End-User Credential Security

This week’s announcement that a Russian crime syndicate has amassed 1.2 billion unique usernames and passwords across 420,000 websites would seem like startling news in 72-point font on the front of major newspapers, if it wasn’t sad it was such a commonplace announcement these days.  With four more months to go and still higher than the estimated 823 million compromised credentials part of 2013 breaches affecting Adobe to Target, it’s from Black Hat 2014 I find myself thinking about what we as ISV’s, SaaS providers, and security professionals can do to protect users in the wake of advanced persistent threats and organized, well-funded thieves wreaking havoc on the digital identities and real assets of our clients and customers.

Unlike Heartbleed or other server-side vulnerabilities, this particular credential siphoning technique obviously targeted users themselves to affect so many sites and at least 542 unique addresses affecting at least half that many unique users.  Why are users so vulnerable to credential-stealing malware?  To explore this issue, let’s immediately discard a tired refrain inside software houses everywhere: users aren’t dumb.  All too often, good application security is watered down to its least secure but most useful denominator for an overabundance of concern that secure applications may frustrate users, lower adoption, and reduce retention and usage.  While it is true that the more accessible the Internet becomes, the wider the spectrum the audience that uses it, from the most expertly capable to the ‘last-mile’ of great grandparents, young children, and the technologically unsophisticated.  However, this should neither be grounds to dismiss end-user credential security as a concern squarely in service provider’s court to address nor should it be an excuse to fail to provide adequately secure systems.  End-user education is our mutual responsibility, even if that means three more screens, additional prompts to confirm identity or action, or an out-of-band verification process.  Keeping processes as stupefying simple as possible because our SEO metrics show that’s the way to marginally improve adoption, reduce cart abandonment, or improve site usage times breeds complacency that ends up hurting us all in the long-run.

Can we agree that 1FA needs to end?  In an isolated world of controlled systems, a username and password combination might have been a fair assertion of identity.  Today’s systems, however, are neither controlled or isolated – the same tablets that log into online banking also run Fruit Ninja for our children, and we pass them over without switching out any concept of identity on a device that can save our passwords and represent them without any authentication.  Small-business laptops often run without real-time malware scanning software, easily harvesting credentials through keystroke logging, MitM attacks, cookie stealing, and a variety of other commonplace techniques.  Username and passwords fail us because they can be saved and cached just as easily as they can be collected and forwarded to command and control servers is Russia or elsewhere.  I’m not one of those anarchists advocating ‘death to the password’ (remember Vidoop?), but using knowledge-based challenges (password, out-of-wallet questions, or otherwise) as the sole factor of authentication needs to end.  And it needs to end smartly: sending an e-mail ‘out of band’ to an inbox loaded in another tab on the same machine, or an SMS message read by Google Voice in another tab means your ‘2FA’ is really just one factor layered twice instead of two-factor authentication.  A few more calls into the call center to help users cope with 2FA will be far cheaper in the long-run than the fallout of a major credential breach that affects your sites users.

We need to also discourage poor password management: allowing users to choose short or non-complex passwords and warning them about their poor choices is no excuse – we should just flatly reject them.  At the same time, we need to recognize that forcing users to establish too complex of a password will encourage them to establish a small number of complex passwords and reuse them across more sites.  This is one of the largest Achilles’s Heels for end-users: when a compromise of one site does occur, and especially if you have removed the option for users to establish a username not tied to their identity (name, e-mail address, or otherwise), you have made it tremendously easier for those who have gathered credentials from one site to have a much higher likelihood of exploiting them on your site.  Instead, we should consider nuances to each of complexity requirements that would make it likely a user would have to generate a different knowledge-based credential for each site.  While that in of itself may increase the chance a user would ‘write a password down’, a user who stores all their passwords in a password manager is still arguably more secure than the user who users one password for all websites and never writes it anywhere.

Finally, when lists of affected user accounts become available in uploaded databases of raw credentials that are leaked or testable on sites such as – ACT.  Find out your users that have overlap with compromised credentials on other sites, and proactively flag or lock their accounts or at least message to them to educate and encourage good end-user credential security.  We cannot unilaterally force users to improve the security of their credentials, but we can educate them, and we can make certain their eventual folly through our inaction.


When to Ride the Service Bus

One of the great things about adding new, senior talent to a storied team working on a large, complex, and successful enterprise application solution is the critical technical review that results in a lot of “why did/didn’t you do it this way?” questions.  You have two options for responding to those questions – ignore or passively dismissing them, or taking the questions seriously as a challenge to prove out why you would make a decision you and your team made 5 years ago the same if you had to consider for the first time today, in today’s frameworks, development methodologies, and the current team makeup and skills inventory.  If you choose to dismiss these opportunities to critically review your prior decisions, it says a lot about your management style, general appreciation of technology and response to its change, and positions your team to take a reactionary, defensive posture to architecture rather than create a team that honors a proactive, continuous improvement perspective.  Far more interesting too are those questions that ask why the system is architected in a general way, rather than a theological debate on whether a particular technology component choice is superior to all over or one’s preferred/familiar choice.

The particular question the new engineer asked was, “Why aren’t we using a service bus?”  Instead of answering him directly, I figured this as a good opportunity to explore the previous decision we made that not only did not include an enterprise service bus (ESB) in the original design, but rejected its inclusion when it was strongly suggested by our first customer because they were standardizing on a service bus-centric architecture themselves.  The primary advantage of a service bus is to layer an abstraction across heterogeneous systems by implementing a centralized communication mechanism across components.  By applying this architectural model, you can get some key benefits including orchestration, queuing to handle intermittent component availability, and extensibility points for message routing to alter dispatch logic or transform messages.  Implementing the service bus pattern requires some kind of adapter to be written for each component of the system, either as a local modification to each component or by choosing to standardize on a communication channel provided by the ESB.  Even in the latter, usually some minor accommodation is required to allow the ESB to receive and encapsulate the native message for delivery to the destination component. Our first customer was a notable player in the community banking market, and was productizing multiple new SaaS-based web applications that depended on data feeds coming from many different customers.  In their scenarios, data was consumed by one application, parsed, and delivered to other applications, which in turn may have created additional data feeds for other products, in a cyclic communication/dependency non-directed graph.  Each application was developed by different teams, and there was no unified technology stack adoption – some teams were developing on EJB and Flex, others were pure .NET, and teams generally had the discretion to choose whatever they could argue would solve the job, without a strong technology leader looking to unify the stack for similar applications that delivered CMS and pseudo-online banking functionality using a common input data set.

