHandleMenuMessages : Pos 9, 1 Bit
+0x004 fDragAndDrop : Pos 10, 1 Bit
+0x004 fAutoDismiss : Pos 11, 1 Bit
+0x004 fAboutToAutoDismiss : Pos 12, 1 Bit
+0x004 fIgnoreButtonUp : Pos 13, 1 Bit
+0x004 fMouseOffMenu : Pos 14, 1 Bit
+0x004 fInDoDragDrop : Pos 15, 1 Bit
+0x004 fActiveNoForeground : Pos 16, 1 Bit
+0x004 fNotifyByPos : Pos 17, 1 Bit
+0x004 fSetCapture : Pos 18, 1 Bit
+0x004 iAniDropDir : Pos 19, 5 Bits
+0x008 ptMouseLast : tagPOINT
+0x010 mnFocus : Int4B
+0x014 cmdLast : Int4B
+0x018 ptiMenuStateOwner : Ptr32 tagTHREADINFO
+0x01c dwLockCount : Uint4B
+0x020 pmnsPrev : Ptr32 tagMENUSTATE
+0x024 ptButtonDown : tagPOINT
+0x02c uButtonDownHitArea : Uint4B
+0x030 uButtonDownIndex : Uint4B
+0x034 vkButtonDown : Int4B
+0x038 uDraggingHitArea : Uint4B
+0x03c uDraggingIndex : Uint4B
+0x040 uDraggingFlags : Uint4B
+0x044 hdcWndAni : Ptr32 HDC__
+0x048 dwAniStartTime : Uint4B
+0x04c ixAni : Int4B
+0x050 iyAni : Int4B
+0x054 cxAni : Int4B
+0x058 cyAni : Int4B
+0x05c hbmAni : Ptr32 HBITMAP__
+0x060 hdcAni : Ptr32 HDC__

This is the main object which xxxEndMenuState will deal with, it will perform a couple of actions using the object and finally attempts to free it with the call to ExFreePoolWithTag. The interaction with the object that occurs prior to the free are the ones we have to analyze deeply as they are our only hope in getting code execution before the imminent crash.

xxxEndMenuState is a destructor, and as such it will first call the destructor of all the objects contained inside the main object before actually freeing their associated allocated memory, for example:


The _MNFreePopup call is very interesting, as PopupMenu objects contain several WND objects and these have Handle references. This is relevant because if this WND object has its lock count equal to one when MNFreePopup is called, at some point it will try to destroy the object that the Handle is referencing. These objects are global to a user session. This means that we can force the deletion of any object within the current windows session, or at the very least decrement its reference count.

+0x000 fIsMenuBar : Pos 0, 1 Bit
+0x000 fHasMenuBar : Pos 1, 1 Bit
+0x000 fIsSysMenu : Pos 2, 1 Bit
+0x000 fIsTrackPopup : Pos 3, 1 Bit
+0x000 fDroppedLeft : Pos 4, 1 Bit
+0x000 fHierarchyDropped : Pos 5, 1 Bit
+0x000 fRightButton : Pos 6, 1 Bit
+0x000 fToggle : Pos 7, 1 Bit
+0x000 fSynchronous : Pos 8, 1 Bit
+0x000 fFirstClick : Pos 9, 1 Bit
+0x000 fDropNextPopup : Pos 10, 1 Bit
+0x000 fNoNotify : Pos 11, 1 Bit
+0x000 fAboutToHide : Pos 12, 1 Bit
+0x000 fShowTimer : Pos 13, 1 Bit
+0x000 fHideTimer : Pos 14, 1 Bit
+0x000 fDestroyed : Pos 15, 1 Bit
+0x000 fDelayedFree : Pos 16, 1 Bit
+0x000 fFlushDelayedFree : Pos 17, 1 Bit
+0x000 fFreed : Pos 18, 1 Bit
+0x000 fInCancel : Pos 19, 1 Bit
+0x000 fTrackMouseEvent : Pos 20, 1 Bit
+0x000 fSendUninit : Pos 21, 1 Bit
+0x000 fRtoL : Pos 22, 1 Bit
+0x000 iDropDir : Pos 23, 5 Bits
+0x000 fUseMonitorRect : Pos 28, 1 Bit
+0x004 spwndNotify : Ptr32 tagWND
+0x008 spwndPopupMenu : Ptr32 tagWND
+0x00c spwndNextPopup : Ptr32 tagWND
+0x010 spwndPrevPopup : Ptr32 tagWND
+0x014 spmenu : Ptr32 tagMENU
+0x018 spmenuAlternate : Ptr32 tagMENU
+0x01c spwndActivePopup : Ptr32 tagWND
+0x020 ppopupmenuRoot : Ptr32 tagPOPUPMENU
+0x024 ppmDelayedFree : Ptr32 tagPOPUPMENU
+0x028 posSelectedItem : Uint4B
+0x02c posDropped : Uint4B

