Tuesday, August 22, 2017

A quick post on Wikipedia-scrubbing and a historical document on binary diffing

I am a huge fan of Wikipedia -- I sometimes browse Wikipedia like other people watch TV, skipping from topic to topic and - on average - being impressed by the quality of the articles.

One thing I have noticed in recent years, though, is that the base-democratic principles of Wikipedia open it up to manipulation and whitewashing - Wikipedia's guidelines are strict, and a person can get a lot of negative information removed just by cleverly using the guidelines to challenge entries. This is no fault of Wikipedia -- in fact, I think the guidelines are good and useful -- but it is often instructive to read the history of a particular page.

I recently stumbled over a particularly amusing example of this, and feel compelled to write about it.

More than a twelve years ago, when BinDiff was brand-new and wingraph32.exe was still the graph visualization tool of choice, there was a controversy surrounding a product called "CherryOS" - which purported to be an Apple emulator. A student had raised the allegation that "CherryOS" had misappropriated source code from an open-source project called "PearPC" on his website, and the founder of the company selling CherryOS (somebody by the name of Arben Kryeziu) had threatened the student legally over this claim.

In order to help a good cause, we did a quick analysis of the code similarities between CherryOS and PearPC, and found that approximately half of the code in CherryOS was verbatim copy & paste from PearPC. We wrote a small report, provided it to the lawyer of the student under allegation, and the entire kerfuffle died down quickly. Wikipedia used to have a page that detailed some of the drama for a few years thereafter.

I recently stumbled over the Wikipedia page of CherryOS, and was impressed: The page had been cleaned of any information that supported the code-theft claims, and offered a narrative where there had never been conclusive consensus that CherryOS was full of misappropriated code. This is not a reflection of what happened back then at all.

Anyhow, in a twist of fate, I also found an old USB stick which still contained a draft of the 2005 note we wrote. For the sake of history, here it is :-)

I had forgotten how painful it was to look at disassembly CFGs in wingraph32. Sometimes, when I am frustrated at the speed at which RE tools improved during my professional life, it is useful to be reminded what the dark ages looked like.

Saturday, October 01, 2016

"Why do you work in security instead of something more lasting ?"

This post grew out of a friend on Facebook asking (I paraphrase) "why do you spend your time on security instead of using your brainpower for something more lasting ?". I tried to answer, and ended up writing a very long reply. Another friend then encouraged me to re-post my reply to a wider audience. The below is a slightly edited and expanded version. It is much less polished than my usual blog posts, more personal, and somewhat stream-of-conscious-y. Apologies for that.

Why do I work in security instead of on something more lasting?

Predictions about what is "lasting" are very difficult to make :-). I think outside of the exploit-of-the-day, there's lasting work to be done in understanding of exploitation (because machines and automata aren't going away, and neither are programming mistakes), and I sincerely hope I'll have opportunity to do that work.

I tried my hand in cryptography / academia, and found it more prone to political trends/fads and less blindly results-oriented than security - to my great disappointment. When all attacks are of theoretical complexity 2^96, verifying and replicating results becomes difficult, and objective truth suffers (see below).

In the following, I will state a few things that I really like about the computer security community. I did not realize this immediately - instead, I learnt this over many years and engagement in other communities.
  1. Original thinkers. I used to joke that there are less than 2 dozen reasons why security as a field doesn't suck, and I know many of them personally. Now, the 2 dozen is bullshit, but what is true that in all the noise & hype, I have met a number of very fun, unconventional, and deeply insightful thinkers of very different backgrounds. They are few and far between, but I wouldn't have met them without security, and I am grateful for having met them. Many exploits require considerable inventiveness, and non-obvious / creative ways of solving problems; they are sometimes like a good joke / magic trick: With an unexpected twist that makes you laugh in disbelief.
  2. Tolerance of non-conformism and diverse educational backgrounds. There are few other industries where people who did not finish high school mix with people with postgraduate degrees, and debate on even terms. With all it's problems and biases, the part of the community I grew up with did not care about gender, skin color, or parental income - everybody was green writing on a black screen.
  3. Intellectual honesty. When discussing attacks, there is "objective truth" - you can establish whether an attack works or does not work, and checking reproducibility is easy. This is not true in many other disciplines, and "truth" becomes a matter of social consensus - even in pure math, where proof should be absolute. Having objective truth is extremely helpful to prevent a discipline to devolve into scholasticism.
Many other fields which may be more "lasting" do not have the luxury of these three points. Also be aware that my visibility into the security community is very skewed:

My skewed view of the security community

It is common to hear negative things about the community - that it is elitist, full of posturing, or of people that are mean / demeaning to others with less experience. This is not the community I experience - and this discrepancy has been puzzling me for a while.

For one thing, everybody is always nice to me. I am not sure why this is the case, but the only non-niceties I encountered in this industry were in leaked email spools. This makes it difficult for me to notice people being mean to newcomers and elitist - and it saddens me to hear that people are being shit to each other.