For this customer, ESB was a solution to a problem – their choices lead to a highly concurrent development process with multiple independent teams – but also supported connecting a heterogeneous environment of interdependent components, each of which accomplished limited objectives.  This organization was running red-hot – developing ancillary products to a highly engaged and fanatic client base of community banks, where their limiting factor was their speed of innovation and delivery.  By agreeing on a common communication mechanism that ESB could provide, there was something, albeit low-level, to which all teams agreed.  In the ‘controlled agile chaos’ they found themselves in, the abstraction bought them flexibility to adapt changing business requirements using orchestration.  In theory anyway – they ended up moving much slower than they anticipated, but this wasn’t the fault of ESB. ESB solves two classes of problems.  The first is the common use case of large, disparate enterprises looking to marry systems established from the dawn of client-server architectures to the newest Node.js hotness, without having to bend the will of any particular system to the communication conventions of any other, which may prove impossible if both systems are proprietary.  This is a common use case for BizTalk, especially in the financial sector.  All the other benefits you can name off from a service bus architecture are really secondary advantages to this key objective.  The second is the use case that any layer of indirection provides: an abstraction you can use to increase the speed of development when requirements are incomplete or prone to pivot.  In each case, you invest in a layer to reduce the cost of future change. This particular customer chose NServiceBus as their message-oriented middleware.  We seriously evaluated both the general architectural concepts ESB as well as the particular technology they suggested and came up with a definitive ‘no’ to that choice.  While it made a lot of sense for our customer, it did not make sense for us because:

  1. We did not require guaranteed event handling.  Our system connected to a system of record that provided transactional consistency, and virtually all state changes were initiated by users through a web browser.  A timeout was preferable to queued command handling system because of the possibility of duplicate transactions that frustrated users may initiate, not realizing their requests were queued.  Second, our interconnected systems did not provide guaranteed event handling, so the guaranteed provided by the ESB would now be honored end-to-end.  Third, we are using the Windows Identity Foundation with sliding time expirations end-to-end from the user’s browser through the lowest layer of service components, which doesn’t bode well for delayed delivery situations, even if the user was willing to wait.
  2. We do require transformation, but not orchestration between our components.  Our system features adapter-based design to allow multiple types of endpoints to be serviced by a single service implementation for those portions that may need to connect to a different type of third-party system through a provider model implementation loaded by dependency injection.  We could have chosen to use ESB for this piece, however, we perceived the long-term maintenance cost of multiple providers with the party-specific transformation logic to be lower than maintaining those transforms in ESB scripting or adapters.  In reviewing this perception today, I believe it was still the right decision because is allowed for us to unit-test our transformation logic without including the ESB.
  3. An ESB is a single point of failure that would independently need to scale for load exponentially proportional to the number of service interconnects in our solution, and would add some amount of latency between each. Because online banking is a mission-critical, customer-facing solution, it cannot have SPOF’s in any portion of the architectural design.  The SPOF nature of an ESB can be mitigated in multiple ways, but we felt that was at least two layers of complexity we could solve in other, simpler ways.
  4. All middleware increases the Mean Time Between Failures (MTBF).  This is not a risk specific to ESB, but of any layer added to a system.  If you add an ORM, IOC, ESB, or even a logging aspect, something can go wrong with them.  Each component has some small, but measurable failure rate, and when inserted into the communication chain between all components, its reliability of 99.999% still contributes to a reduction in the overall reliability of a serial system.  This is where the KISS principle shines – complexity creates unreliability, so all complexity must generate a compelling benefit in excess of its potential to fail.
  5. We wanted our application layer to be the platform, we did not want ESB to be the platform.  This was a business case / competitive advantage decision that we wanted to build as a feature of our system that the same services layer that supported our front-end user interfaces was also an open and extensible platform upon which our clients could integrate to, which would increase the overall value proposition of online banking not only as a sticky end-user experience, but also as a value proposition to capitalize on our solution as the middleware that marries together all the disparate systems within a financial institution, which ultimately online banking does like no other piece of technology within a bank or credit union.  We felt that by positioning everything behind an ESB, the perceived value of our technology piece would be lessened without additional client education.
  6. MSMQ made us feel dirty enough, and we did not want to mandate it for each component because it was in 2009 and still is relatively difficult to debug, and lately we have learned, queues do not work well with used with Layer 7 network load balancing.  The new hotness of 0MQ wasn’t around then, and while RabbitMQ was, it was arguably not production ready by that time.  For us, production-ready isn’t just whether a component is capable, but whether it will have general acceptance from the IT departments of our large clients – many newer technologies that are FOSS or from vendors without an establish track record require a ‘sale’ and buy-in during due diligence, long before ink is applied to a contract.  Even if they were options for the ESB queuing mechanism, they would not resolve the larger aforementioned concerns.
  7. At the time we made this choice, AMQP was an amorphous draft that did not solidify until later.  The lack of a vendor-independent protocol between components and an ESB made the choice to utilize an ESB subject to vendor lock-in, which we were not willing to tolerate for such a critical component.
  8. Because our product was both the end-user experience and the middleware we were writing, we felt strongly that the application protocol should provide descriptive metadata and support fast client proxy generation using .NET-based tools.  REST support was archaic at best (HttpRequest anyone?) in .NET 3.5, and to this day, consuming SOAP services is intrinsically more verbose in C# and VB.NET (HttpClient) than consuming REST or AMQP services due to a lack of better library and integrated language support for it.  Looking back on this, with a large amount of iterative change we went through from ideation to Version 1.0 of our solution, we could not have moved as fast without a fast way to regenerate proxies that would cause build failures to alert us of service operation signature changes — tracking these down at runtime (REST) or having to debug a secondary system (ESB) to find these would have bogged down our delivery timelines.
  9. A lesser concern was we felt that tracing SOAP messages, while definitely more difficult than REST, would be more difficult debug any issues in AMQP or other ESB encapsulation protocols than inspecting SOAP envelopes with built-in WCF tools already present in the .NET development stack.