In order to understand why this is so useful, let's analyze what happens when a WND object is destroyed:

pWND __stdcall HMUnlockObject(pWND pWndObject)
pWND result = pWndObject;


if (!pWndObject->cLockObj)
result = HMUnlockObjectInternal(pWndObject);
return result;

The first thing done is a decrement of the cLockObj counter, and if the counter is then zero the function HMUnlockObjectInternal is called.

pWND __stdcall HMUnlockObjectInternal( pWND pWndObject)
pWND result;
char v2;

result = pWndObject;

unsigned int entryIndex;
pHandleEntry entry;

entryIndex = pWndObject->handle & 0xFFFF;
entry = gSharedInfo.aheList + gSharedInfo.HeEntrySize * entryIndex

if ( entry->bFlags & HANDLEF_DESTROY )
if ( !(entry->bFlags & HANDLEF_INDESTROY) )
result = 0;
return result;

Once it knows that the reference count has reached zero, it has to actually destroy the object. For this task it gets the handle value and applies a mask in order to get the index of the HandleEntry into the handle table.
Then it validates some state flags, and calls HMDestroyUnlockedObject.
The HandleEntry contains information about the object type and state. This information will be used to select between the different destructor functions.

int __stdcall HMDestroyUnlockedObject(pHandleEntry handleEntry)
int index;
index = 0xC * handleEntry->bType
handleEntry->bFlags |= HANDLEF_INDESTROY;

return (gahti[v2])(handleEntry->phead);


The handle type information table (gahti) holds properties specific to each object type, as well as their Destroy functions. So this function will use the bType value from the handleEntry in order to determine which Destroy function to call.

At this point it is important to remember that we have full control over the MenuState object, and that means we can create and fully control its inner PopupMenu object, and in turn the WND objects inside this PopupMenu. This implies that we have control over the handle value in the WND object.

Another important fact is that entry zero on the gahti table is always zero, and it represents the FREE object type.

So our strategy in order to get code execution here is to, by some means, create an object whose HandleEntry in the HandleEntry table has a bType = 0x0, and bFlags = 0x1. If we can manage to do so we can then create a fake WND object with a handle that makes reference to this object of bType=0x0. When the HMDestroyUnlockedObject is called it will end up in a call gahti[0x0]. As the first element in gahti table is zero, this ends up as a "call 0". In other words we can force a path that will execute our controlled data at address zero.

What we need

We need to create a user object of bType=FREE (0x0) and bFlags= HANDLEF_DESTROY (0x1).
This is not possible directly, so we first focus on getting an object with the bFlag value equal to 0x1. For this purpose we create a Menu object, set it to a window, and then Destroy it. The internal reference count for the object did not reach zero because it is still being accessed by the window object, so it is not actually deleted but instead flagged as HANDLEF_DESTROY on the HandleEntry. This means the bFlag will equal to 0x1.

The bType value is directly associated to the Object Type. In the case of a menu object the value is 0x2 and there is no way of creating an object of type 0x0. So we focus on what ways we have to alter this value using some of the functions being called before destroying the WND object.

As you can probably remember from the PopupMenu structure shown before, it contains several WND objects, and one of the first actions performed when HMUnlockObject(pWnd) is called is to decrement the lockCount. So we simply set-up two fake WND objects in such a way that the lockCount field will be pointing to the HandleEntry->bType field. When each of those fake WND objects is destroyed it will actually perform a “dec” operation over the bType of our menu object, thus decrementing it from 0x2 to 0x0. We now have a bFlag of 0x1 and a bType of 0x0.

Using this little trick we are able to create a User object with the needed values on the HandleEntry.


First we will create a MenuObject and force it to be flagged as HANDLEF_DESTROY.

Then we will trigger the vulnerability, where xxxEndMenuState will get a reference to the menuState structure from a global thread pointer, and its value will be zero. So we map this address and create a fake MenuState structure at zero.

XxxEndMenuState will call FreePopup(..) on a popup object instance we created, and will in turn try to destroy its internal objects. Three of these objects will be fake WND objects which we also create. The first two will serve the purpose of decrementing the bType value of our menu object, and the third one will trigger a HMDestroyUnlockedObject on this same object. This will result on code execution being redirected to address 0x0 as previously discussed.

We have to remember that when we redirect execution to address 0, this memory also servers as a MenuState object. In particular the first field is a pointer to the PopupMenu object that we need to use. So what we do is to choose the address of this popup menu object in such a way that the least significant bytes of the address also represent a valid X86 jump opcode (e.g. 0x04eb translates to eb 04 in little endian memory ordering which represents a jump 4).

Finish him!

Once we achieve execution at ring 0 we patch the Enabled field on the _SEP_TOKEN_PRIVILEGES structure from the MOSDEF callback process in order to enable all the privileges for the process. We fix up the HandleEntry we modified before, and restore the stack in order to return after the PoolFree thus skipping the BSOD.

Once all of this is done we return to user-land, but now our MOSDEF process has all the privileges, this allows us to for example migrate to LSASS and get System privileges.

-- Matias Soler