People weren't always nice to me - like any group of teenagers, 1990's IRC was very often not a friendly place, and #cracking would kickban you for asking a question. I found a home of sorts in a channel called #cracking4newbies - a very welcoming environment dedicated to joint learning. It was great for me: I could ask questions, and either got answers or links to documentation. A few members of #cracking were no longer active, and held status in the channel for historical reasons, #cracking4newbies on the other hand was full of eager & active youngsters.

I somehow managed to avoid being around the posturing and status games much, and in some bizarre stroke of luck, have managed to do so up to this day. The people in the security community I spend time with are genuinely interested in the technical challenges, genuinely curious, and usually do not care about the posturing part. The posturing may happen at industry conferences, but I tend to not notice - the technically interesting talks tend to adhere to substance-over-style, and the rest is as relevant to me as big advertisements for broken content inspection appliances.

All I want to say with this section is: I do not know how I managed to avoid experiencing the bad sides of the security community much. Some of it was luck, some of it was instinct. There are plenty of things I find annoying about the security community (but that is for another post :-), but in my day-to-day life, I don't experience much of it. If you are in security, and feel that the community is elitist or demeaning to people learning, I hope you succeed in seeking out the (many) people I encountered that were happy to share, explain, and just jointly nerd out on something. Feel free to reach out any time.

On building vs. breaking

I quite often hear the phrase "I quit security and I am much happier building instead of breaking things". This is a normal sentiment - but for me, security was never about "just" breaking things. Tooling was always inadeqate, workflows horribly labour-intensive, and problems were always tackled on the lowest level of abstraction, missing the forest for the trees.

In my reverse engineering classes, I always encourage people to be tool builders. Most of security work today is akin to digging trenches with chopsticks. Invest in designing and building shovels. Perhaps we will even get a bulldozer in my lifetime. Slowly but surely, the industry is changing in that direction: Microsoft is commercializing SAGE, no code auditor is more productive (even though more in-depth) than a farm of computers running AFL - but the discrepancy between the quality and quantity of tools that developers have available vs. the tools that security review has available is still vast.

I like my work most when I can cycle through building / breaking phases: Try to break something, notice how insanely badly the tooling is, cycle through an iteration of tool development, return to the breaking etc.

I realize this isn't the path for everybody, but I don't think that security is "always just about breaking". The most persistent person gets bored of chopstick-trench-digging. Invest in tooling. Being a better developer makes you a better hacker. And perhaps you like building more than breaking, and I can't fault you for that.

My friend Sören happens to be one of the best C++ developers I know. When we first met in undergraduate math class, I described what I do for a living to him (reading code for subtle mistakes), and he said "that sounds like one of the worst imaginable jobs ever". He is a builder, and I have nothing but admiration and respect for him - and from the builder's perspective, his assessment is right.

I still like finding subtle bugs. To paraphrase another person who I respect a lot: "People still search for new stuff in Shakespeare hundreds of years later".

Using security as an excuse for broad learning

I once read that "cryptography gathers many very different areas of mathematics like a focal lens". The same is very true of security and computer science. Security happens at the boundaries between layers, and I have used working in security as an excuse to learn about as many layers as possible: Low-level assembly, high-level stuff on formal verification, and even electrical engineering problems and their implications on security.
People talk about "full stack engineers" a lot; security allows me to roam the full stack of abstractions in computer science without guilt. All layers are relevant for security, all layers are interesting in their own right, and each layer has it's own funny quirks.


Given the length of this blog post, it is evident that I have asked myself the question "why do I do this" many times. And I have thought about devoting attention to other things often enough. Who knows, I am 35, so I have about 30 years of professional activity ahead of me - which may be enough to fail in one or two other fields before returning to give grandfather-security keynotes. :-)

But right now, I am actually enjoying having my hands dirty and thinking about heap layout for the first time in years.

Saturday, September 03, 2016

Essays about management in large(r) organisations (1): Process and flexibility

Even though I often profess that my primary interests are technical, by this point in my life I have been exposed to a variety of different organisations and management styles: From the self-organizing chaos of the 1996-2002 cracking/hacking groups, through the small engineering-centric startup zynamics, via the various organisations (both governmental and industry) I consulted for at some point, to the large (but nonetheless engineering-centric) culture at Google.

I enjoy thinking about organisations - their structure, how information flows, their strengths and dysfunctions. Part of it may be the influence of my father (who wrote extensively on matrix organisations, but also on organisations that fail); the other part is certainly the recognition that both company culture and organisational culture matter. In any organisation, setting the culture and organisational structure - and keeping it healthy - is paramount, and probably the key element that will allow long-term success. Ignore culture and organisation structure (both explicit and implicit) at your peril.

I had a lot of time to think in the last year, so in the coming months I will write a few posts / essays about company culture and management.

The first post is about organisational processes - why they are important, but also how they can take on a life of their own and strangle flexibility.