So, that’s quite a case against an ESB, but they do have compelling uses for certain environments – just not ours.  Like all technology selection decisions, it’s important to pick the right tool for the job, and improve your tools as needed.  A standalone ESB would have provided significant benefits if we were developing with proprietary/closed third-party systems that were part of a call chain that required orchestration, or if we were developing with a heterogeneous mix of technologies.  In our case, we had a predictable homogeneous .NET environment based on web services, consumers of our API are our own technologies or a limited number of customers who were also using .NET, and we had no legacy baggage.  With the widespread adoption of WS-* standards, we have chosen to obtain some of the benefits, such as federation, from those standards rather than an ESB feature, which ultimately we believe makes our platform easier to support and distribute for our future API consumers.  Other side benefits such as logging are handled as separated concerns through dependency injection rather than external interceptors in a communication channel, a possibility for us only because we control the portion of the stack that requires orchestration.  And finally, by keeping all communication as SOAP over HTTP/HTTPS, we gain features like load balancing from Layer 7 network devices instead of an ESB process, which are much easier to switch out and upgrade.

The central design decision we made was that ESB’s provide some great features and that ties you into an ESB, but if we could get those features another way that was just as convenient or more so, we’d prefer the plug-and-play flexibility of leveraging existing solutions for components such as caching and load balancing in the environment our solution operates, or pick those pieces ad-hoc for those concerns rather than pick the best omnibus solution and work around any specific shortcomings for any one of them. In reviewing the current industry literature and blog posts and looking at general trends, it would seem our decision not to marry our solution is generally the path many take when not required to integrate legacy systems as part of an orchestration chain or when using non-HTTP based transport mechanisms.  If you’re using one, hopefully it’s for a good and necessary reason!  For us, though, we decided not to hop on a service bus that could take us somewhere we already arrived.

* As an aside, we actually did end up rolling our own small “ESB” as a TCP port multiplexer that queues and portions out connectivity to a socket-based, legacy third-party component that has no listener back-queue and no port concurrency, highly unusual for a server process.  Each connection consumes the port fully for the duration of the short transaction, and we had to write a way to buffer M number of requests and hand them off to (M-N) number of available ports as they became available,in a specialized type of producer-consumer problem. In hindsight, this was an opportunity to use an ESB, but in our case, we only required message routing and load leveling, and in a few hundred lines of code, we implemented what we needed for this particular third-party system what would have taken us far longer to do as our first time using an ESB. That being said, should we encounter this with another vendor, it would make sense to review using an ESB for this type of functionality in the future.
Leave a comment

Posted by on April 1, 2014 in Programming


Tags: , ,

The Wires Cannot Be Trusted; Does DRM Have Something to Teach Us?

In the continuing revelations about the depth to which governments have gone to subjugate global communications in terms of privacy, anonymity, and security on the Internet, one thing is very clear: nothing can be trusted anymore.

Before you wipe this post off as smacking of ‘conspiracy theorist’, take the Snowden revelations disclosed since Christmas, particularly regarding the NSA’s Tailored Access Operations catalog that demonstrates the ways they can violate implicit trust in local hardware by infecting firmware at a level where even reboots and factory ‘resets’ cannot remove the implanted malware, or their “interdiction” of new computers that allow them to install spyware between the time it leaves the factory and arrives at your house.  At a broader level, because of the trend in global data movement towards centralizing data transit through a diminishing number of top tier carriers – a trend is eerily similar to wealth inequality in the digital era – governments and pseudo-governmental bodies have found it trivial to exact control with quantum insert attacks.  In these sophisticated attacks, malicious entities (which I define for these purposes as those who exploit trust to gain illicit access to a protected system) like the NSA or GCHQ can slipstream rogue servers that mimic trusted public systems such as LinkedIn to gain passwords and assume identities through ephemeral information gathering to attack other systems.

Considering these things, the troubling realization is this is not the failure of the NSA, the GCHQ, the US presidential administration, or the lack of public outrage to demand change.  The failure is in the infrastructure of the Internet itself.  If anything, these violations of trust simply showcase technical flaws we have chosen not to acknowledge to this point in the larger system’s architecture.  Endpoint encryption technologies like SSL became supplanted by forward versions of TLS because of underlying flaws not only in cipher strength, but in protocol assumptions that did not acknowledge all the ways in which the trust of a system or the interconnects between systems could be violated.  This is similarly true for BGP, which has seen a number of attacks that allow routers on the Internet to be reprogrammed to shunt traffic to malicious entities that can intercept it: a protocol that trusts anything is vulnerable because nothing can be trusted forever.

When I state nothing can be trusted, I mean absolutely nothing.  Your phone company definitely can’t be trusted – they’ve already been shown to have collapsed to government pressure to give up the keys to their part of the kingdom.  The very wires leading into your house can’t be trusted, they could already or someday will be tapped.  Your air-gapped laptop can’t be trusted, it’s being hacked with radio waves.

But, individual, private citizens are facing a challenge Hollywood has for years – how do we protect our content?  The entertainment industry has been chided for years on its sometimes Draconian attempts to limit use and restrict access to data by implementing encryption and hardware standards that run counter to the kind of free access analog storage mediums, like the VHS and cassette tapes of days of old, provided.  Perhaps there are lessons to be learned from their attempts to address the problem of “everything, everybody, and every device is malicious, but we want to talk to everything, everybody, on every device”.  One place to draw inspiration is HDCP, a protocol most people except hardcore AV enthusiasts are unaware of that establishes device authentication and encryption across each connection of an HD entertainment system.  Who would have thought when your six year old watches Monsters, Inc., those colorful characters are protected by such an advanced scheme on the cord that just runs from your Blu-ray player to your TV?

While you may not believe in DRM for your DVD’s from a philosophical or fair-use rights perspective, consider the striking difference with this approach:  in the OSI model, encryption occurs at Layer 6, on top of many other layers in the system.  This is an implicit trust of all layers below it, and this is the assumption violated in the headlines from the Guardian and NY Times that have captured our attention the most lately: on the Internet, he who controls the media layers also controls the host layers.  In the HDCP model, the encryption happens more akin to Layer 2, as the protocol expects someone’s going to splice a wire to try to bootleg HBO from their neighbor or illicitly pirate high-quality DVD’s.  Today if I gained access to a server closet in a corporate office, there is nothing technologically preventing me from splicing myself into a network connection and copying every packet on the connection.  The data that is encrypted on Layer 6 will be very difficult for me to make sense of, but there will be plenty of data that is not encrypted that I can use for nefarious purposes: ARP broadcasts, SIP metadata, DNS replies, and all that insecure HTTP or poorly-secured HTTPS traffic.  Even worse, it’s a jumping off point for setting up a MITM attack, such as an SSL Inspection Proxy.  Similarly, without media-layer security, savvy attackers with physical access to a server closet or the ability to coerce or hack into the next hop in the network path can go undetected if they redirect your traffic into rogue servers or into malicious networks, and because there is no chained endpoint authentication mechanism on the media-layer, there’s no way for you to know.

These concerns aren’t just theoretical either, and they’re not to protect teenagers’ rights to anonymously author provocative and mildly threatening anarchist manifestos.  They’re to protect your identity, your money, your family, and your security.  Only more will be accessible and controllable on the Internet going forward, and without appropriate protections in place, it won’t just be governments soon who can utilize the assumptions of trust in the Internet’s architecture and implementation for ill, but idealist hacker cabals, organized crime rings, and eventually, anyone with the right script kiddie program to exploit the vulnerabilities once better known and unaddressed.

Why aren’t we protecting financial information or credit card numbers with media-layer security so they’are at least as safe as Mickey Mouse on your HDTV?


Tags: , , ,

Scaling Enterprise Database-Bound Applications: I/O

Optimizing Slow Accesses

While most software developers like to think of themselves as computer scientists in the purest sense of the term, with job duties that would include intimately understanding and exploiting efficiencies of the x64 processor platform, optimizing that critical-path O(log n) algorithm to perform in O(log log n) time, and other acts of mathematical creativity and scientific application, that’s not what most software developers do (or should be doing if they are).

Most software developers are building business (retail B2C, B2B API’s, or LOB‘s), not scientific applications — and that means most are developing I/O-bound, not CPU-bound applications.  Specifically, most business applications are creative user or application programming interfaces around relatively mundane CRUD operations on a data store.  Even more complex applications that perform data synchronization or novel calculations of co-variance or multivariate regression consume maybe 5% of their time crunching data, and the other 95% of the time retrieving and sending it on.

So, when you design an enterprise application and get past the ideation phase and start scaling out your next-generation game-changing application from a cute demo to a serious and robust application serving millions of requests, why would you bother with refactoring your string concatenation in loops into string builders, aiming for zero-copy, or optimizing for CPU performance?  You should not and you should:  You should not be optimizing for CPU performance, unless you have optimized all your slow accesses away — and you should be optimizing for CPU performance because hopefully you’ve already squeezed all the blood out of the I/O turnip you can.

But you haven’t.  I know you haven’t.  You know you haven’t if you are being honest.  Have you ever looked at your database queries per second for specific-entity queries?  For instance, let’s say a user logs into your enterprise application, and a service on your application tier needs to retrieve the record of a user.  That service might call another service to make a record of the user’s login.  Then the user navigates to another page in your application 60 seconds later.  How many times did any component of your system retrieve the user by their unique identifier?  If the answer is, “I don’t know”, you haven’t scratched the surface of scaling an enterprise application, much less my most important axiom of doing so: “Don’t Repeat Requests“.

This is a lot harder than you might think, because enterprise web application development lends itself to repeating requests, and it is not an easy problem to solve, because you are essentially creating state on an application tier for a web tier that hosts a stateless HTTP application protocol.  When functionality is segregated into multiple services with distinct responsibilities, there is some duplication of I/O access that happens to fulfill a request that is unavoidable.  Unless you and everyone on your team completely understands this disjoint and works collectively to design solutions that do not repeat requests, you will repeat requests as part of the natural design of any system.

Caching Isn’t a Magic Bullet, But It Is a Bullet

If you thought this post was going to end at “implement second-level caching on your ORM of choice”, you’re wrong, but you should be doing that for sure.  This is usually as easy installing a caching server like Couchbase, configuring your ORM in a few lines of code or configuration files, and wala – you are still repeating your requests, but this time, answering your repeated requests will be a lot faster than any SSD-backed database server will ever be.

(I say ‘usually’, because this depends on how you’re using your ORM.  If you use your ORM as an expensive way to execute stored procedures, your ORM will be at best a pass-through for database methods and will not give you the benefit of entity caching that could be reused for multiple queries that include that entity as a result.  As with all caching, YMMV depending on how you have designed your layers.)

Once you enable caching, measure.  Measure how many times you ask for that user record when a user logs in and performs some actions over time.  You’ll be amazed that when you view this from a database request level, you will still be asking for the same user over and over again as long as not every component is using the cache for database entities with a consistent cache key.  It’s very hard to get right, both from an application configuration and a caching server configuration perspective — do not assume, but do measure.

Remember, the most important thing to remember is not to get really fast answers to your repeated questions, but stop asking the same questions over and over again!  Caching at the ORM is your tourniquet to stop the bleeding of your performance into database I/O buffers and wait times, but caching at the inter-component request level is critical.  Let’s say you have an enterprise web application that retrieves a forecast for a city for a given period of time.  The web client makes the request for the locale and date range to your application tier, which translates that into queries of whatever entities comprise your data model.  With ORM second-level caching in effect, the next request for the same locale and date range will not ask the question of the database this time, but the answer will come instead from the second-level cache… but stop right there.  The question was asked again at a higher level, you’re just answering it in a more intelligent way the second time around.