A technical anecdote to start with

In early 2004, the first prototype of BinDiff started to work properly - just when Microsoft released MS04-001: A series of amusing little memory corruptions inside the H.323 parsing component of Microsoft ISA server (a now-discontinued firewall product). Using BinDiff on the patch, it was evident that the problems were inside the ASN.1 PER parsing routines in a central library - but instead of fixing the library, the patch fixed the issue inside ISA server.
The patch fixed only one exploit path, but the actual vulnerability was still there. This meant that any other program using the same library remained vulnerable, and the patch had now effectively disclosed the security issue. I started searching for other applications that used this library. The first program I found which was also affected by this vulnerability was Netmeeting - Microsoft had inadvertently given a remote code execution bug in Netmeeting to everybody. It wasn't until MS04-011, at some point in April, that this vulnerability got fixed in the correct place -- the library.

The technical details of the bug are not terribly interesting - what is interesting is what the mistake revealed about flaws in Microsoft’s organisational structure, and how they reacted to the bug report.

What could we deduce/learn/extrapolate from this event?

  • Bug reports were likely routed to the product teams - e.g. if a bug is reported in your product, the bug report is routed to you.
  • Responsibility for fixing a bug appears to lie with the product teams (see above), and teams are incentivized (either directly or indirectly through feature deadlines etc.) to get bug reports “off their desk” quickly.
  • Patching shared central code is harder than patching code you own (for various reasons - perhaps compatibility concerns, other priorities from other teams, or perhaps even a heavyweight process to ask for changes in critical code).

What likely happened is that the ISA team decided that dealing with the issue on their side is enough - either because they did not realize that the same issue will affect others, or because dealing with the other team / the library is a pain, or for some other unknown reason. Microsoft’s bug fixing process incentivized “shallow” fixes, so for attackers, finding the ultimate root cause of a vulnerability could expose other vulnerable programs.

This is a classical example of making a locally convenient decision that adversely affects the larger organisation.

From what I heard, Microsoft learned from this event and made organisational changes to prevent similar mistakes in the future. They introduced a process where all patches are reviewed centrally before they go out to ensure that they don't inadvertently fix a bug in the wrong spot, or disclose a vulnerability elsewhere.

Processes as organisational learning

In what an MBA would call ‘organisational learning’, a process was created out of the experience with a previous failure in order to prevent the mistake from happening again. A process is somewhat similar to organisational scar tissue - the organisation hurt itself, and to prevent such injury in the future, the process is established.

Surprisingly, most organisations establish processes without documenting explicitly what sort of failure and what sort of incident caused the process to be established. This knowledge usually only lives in the heads of individuals that were there, or in the folklore of those that talked to those that were there. After a half a decade or so, nobody remembers the original incident - although the process will be alive and kicking.

A process can prevent an organisation from doing something stupid repeatedly - but all too often, the process takes on a life of its own: People start applying the process blindly, and in the hands of an overly-literally-minded person, the process becomes an obstacle to productivity or efficiency. The person in charge of applying and enforcing the process may themselves not know why it is there - just that it is "the process", and that bad things can happen when one doesn't follow it.

My grandfather used to say (I will paraphrase) : "a job with responsibility is a job where you don’t simply apply the rules, but need to make judgements about how and where to make exceptions". This quote carries an important truth:

People at all places in an organisation need to be ...
  1. Empowered to make exceptions: After demonstrating sound judgement, people need to feel empowered to make exceptions when the letter of a process gets in the way of the greater good and changing the process would be excessive (for example, in a one-off situation).
  1. Empowered to challenge processes: The reasoning behind a process must to be accessible to organisation members, and there needs to be a (relatively pain-free) method to propose changing the process. Since powerlessness is one of the main drivers of occupational burnout, this will help keep individuals and the organisational structure healthy.

Some organisations get the “exception” part right - most big organisations only function because people are regularly willing to bend / twist / ignore processes. Very, very few organisations get the “challenge” part right-- making sure that every employee knows and understands that processes are in the service of the company, and that improvements to processes are welcome.

I think that the failure to achieve the challenge-process frequently arises due  to "lack of institutional memory". When organisations fail to keep track of why a process was created, all sorts of harmful side-effects arise:
  1. Nobody can meaningfully judge the spirit of the process - what was it designed to prevent?
  2. Making an exception to the process is riskier - if you do not know what it was designed to prevent, how can you know that in this particular case that risk does not apply?
  3. Amending the process becomes riskier. (Same reason as above.)
  4. Challenging the process cannot happen in a decentralized / bottom-up fashion: It is often the most junior employees who may have the freshest eyes for obstructive processes - but since they do not know the history of why the processes exists, they often can't effectively propose a change since they don’t know the organisation well enough to rule out unwanted side-effects. This directly sabotages decentralised, bottom-up improvements of workflows.