Enterprise web applications need to cache the responses of service requests using a cache key that accounts for the parameters of the request.  Hopefully your web application faithfully implements a repository pattern, and if so, you implement a cache into this layer to eliminate repeated requests to the service layer to start with. This is not easy.  This is hard because your ORM’s database caching is likely a black box implementation of complex cache expiration logic that performs all sorts of clever tricks to know when an entity has become ‘dirty’ and needs to be retrieved again from the underlying database rather than use the cached copy.  If you’re developing business applications, you’re probably not accustomed to being clever at this level, and you will need to spend the time to implement this manually throughout your repository pattern (unless you thought ahead and can add caching as an aspect) and to bust your caches.

Challenges of Busting Caches

Busting your own caches – that is, invalidating a cached entry when you have reason to know the cached version is no longer good – is one of the trickiest things to get right in this stage of Don’t Repeat Requests.  Let’s take a service method called GetUser() that returns the user and an object graph of some interesting things that cover multiple data entities from the database.  At the web tier, we start caching that call when we make it so subsequent calls from the web tier won’t even request this from the service while its in cache.  But what else could change the User object in the database?  If the user themselves can, then that’s easy enough to know to bust a cache on a User repository .Save() method, but if other unrelated processes can, such as say, a back-end service process that bulk-updates users for some reason, then this gets more challenging to ensure you’ve identified all the paths that could invalidate the data and make sure each have access to bust the cache for the GetUser() response as cached by the web tier as well as the User entity as represented in any other request (think GetUser(), GetUsersByWhatever(), and all the other variants that may also need cache busting).  When GetUser() actually includes data sourced from other entities, you have to think about the dependent object graph in the data model and ensure you’ve accounted for these as well.  You just have to consider but not handle this recursive analysis for deep object graphs — it only matters as much as it matters for the user experience.

This kind of task must be reserved for the architects and most senior engineers who know your system design and inter-dependencies inside and out to avoid data consistency errors.  A key point is as long as all data validation logic is performed at the lowest layer under any custom caching work you perform, data consistency errors will at worst create a poor user experience.  If you don’t – if you have critical client-side validation that is not mirrored under caching on the service-side of your architecture, you have bigger security risks and other problems than caching, but this will definitely impede your ability to deploy service request caching and scale your application.

Caching From Within

Within any area of your application, beware anti-patterns that repository patterns can create.  If you author MethodA() that calls MethodB() that calls MethodC(), all of which individually call UserRepository.GetUser(), then you’re recursively repeating yourself.  Repository patterns are nice because they reduce the repetitive session and connection management functions involved with making a web service or database call, but they make it easy to forget that they’re very, very heavy methods.

Do not be afraid to accumulate.  Do not be afraid to pass object graphs through method parameters to save I/O.  You could think about the call stack as your cache here, and while you shouldn’t load it up as an unnecessarily heavy omnibus object to pass around to any method, and while you definitely should not front-load all your I/O before calling a logical method chain before conditional logic or exception management could make some of the calls unnecessary, intelligently design methods so they don’t take the smallest parameter set possible, but create the best scalability when working in concert.

Caching Outside Your Boundaries

If you’re writing enterprise web applications for a product that is not dying or decaying, you’re writing it in HTML5 today.  And if your web design isn’t from a Frontpage 98 template, you’re probably using AJAX requests either to improve user experiences and reduce perceived page load times or maybe you’ve gone whole-hog into an SPA design.  With HTML5 and a relatively modern web browser, you have LocalStorage.  Use LocalStorage.

You should be using LocalStorage to cache and bust non-error responses to AJAX requests to your web services and REST endpoints. Just because you’ve thinned out the pipes from services to the database and from the web tier to the services tier, why stop there?  Why continue to allow browsers to repeat requests to your web tier as a user moves back and forth between areas or pages?  If you rest on your laurels on a job-well-done, but still repeat unnecessary I/O queries at a level higher up in the chain, then you’ve made your application more performant but not truly scalable — you’ve just shifted the blame.

The F5 Test

I propose what I will call the “F5 Test” for scalability.  When you’ve cached all you can cache, and every layer is implementing the “Don’t Repeat Requests” mantra, open up your database profiler and your Couchbase cache hit dashboard.  Log into your application’s dashboard, reporting, or whatever page you want to test, then clear your profiler and cache hit counters.  Press F5.  You should see very, very little activity on a reload, and you should be able to explain what you do see.

But, for what you do see, justify each and don’t make excuses for yourself:

  1. If your dashboard makes repeated requests because you feel it “always needs to be up-to-date”, then you’re doing it wrong.  Cache and use server-side events to refresh your cached copy.
  2. If you load a user object to determine whether they have a login session, then do you have a good reason for not using browser evidence such as a signed SAML assertion to validate a session instead of using a database lookup to verify a user exists and is authorized?
  3. If you see something you can’t explain, investigate.  I wish this was as obvious as it is intuitive, but many times software developers will be content with an arbitrary improvement (I made 232 database calls on login go down to 47) rather than to do the homework to find out why 47 isn’t 5.  Maybe there are 42 extraneous requests made by a service that doesn’t use the cache even though you thought it did.  Maybe one of those 42 requests causes database locking escalations that won’t scale with load.

Optimizing Query Plans

Oh yeah, and optimize query plans.  This is important work, but it’s not the outer-most layer of the onion.  It’s important to remember the difference between scalability and performance:

  • Performance should be determined by the user experience from dispatching of the request to final rendering of the result to the user in their browser.  Performance is not “how much CPU does the system use under load” – that is resource utilization, though many people use performance for both concepts.
  • Scalability is two-fold: How many users can I get a certain level performance on a certain hardware basline (scaling up), and can I and how often will I have to throw money at more hardware to handle more users at the same level of performance (scaling out)?
  • Improving performance may or may not improve scalability
  • Improving scalability rarely improves performance
  • Management will not understand the difference

Optimizing query plans can impact both: improving a query plan from 6 seconds to 1 second improves performance.  It could improve scalability if your queries are over complex joins or large data sets that couldn’t be pinned in memory automagically in your database server.  But optimizing query plans for speed alone is not a function of scalability — optimizing them for I/O is where it’s at.  Simple improvements like changing JOIN’s to EXISTS’s where feasible allow the query engine to skip unnecessary I/O is what opens up buffers and improves throughput through the disk subsystems where the big performance and scalability penalties hit.  It just so happens complex queries that have I/O in intermediate steps also have high CPU due to hash matching, rewinds, and other operations that perform calculations on large amounts of data generated from unnecessary I/O.

It’s work you should do, but you shouldn’t do it first for scalability reasons.

After-Thoughts: Don’t Report Stupid Results

Building highly-scalable applications from the ground up with a large team is impossible.  You iterate scalability just as you iterate product features.  Actually, hopefully you iterate scalability tasks along with user stories, but in actuality, complex enterprise web applications are usually architected with the best of intentions with intelligent designs, but reach a breaking point at some level of load on some hardware platform that cause a stop-drop-and-roll effort to improve the scaling up and out of an application.  Companies with deadlines and tight deliverable schedules don’t consistently evaluate and factor the required work to make and keep an application scalable over time into iterations.  If someone tells you differently, they’re probably in sales and they’re definitely lying.

That being said, software developers, do not succumb to the pressure to deliver scalability improvements by reporting true but irrelevant statistics to management.

  • “I sped up database calls for GetUser() by 300%!” suggests anything that gets a user should see a three-fold improvement in speed.  If that database call is 1% of the login process time, then it will have no material impact.
  • “I reduced the size of page requests from 500K to 250K!” means “I doubled the performance or scalability of the application” to management, but in reality, it means neither.
  • “I found a problem between ServiceA and ServiceB and cut out three extraneous calls between them!” means nothing to anyone.  Did you remove three calls that are made once an hour by a batch process, or three calls made for every user login?  What was the impact of those calls on performance and scalability before and after the optimization?
  • “ServiceA is a big problem and has a lot of errors.  I removed a lot of exceptions on ServiceA.  Exceptions cause performance problems.” is problematic on several levels.  Why were the exceptions being thrown?  Did removing them fix or just sweep a real problem under the carpet?  If it was justified, what improvement did it have on the overall system?

When software developers communicate their changes, it implies they have meaningful impact. However, many software developers fail to measure the before and after impact of their changes on the whole system, but typically only evaluate them in the microcosm of the area they changed.  This is about as useful as management suggesting areas they should fix based on intuition or high-level reporting tools.

While most devs don’t do scientific computing, scaling applications is an empirical task that demands meaningful measurement in a realistic testing context.  There is spec document or product owner guidance on improving scalability: you must treat it as a scientific experiment.  Observe, hypothesize, have a control (the pre-change measurement), experiment, report data.  If you fail to discretely value each change with before and after metrics, you’re just shooting in the dark.  Cowboy coding gets teams into scalability messes, not out of them.

Especially, though, don’t give updates on enhancements that you cannot verify improve scalability with before and after numbers.  If you fix a problem that doesn’t improve the overall system scalability, which happens often in scalability improvement iterations, highlighting your accomplishments when there is no observable improvement suggest you are either ineffective or not working on the right items.  Worse, in crunch times, providing such updates gives a false sense of accomplishment to management.  Improving scalability, or performance for that matter, has no done-state.  But providing meaningless accomplishment notes to management will accelerate the sense of “we’re done enough”, when in fact, you may not have even identified the most significant issue to your scalability for your particular scenario.

And if you haven’t, let me do it for you:  You’re repeating your requests.  Trust me on that one. 🙂

Leave a comment

Posted by on December 4, 2013 in Programming


A Brief Introduction to Part-of-Speech Tagging

A field of computer science that has captured my attention lately is computational linguistics — the inexact science of how to get a computer to understand what you mean.  This could be something as futuristic as Matthew Broderick’s battle with the WOPR, or with something more practical, like Siri.  Whether it be text entered by a human into a keyboard or something more akin to understanding the very unstructured format of human speech, understanding the meaning behind parsed words is incredibly complex — and to someone like me — fascinating!

My particular interest as of late is parsing — which from a linguistic perspective, means the breaking down of a string of characters into words, their meanings, and stringing them together in a parse tree, where the meanings of individual words as well as the relationships between words is composed into a logical construct that allows higher order functions, such as a personal assistant.  Having taken several foreign language classes before, then sitting on the other side of the table as an ESL teacher, I can appreciate the enormous ambiguity and complexity of any language, and much more so English among Germanic languages, as to creating an automated process to parse input into meaningful logical representations.  Just being able to discern the meaning of individual words given the multitude of meanings that can be ascribed to any one sequence of characters is quite a challenge.

Parsing Models

Consider this:  My security beat wore me out tonight.

In this sentence, what is the function of the word beat?  Beat functions as either a noun or a verb, but in this context, it is a noun.  There are two general schools of thought around assigning a tag as to what part of speech (POS) each word in a sentence functions as — iterative rules-based methods and stochastic methods.  In rules-based methods, like Eric Brill’s POS tagger, a priority-based set of rules that set forth language-specific axioms, such as “when a word appears to be a preposition, it is actually a noun if the preceding word is while”.  A complex set of these meticulously constructed conditions is used to refine a more course dictionary-style assignment of POS tags.

Stochastic methods, however, are more “fuzzy” methods of building advanced statistical models of how words should be tagged not based on a procedural and manual analysis of edge cases and their mitigations, but using training models over pre-tagged corpra, in a manner hearkening to the training sets applied to neural networks.  These trained models are then used as a baseline for assigning tags to incoming text, but no notable option for correction of any specific error or edge case other than retraining the entire model is available for refinement.  One such very interesting concept is treating the tagging of parts of speech as Hidden Markov Models, which is a probabilistic model that strives to explain how a process with a defined pattern that is not known other than sparse characteristics of the model and the inputs and the outputs through the process.

This continues to be a good candidate for doctorial theses in computer science disciplines.. papers that have caused me to lose too much sleep as of late.