What is a healthy way to deal with processes?
  1. Realize that they are a form of “organisational memory”: They are often formed as reaction to some unpleasant event - with the intent of preventing this event from repeating. It is also important to realize that unchecked and unchallenged processes can become organisational “scar tissue” - more hindrance than help.
  2. Keep track of the exact motivation for creating each process -- the “why”. This will involve writing half a page or more, and checking with others involved in the creation of the process that the description is accurate and understandable.
  3. The motivations behind the process should be accessible to everybody affected by it.
  4. Everybody should know that company processes are supposed to support, not hinder, getting work done. Everybody should feel empowered to suggest changes in a process - ideally while addressing why these changes will not lead to a repeat of the problem the process was designed to prevent.
  5. People should be empowered to deviate from the process or ignore it - but frequent or even infrequent-but-recurring exceptions are a red flag that the process needs to be improved. Don't accumulate "legacy process" and "organisational debt" through the mechanism of exception-granting.
  6. Everybody should be aware that keeping processes functional and lean is crucial to keeping the organisation healthy. Even if a process is unreasonable and obstructive, most people instinctively try to accept it - but the first instinct should ideally be to change it for the better. Constructively challenging a broken process is a service to the organisation, not an attack.
  7. It may be sensible to treat processes a bit like code - complete with ownership of the relevant process, and version control, and handover of process ownership when people change jobs. Amendments to processes can then be submitted as text, reviewed by the process owner, discussed, and eventually approved - much like a patch or removal of dead code.

Keeping an organisation healthy is hard. The most crucial ingredient to keeping it healthy, though, is that the members of the organisation care to keep it healthy. Therefore it is absolutely critical to encourage fixing the organisation when something is broken - and to not discourage people into "blindly following the process".

Wednesday, January 27, 2016

An attempt at fixing Wassenaar

Last year in May, I wrote extensively about the many ways in which the 2013 "intrusion software" amendments to the Wassenaar Arrangement were broken and downright dangerous to all efforts at security the global IT infrastructure. Since then, the debate has heated up from all sides -- extending to a hearing in front of US congress (which was pretty unanimous in condemning these amendments), but also including voices such as James Bamford arguing for these controls in an op-ed. The landscape of the discussion is too complex now to be summarized here (the interested reader can find a partial survey of recent developments here).

Common ground between the different sides of the discussion is not large, but the thing that almost everybody agrees to is "it is bad when despotic regimes that couldn't otherwise get advanced surveillance software purchase sophisticated surveillance software from abroad". How to prevent this is up for discussion, and it is unclear whether export control (and specificially the Wassenaar Arrangement) is the right tool for the task.

To find out whether the export control language can be made to work, my colleagues Mara Tam and Vincenzo Iozzo and me have worked jointly and tried to come up with an amendment to the language of the Wassenaar Arrangement that would satisfy the following criteria:

  • Make sure that click-and-play surveillance frameworks such as the ones marketed by HackingTeam or Gamma are caught and controlled.
  • Make sure that no technology that is required for defending networks (including bugs, proof-of-concept exploits, network scanners etc.) is caught and controlled.
In order to achieve this, we had to depart from the "traditional" Wassenaar language (that is focused on performance metrics and technical properties) and include much greater emphasis on "intent" and especially "informed consent by the user". We draw the line between good and bad if the design intent of the software in question is to be used against people that did not consent.

As of today, we are circulating our draft more widely. We are not 100% sure that our language achieves what we want to achieve, and we are not even sure whether what we want to achieve can be achieved within the language of export control -- but we have made a very thorough effort at testing our language against all scenarios we could come up with, and it worked well.

We are hoping that by circulating our proposal we can somewhat de-polarize the discussion and attempt to find a middle ground that everybody can be happy with -- or, failing that, to show that even with a lot of effort the 2013 amendments may end up being unfixable.

Anyhow, if you are interested in our document, you can download it here. As we get more feedback, the document will be updated and replaced with newer versions.

Tuesday, December 22, 2015

Open-Source BinNavi ... and fREedom

One of the cool things that my former zynamics colleagues (now at Google) did was the open-sourcing of BinNavi - a tool that I used to blog about quite frequently in the old days (here for example when it came to debugging old ScreenOS devices, or here for much more - kernel debugging, REIL etc.).

BinNavi is a GUI / IDE for performing multi-user reverse engineering, debugging, and code analysis. BinNavi allows the interactive exploration and annotation of disassemblies, displayed as browsable, clickable, and searchable graphs - based on a disassembly read from a PostgreSQL database, which can, in theory, be written by any other engine.

Writing UIs is hard work, and while there are many very impressive open-source reverse engineering tools around (Radare comes to mind first, but there are many others), the UI is often not very pretty - or convenient. My hope is that BinNavi can become the "default UI" to a plethora of open-source reverse engineering tools, and grow to realize it's full potential as "the open-source reverse engineering IDE".

One of the biggest obstacles to BinNavi becoming more widely adopted is the fact that IDA is the only "data source" for BinNavi - e.g. while BinNavi is FOSS, somebody that wishes to start reverse engineering still needs IDA to fill the Postgres database with disassembly.