Parsing Syntax

Even describing parts of speech can be as mundane as your elementary school grammar book, or as rich as the C7 tagset, which provides 146 unique ways to describe a word’s potential function.  While exceptionally expressive and specific, I have become rather fond of the Penn Treebank II tagset, which defines 45 tags that seem to provide enough semantic context for the key elements of local pronoun resolution and larger-scale object-entity context mapping.  Finding an extensively tagged Penn Treebank corpus proves difficult, however, as it is copyright by the University of Pennsylvania, distributed through a public-private partnership for several thousand dollars, and the tagged corpus is almost exclusively a narrow variety of topics and sentence structures — Wall Street Journal articles.  Obtaining this is critical to use as a reference check for writing a new Penn Treebank II part-of-speech tagger, and it prevents the construction of a more comprehensive Penn-tagged wordlist, which would be a boon for any tagger implementation.  However, the folks at the NLTK has provided a 10% free sample under Fair Use that has provided somewhat useful for both checking outputs in a limited fashion, but also for generating some more useful relative statistics about relationships between parts of speech within a sentence.

To produce some rudimentary probabilistic models to guide ambiguous POS-mappings for individual words, I wrote a five-minute proof of concept that scanned the NLTK-provided excerpt of the WSJ Penn Treebranch corpus to produce probabilities of what the next word’s part of speech would be given the previous word’s tag. The full results are available in this gist.

Future Musings

My immediate interest, whenever I get some free time on a weekend (which is pretty rare these days due to the exceptional pace of progress at our start-up), is pronoun resolution, which is the object of this generation’s Turing Test — the Winograd Schemas.  An example of such a challenge is to get a machine to answer this kind of question — Joe’s uncle can still beat him at tennis, even though he is 30 years older. Who is older? This kind of question is easy for a human to answer, but very, very hard for a machine to infer because (a) it can’t cheat to Google a suitable answer, which some of the less impressive Turing Test contestant programs now do, and (b) it requires not only the ability to successfully parse a sentence into its respective parts of speech, phrases, and clauses, but it requires the ability for a computer to resolve the meaning of a pronoun.  That’s an insanely tough feat!  Imagine this:

“Annabelle is a mean-spirited person.  She shot my dog out of spite.”

A program could infer “my dog” is a dog belonging to the person providing the text.  This has obvious applications in the real world if you can do this, and it has been done before.  But, imagine the leap in context that is exponentially harder to overcome when resolving “She”.  This requires not only an intra-sentence relationship of noun phrases, possessive pronouns, direct objects, and adverbial clauses, but it also requires the ability to carry context forward from one sentence to the next, building a going “mental map” of people, places, things — and building a profile of them as more information or context is provided.  And, if you think that’s not hard enough to define .. imagine the two additional words appended on to this sentence:

, she said.

That would to a human indicate dialog, which requires a wholly separate frame of Inception-style reference between contextual frames.  The parser is reading text about things which is actually being conveyed by other things — both sets of frames have their own unique, but not necessarily separate, domains and attributes.  I’m a very long-way off from ever getting this diversion in my “free time” anywhere close to functioning as advertised… but, then again, that’s what exercises on a weekend are for — not doing, but learning. 🙂


Posted by on August 22, 2013 in Programming


Robustness in Programming

(For my regular readers, I know I promised this post would detail ‘a method by which anyone could send me a message securely, without knowing anything else about me other than my e-mail address, in a way I could read online or my mobile device, in a way that no one can subpoena or snoop on in between.’  A tall order, for sure, but still something I am working to complete in an RFC format.  In the meantime…)

I have the benefit of supporting an engineering group that is seeing tremendous change and growth well past ideation and proof of concept, but at the validation and scaling phases of a product timeline.  One observation I’ve made about the many lessons taught and learned as part of this company and product growth spurt have been the misapplication of the Jon Postel’s Robustness Principle.  Many technical folks are at least familiar with, but often can quote the adage: “Be conservative in what you do, be liberal in what you accept from others“.  Unfortunately, like many good pieces of advice, this is taken out of context when it relates to software development.

First off, robustness, while it sounds positive, it not a trait you always want.  This can be confusing for the uninitiated, considering antonyms of the word include “unfitness” and “weakness”.  On a macro-scale, you want a system to be robust; you a product to be robust.  However, if you decompose an enterprise software solution into its components, and those pieces into their individual parts, the concerns do not always need to, and in some cases should not, be robust.

For instance, should a security audit log be robust?  Imagine a highly secure software application that must carefully log each access attempt to the system.  This system is probably designed so that many different components of the system can write data to this log, and imagine the logging system is simple and writes its output to a file.  If this particular part of the system were robust, as many developers define it, it must, as well as possible, attempt to accept and log any messages posted to it.  However, implemented this way, it is subject to CRLF attacks, whereby a component that can connect to it and insert a delimiter that would allow it to add false entries to the security log.  Of course, you developers say, you need to do input checking and not allow such a condition to pass through to the log.  I would go much further and state you must be as meticulous as possible about parsing and throwing exceptions or raising errors for as many conditions as possible.  Each exception that is not thrown is an implicit assumption, and assumptions are the root cause of 9 out the OWASP Top 10 vulnerabilities in web applications.

Robustness can, and is often, an excuse predicated by laziness.  Thinking about edge cases and about the assumptions software developers make with each method they write is tedious.  It is time consuming.  It does not advance a user story along its path in an iteration.  It adds no movement towards delivering functionality to your end users.  Recognizing and mitigating your incorrect assumptions, however, is an undocumented but critical requirement for the development of every piece of a system that does store, or may ever come in contact with, protected information.  Those that rely on the Robustness Principle must not interpret “liberal” to mean “passive” or “permissive”, but rather “extensible”.

In the previous example I posited about a example logging system, consider how such a system could remove assumptions but still be extensible.  The number and format of each argument that comprises a log entry should be carefully inspected – if auditing text must be descriptive, then shouldn’t such a system reject a zero or two-character event description?  While information systems should be localizable and multilingual, shouldn’t all logs be written in one language and any characters that are not of that language omitted and unique system identifiers within the log languages’ character set used instead?  If various elements are co-related, such as an account number and a username, shouldn’t they be checked for an association instead of blindly accepting them as stated by the caller?  If the log should be chronological, shouldn’t an event specified in the future or too far in the past be rejected?  Each of these leading questions exposes a vulnerability a careful assessment of input checking can address, but which is wholly against most developers’ interpretations of the Robustness Principle.