To remedy this situation, Dave Aitel put up a contest: Anybody that either builds a Capstone-to-BinNavi-SQL-bridge or that adds decompilation as a feature to BinNavi gets free tickets to INFILTRATE 2016.

Last week Chris Eagle published fREedom, a Python-based tool to disassemble x86 and x86_64 programs in the form of PE32, PE32+, and ELF files. This is pretty awesome - because it means that BinNavi moves much closer to being usable without any non-free tools.

In this blog post, I will post some first impressions, observations, and screenshots of fREedom in action.

My first test file is putty.exe (91b21fffe934d856c43e35a388c78fccce7471ea) - a relatively small Win32 PE file, with about ~1800 functions when disassembled in IDA.

Let's look at the first function:
Left: IDA's disassembly output. Right: fREedom's disassembly output
So disassembly, CFG building etc. has worked nicely. Multi-user commenting works as expected, as does translation to REIL. Callgraph browsing works, too:

The great thing about having fREedom to start from is that further improvements can be incremental and layered - people have something good to work from now :-) So what is missing / needs to come next? 
  1. fREedom: Function entry point recognition is still relatively poor - out of the ~1800 functions that IDA recognizes in putty, only 430 or so are found. This seems like an excellent target for one of those classical "using Python and some machine learning to do XYZ" blog posts.
  2. fREedom: The CFG reconstruction and disassembly needs to be put through it's paces on big and harder executables.
  3. BinNavi: Stack frame information should be reconstructed - but not by fREedom, but within BinNavi (and via REIL). This will require digging into (and documenting) the powerful-but-obscure type system design.
  4. BinNavi: There has been some bitrot in many areas of BinNavi since 2011 - platforms change, systems change, and there are quite some areas that are somewhat broken or need updating (for example debugging on x64 etc.). Time to brush off the dust :-)
Personally, I am both super happy and pretty psyched about fREedom + BinNavi, and I hope that the two can be fully integrated so that BinNavi always has fREedom as default disassembly backend.

Wednesday, December 16, 2015

A decisionmaker's guide to buying security appliances and gateways

With the prevalence of targeted "APT-style" attacks and the business risks of data breaches reaching the board level, the market for "security appliances" is as hot as it has ever been. Many organisations feel the need to beef up their security - and vendors of security appliances offer a plethora of content-inspection / email-security / anti-APT appliances, along with glossy marketing brochures full of impressive-sounding claims.

Decisionmakers often compare the offerings on criteria such as easy integration with existing systems, manageability, false-positive-rate etc. Unfortunately, they often don't have enough data to answer the question "will installing this appliance make my network more or less secure?".

Most security appliances are Linux-based, and use a rather large number of open-source libraries to parse the untrusted data stream which they are inspecting. These libraries, along with the proprietary code by the vendor, form the "attack surface" of the appliance, e.g. the code that is exposed to an outside attacker looking to attack the appliance. All security appliances require a privileged position on the network - a position where all or most incoming and outgoing traffic can be seen. This means that vulnerabilities within security appliances give an attacker a particularly privileged position - and implies that the security of the appliance itself is rather important.

Installing an insecure appliance will make your network less secure instead of safer. If best engineering practices are not followed by the vendor, a mistake in any of the libraries parsing the incoming data will compromise the entire appliance.

How can you decide whether an appliance is secure or not? Performing an in-depth third-party security assessment of the appliance may be impractical for financial, legal, and organisational reasons.

Five questions to ask the vendor of a security appliance

In the absence of such an assessment, there are a few questions you should ask the vendor prior to making a purchasing decision:

  1. What third-party libraries interact directly with the incoming data, and what are the processes to react to security issues published in these libraries?
  2. Are all these third-party libraries sandboxed in a sandbox that is recognized as industry-standard? The sandbox Google uses in Chrome and Adobe uses in Acrobat Reader is open-source and has undergone a lot of scrutiny, so have the isolation features of KVM and qemu. Are any third-party libraries running outside of a sandbox or an internal virtualization environment? If so, why, and what is the timeline to address this?
  3. How much of the proprietary code which directly interacts with the incoming data runs outside of a sandbox? To what extent has this code been security-reviewed?
  4. Is the vendor willing to provide a hard disk image for a basic assessment by a third-party security consultancy? Misconfigured permissions that allow privilege escalation happen all-too often, so basic permissions lockdown should have happened on the appliance.
  5. In the case of a breach in your company, what is the process through which your forensics team can acquire memory images and hard disk images from the appliance?
A vendor that takes their product quality (and hence your data security) seriously will be able to answer these questions, and will be able to confidently state that all third-party parsers and a large fraction of their proprietary code runs sandboxed or virtualized, and that the configuration of the machine has been reasonably locked down - and will be willing to provide evidence for this (for example a disk image or virtual appliance along with permission to inspect).

Why am I qualified to write this?

From 2004 to 2011 I was CEO of a security company called zynamics that was acquired by Google in 2011. Among other things, we used to sell a security appliance that inspected untrusted malware. I know the technical side involved with building such an appliance, and I understand the business needs of both customers and vendors. I also know quite a bit about the process of finding and exploiting vulnerabilities, having worked in that area since 2000.

Our appliance at the time was Debian-based - and the complex processing of incoming malware happened inside either memory-safe languages or inside a locked-down virtualized environment (emulator), inside a reasonably locked-down Linux machine. This does not mean that we never had security issues (we had XSS problems at one point where strings extracted from the malware could be used to inject into the Web UI etc.) - but we made a reasonable effort to adhere to best engineering practices available to keep the box secure. Security problems happen, but mitigating their impact is not rocket science - good, robust, and free software exists that can sandbox code, and the engineering effort to implement such mitigations is not excessive.

Bonus questions for particularly good vendors

If your vendor can answer the 5 questions above in a satisfactory way, his performance is already head-and-shoulders above the industry average. If you wish to further encourage the vendor to be proactive about your data security, you can ask the following "bonus questions":
  1. Has the vendor considered moving the Linux on their appliance to GRSec in order to make privilege escalations harder?
  2. Does the vendor publish hashes of the packages they install on the appliance so in case of a forensic investigation it is easy to verify that the attacker has not replaced some?

Monday, May 25, 2015

Why changes to Wassenaar make oppression and surveillance easier, not harder

Warning to EU readers: EU writing culture lays out arguments to draw a strong statement as conclusion, US writing culture seems to prefer a strong statement up front, followed by the arguments. This article follows US convention.

Adding exploits to Wassenaar was a mistake if you care about security

The addition of exploits to the Wassenaar arrangement is an egregious mistake for anyone that cares about a more secure and less surveilled Internet. The negative knock-on effects of the agreement include, but are not limited to, the following list:

  • It provides governments with a massive coercive tool to control public security research and disadvantage non-military security research. This coercive power need not be exercised in order to chill public research and vulnerability disclosure.
  • It tilts the incentive structure strongly in favor of providing all exploits to your host government, and makes disclosure or collaborative research across national boundaries risky
  • It provides a way to prohibit security researchers from disseminating attack tools uncovered on compromised machines.
  • It risks fragmenting, balkanizing, and ultimately militarizing the currently existing public security research community.

The intention of those that supported the amendment to Wassenaar was to protect freedom of expression and privacy worldwide; unfortunately, their implementation achieved almost the exact opposite. 

With friends of such competence, freedom does not need enemies. The changes to Wassenaar need to be repealed, along with their national implementations.

A pyrrhic victory with unintended consequences

In December 2013, activists worldwide celebrated a big success: Intrusion Software was added to the list of technologies regulated by the Wassenaar Arrangement. In the cyber activist community, people rejoiced: Finally, the people they call "cyber arms dealers" would no longer be able to act with impunity. Oppressive regimes would no longer be able to buy the software that they use for mass surveillance and repression. A victory for freedom and democracy, no doubt.

Unfortunately, the changes to the regulation have horrible knock-on effects for security research, privacy, and society at large. The change to the Wassenaar Arrangement achieves the exact opposite of what it was intended to do.

The difficulties of being a security researcher

To discuss the many ways in which this regulation is flawed requires some background on the difficulties faced by security researchers worldwide:

Security research is an activity fraught with many difficulties. There are few historical precedents where a talented 20-year old can accidentally conceive a method capable of intruding into and exfiltrating information out of hundreds of well-fortified institutions. The best (if very cheesy) analogy I can come up with that explains the difficulties of young researchers are the X-Men comic books -- where teenagers discover that they have a special ability, and are suddenly thrust into a much bigger conflict that they have to navigate. One week you're sitting in your room doing something technically interesting, a few weeks later people in coffee shops or trains may strike up a conversation with you, trying to convince you that government X is evil or that they could really be helpful fighting terrorist organisation Y.

Security researchers face a fundamental problem: In order to prove exploitability, and in order to be 100% sure that they are not crying wolf, they need to demonstrate beyond any doubt that an attack is indeed possible and reliable. This means that the researcher needs to build something that is reliable enough to be dangerous.

Once successful, the researcher is in a very difficult spot -- with no evident winning move. What should he/she do with the exploit and the vulnerability?

Different people will posit different things as "the right behavior" at this point. Most people argue the researcher should provide the vulnerability (and sometimes the exploit) to the software vendor as quickly as possible, so that the vulnerability can be fixed. This often comes with risk -- for many closed-source programs, the researcher had to violate the EULA to find the vulnerability, and many vendors perceive vulnerability reports as attention-seeking at best and blackmail at worst. In the best case, reporting simply involves some extra work that will not be compensated, in the worst case the researcher will face legal and/or extralegal threats by the party he/she is reporting the vulnerability to.

After the researcher hands over the vulnerability and exploit, he/she is often made to wait for many months -- wondering if the people he provided his code to will fix the issue as swiftly as possible -- or if they are silently passing on information to third parties. In many cases, he/she will receive little more than a handshake and a "thank you" for his efforts.

At the same time, various parties are likely to offer him money for the vulnerability and the exploit -- along with a vague promise/assurance that it will only used for "lawful purposes". Given the acrobatics and risks that responsible disclosure carries, it is unsurprising that this offer is tempting. Anything is better than "legal risk with no reward".

Partly to alleviate this imbalance, more mature vendors have begun introducing bug bounties -- small payments that are meant to encourage disclosing the bug to the vendor. The sums are smaller than in the grey market, but -- by and large -- enough to compensate for the time spent and to offer the researcher positive recognition. Talented researchers can scrape out a living from these bounties.

Security researchers are in an odd position: In order to check the validity of their work, they need to create something with the inherent potential for harm. Once the work is shown to be valid, the result becomes the object of desire of many different military and intelligence organisations which, given the general scarcity of "cyber talent", would love to get these researchers to cooperate with them. The software industry has grudgingly accepted the existence of the researchers, and the more mature players in that industry have understood the value of their contributions and tried to re-shape the playing field so that getting security issues reported and fixed is incentivized.

The researcher has to balance a lot of difficult ethical deliberations -- should the exploit be sent to the vendor? Can the people that wish to buy the exploit on the grey market be trusted when they claim that the exploit will save lives as it will be used to prevent terrorist strikes? What is a reasonable timeframe for a vendor to fix the issue? Can disclosure accelerate the process of producing a fix and thus close the window of opportunity for the attacker more quickly? Is fixing the issue even desirable?

There are no 100% simple answers to any of the above questions -- each one of them involves a long and difficult debate, where reasonable people can disagree depending on the exact circumstances.

Adding exploits to Wassenaar, and shoddily crafted regulation

The Wassenaar Arrangement is not a law by itself -- it is an agreement by the participating countries to pass legislation in accordance with the Wassenaar Arrangement, which then stipulates that "export licenses" have to be granted by governments before technology listed in the agreement can be exported.

From an engineering perspective, it is good to think of Wassenaar as a "reference implementation" of a law -- different legal systems may have to adapt slightly different wording, but their implementation will be guided by the text of the arrangement. The most damaging section in the new version reads as follows:
Intrusion software:
"Software" specially designed or modified to avoid detection by 'monitoring tools', or to defeat 'protective countermeasures', of a computer or network capable device, and performing any of the following:
a. The extraction of data or information, from a computer or network capable device, or the modification of system or user data; or
b. The modification of the standard execution path of a program or process in order to allow the execution of externally provided instructions.
This formulation is extremely broad -- any proof-of-concept exploit that defeats ASLR or a stack canary (or anything else for that matter) and achieves code execution falls under this definition. Even worse -- by using a formulation such as "standard execution path" without properly defining how this should be interpreted, it casts a shadow of uncertainty over all experimentation with software. Nobody can confidently state that he knows how this will be interpreted in practice.

Legal uncertainty and coercive power to nation states

Legal uncertainty is almost always to the benefit of the powerful. Selective enforcement of vague laws that criminalize common behavior is a time-honored technique -- perhaps best embodied in Beria's famous statement "Find me the man and I will find you the crime".

The principle is simple: As long as a person conforms to the expectations of society and the powerful, the laws are not stringently applied -- but as soon as the person decides to challenge the status quo or not cooperate with the instruments of power, the laws are applied with full force.

I live in Switzerland. Most other security researchers that I like to collaborate with are spread out all over the world, and I routinely travel accross international borders. Occasionally, I carry working exploits with me -- I have to do this if I wish to collaborate with other researchers, discuss new discoveries, or perform investigations into reliable methods of exploitation. Aside from the physical crossing of borders, it is routine to work jointly on a shared code repository accross borders and timezones.

The newly created legal situation criminalizes this behavior. It puts me personally at huge risk -- and gives local governments a huge coercive tool to make me hand over any information about zero-day I may find ahead of time: If I do not apply for an export license before visiting another researcher in Germany or France or the US, I will be breaking export control regulations.

The cyber activists that celebrated Wassenaar have mainly made sure that every security researcher that regularly leaves his country to collaborate with others can be coerced into cooperation with their host government easily. They have made it easier for all governments to obtain tools for surveillance and oppression, not harder.

A story of Elbonia and Wassenaar

Let us imagine a young researcher Bob in a country named Elbonia. Elbonia happens to be a Wassenaar member, and otherwise mostly under the rule of law with comparatively solid institutions. 
Bob finds a number of vulnerabilities in a commonly used browser. He reports these to the vendor in the US, and the vendor publishes fixes and awards bug bounties.

The domestic intelligence service of Elbonia has a thorny problem on the counterterrorism side -- separatists from one of their provinces have repeatedly detonated bombs in the last years. The intelligence service could really use some good exploits to tackle this. Unfortunately, they have had difficulties hiring in recent years, and they do not have the tooling or expertise -- but they do see the security bulletins and the bug bounties awarded to Bob.

They decide to have a coffee with Bob -- who does not want to work with the intelligence service and would prefer to get the problems fixed as quickly as possible. The friendly gentlemen from the service then explain to the researcher that he has been breaking the law, and that it would be quite unfortunate if this led to any problems for him -- but that it would be easy to get permission for future exports, provided that a three-month waiting period is observed in which the Elbonian intelligence service gets to use the exploits for protecting the national security and territorial integrity of Elbonia.

What would the researcher do?

Balkanising and nationalising an international research community

The international security research community has greatly contributed to our understanding of computer security over the last 20+ years. Highly international speaker line-ups are the norm, and cooperation between people from different nations and continents is the norm rather than the exception. 

The criminalization of exporting exploits risks balkanising and nationalising what is currently a thriving community whose public discussion of security issues and methods for exploitation benefits everybody.

The implementation of Wassenaar makes it much easier and less risky to provide all exploitable bugs and associated exploits to the government of the country you reside in than to do any of the following:
  • report the vulnerability to a vendor and provide a proof-of-concept exploit
  • perform full disclosure by publishing an exploit in order to force a fix and alert the world of the problem
  • collaborate with a researcher outside of your home country in order to advance the state of our understanding of exploitation and security

Wassenaar heavily tilts the table toward "just sell the exploit to your host government".

Making "defense" contractors the natural place to do security research

With the increased risk to the individual conducting research, and a fragmentation of the international research community, what becomes the natural place for a security researcher to go to pursue his interest? What employer is usually on great terms with his host government, and can afford to employ a significant number of other researchers with the same nationality?

The changes to Wassenaar make sure that security researchers, which in the past few years have been recruited in large numbers into large, private-sector and consumer-facing companies, will have much less attractive prospects outside of the military-industrial complex. A company that does not separate people by nationality internally and is unused to heavy classification and compartmentalisation will simply not want to run the risk of violating export-control rules -- which means that the interesting jobs will be offered by the same contracting firms that dominate the manufacturing of arms. These companies are much less interested in the security of the global computing infrastructure -- their business model is to sell the best method of knocking out civilian and military infrastructure of their opponents.

Will the business of the defense contractors be impacted? This is highly doubtful -- the most likely scenario is that host governments for defense contractors will grant export licenses, provided that the exploits in question are also provided to the host government. The local military can then defend their systems while keeping everybody else vulnerable -- while keeping good export statistics and "protecting Elbonian jobs".

Everybody that cares about having more secure systems loses in this scenario.

Weakening defensive analysis

The list of knock-on effects that negatively affect the security of the Internet can be extended almost arbitrarily. The changed regulation also threatens to re-balance the scales to make the public analysis of surveillance and espionage toolkits harder and riskier.

The last few years have seen a proliferation of activist-led analysis of commercial surveillance tools and exploit chains -- publicly accessible analysis reports that disassemble and dissect the latest government-level rootkit have become a regular occurrence. This has, no doubt, compromised more than one running intelligence-gathering operation, and in general caused a lot of pain and cost on the side of the attackers.

Some implants / rootkits / attack frameworks came packaged with a stack of previously-unknown vulnerabilities, and most came with some sort of packaged exploit.

It is very common for international groups of researchers to circulate and distribute samples of both the attack frameworks and the exploits for analysis. Quick dissemination of the files and the exploits they contain is necessary so that the public can understand the way they work and thus make informed decisions about the risks and capabilities.

Unfortunately, sending a sample of an attack framework to a researcher in another country is at risk of being made illegal.

On export controls and their role in protecting privacy and liberty

There is a particular aspect in the lobbying for adding exploits to Wassenaar that I personally have a very hard time understanding: How did anyone convince himself that the amendmends to Wassenaar were a good idea, or that export control would be helpful in preventing surveillance?

Any closer inspection of the agreement and the related history will bring to light that it was consistently used in the past to restrict the export of encryption -- as you can read here, Wassenaar restricted export of any mass-market encryption with key sizes in excess of 64 bits in the late 1990s. 

Everybody with any understanding of the cryptographic history knows that export licenses were used as a coercive mechanism by governments to enhance their surveillance ability -- "you can get an export license if you do key escrow, or if you leak a few key bits here and there in the ciphertext". To this day, security of millions of systems is regularly threatened by the remains of "export-grade encryption".

Even today, there are plenty of items on the Wassenaar restrictions list that would have great pro-privacy and anti-surveillance implications -- like encrypted, frequency-hopping radios. 

I have a difficult time understanding how anyone that claims to support freedom of expression and wishes to curb surveillance by governments would provide additional coercive power to governments, instead of advocating that encrypted, frequency-hopping radios should be freely exportable and available in places of heavy government surveillance.


While the goal of restricting intrusive surveillance by governments is laudable, the changes to Wassenaar threaten to achieve the opposite of their intent -- with detrimental side effects for everybody. The changes need to be repealed, and national implementations of these changes rolled back.

(Thanks to all pre-reviewers for useful comments & suggestions :-)