However, robustness is not about taking whatever is given to you, it is about very carefully checking what you get, and if and only if it passes a litany of qualifying checks, accepting it as an answer to an open-ended question, rather than relying on a defined set of responses, when possible.  A junior developer might enumerate all the error states he or she can imagine in a set list or “enum”, and only accept that value as valid input to a method.  While that’s a form of input checking, it is wholly inextensible, as the next error state any other contributor wishes to add will require a recompile/redeploy of the logging piece, and potentially every other consumer of that component.  Robustness need not require all data be free-form, it must simply be written with foresight.

Postel, wrote his “law” with reference to TCP implementations, but he never suggested that TCP stack implementers liberally accept TCP segments with such boundless blitheness that they infer the syntax of whatever bits they received, but rather, they should not impose an understanding of the data elements that were not pertinent to the task at hand, nor enforce one specific interpretation of a specification upon upstream callers.  And therein lies my second point — robustness is not about disregarding syntax, but about imposing a convention.  Robust systems must fail as early and as quickly as possible when syntax, especially, has been violated or cannot be accurately and unambiguously interpreted, or if the context or state of a system is deemed to be invalid for the operation.  For instance, if a receives a syntactically valid message but can determine the context is wrong, such as a request for information from a user who lacks an authorization to that data, every conceivable permutation of invalid context should be checked, not fail to consider each in a blasé fashion to leave room for a future feature that may, someday, require an assumption made in the present, if it is ever to be developed.  This crosses another threshold beyond extensibility to culpable disregard.

In conclusion, building a robust system requires discretion in interpretation of programming “laws” and “axioms”, and an expert realization that no one-liner assertions were meant by their authors as principles so general to apply to every level of technical scale of the architecture and design of a system.  To those who would disagree with me, I would say, then to be “robust” yourself, you have to accept my argument. 😉

Leave a comment

Posted by on August 7, 2013 in Programming


When All You See Are Clouds… A Storm Is Brewing

The recent disclosures that the United States Government has violated the 4th amendment of the U. S. Constitution and potentially other international law by building a clandestine program that provides G-Men at the NSA direct taps into every aspect of our digital life – our e-mail, our photos, our phone calls, our entire relationships with other people and even with our spouses, is quite concerning from a technology policy perspective.  The fact that the US Government (USG) can by legal authority usurp any part of our recorded life – which is about every moment of our day – highlights several important points to consider:

  1. Putting the issue of whether the USG/NSA should have broad access into our lives aside, we must accept that the loopholes that allow them to demand this access expose weaknesses in our technology.
  2. The fact the USG can perform this type of surveillance indicates other foreign governments and non-government organizations likely can and may already be doing so as well.
  3. Given that governments are often less technologically savvy though much more resource-rich than malevolent actors, if data is not secure from government access, is it most definitely not secure from more cunning hackers, identity thieves, and other criminal enterprises.

If we can accept the points above, then we must accept that the disclosure of PRISM and connotation through carefully but awkwardly worded public statements about the program present both a problem and an opportunity for technologists to solve regarding data security in today’s age.  This is not a debate of whether we have anything to hide, but rather a discussion of how can we secure data, because if we cannot secure it from a coercive power (sovereign or criminal), we have no real data security at all.

But before proposing some solutions, we must consider:

How Could PRISM Have Happened in the First Place?

I posit an answer devoid of politics or blame, but on an evaluation of the present state of Internet connectivity and e-commerce.  Arguably, the Internet has matured into a stable, reliable set of services.  The more exciting phase of its development saw a flourishing of ideas much like a digital Cambrian explosion.  In its awkward adolescence, connecting to the Internet was akin to performing a complicated rain dance that involved WinSock, dial-up modems, and PPP, sprinkled with roadblocks like busy signals, routine server downtime, and blue screens of death.  The rate of change in equipment, protocols, and software was meteoric, and while the World Wide Web existed (what most laypeople consider wholly as “the Internet” today), it was only a small fraction of the myriad of services and channels for information to flow.  Connecting to and using the Internet required highly specialized knowledge, which both increased the level of expertise of those developing for and consuming the Internet, while limiting its adoption and appeal – a fact some consider the net’s Golden Age.

But as with all complex technologies, eventually they mature.  The rate of innovation slows down as standardization becomes the driving technological force, pushed by market forces.  As less popular protocols and methods of exchanging information give way to young but profitable enterprises that push preferred technologies, the Internet became a much more homogeneous experience both in how we connect to and interact with it.  This shapes not only the fate of now-obsolete tech, such as UUCP, FINGER, ARCHIE, GOPHER, and a slew of other relics of our digital past, but also influenced the very design of what remains — a great example being identification and encryption.

For the Internet to become a commercializable venue, securing access to money, from online banking to investment portfolio management, to payments, was an essential hurdle to overcome.  The solution for the general problem of identity and encryption, centralized SSL certificate authorities providing assurances of trust in a top-down manner, solves the problem specifically for central server webmasters, but not for end-users wishing to enjoy the same access to identity management and encryption technology.  So while the beneficiaries like Amazon, eBay, PayPal, and company now had a solution that provided assurance to their users that you could trust their websites belonged to them and that data you exchanged with them was secure, end-users were still left with no ability to control secure communications or identify themselves with each other.

A final contributing factor I want to point out is that other protocols drifted into oblivion, more functionality was demanded over a more uniform channel — the de facto winner becoming HTTP and the web.  Originally a stateless protocol designed for minimal browsing features, the web became a solution for virtually everything, from e-mail (“webmail”), to searching, to file storage (who has even fired up an FTP client in the last year?).  This was a big win for service providers, as they, like Yahoo! and later Google, could build entire product suites on just one delivery platform, HTTP, but it was also a big win for consumers, who could throw away all their odd little programs that performed specific tasks, and could just use their web browser for everything — now even Grandma can get involved.  A more rich offering of single-shot tech companies were bought up or died out in favor of the oligarchs we know today – Microsoft, Facebook, Google, Twitter, and the like.

Subtly, this also represented a huge shift on where data is stored.  Remember Eudora or your Outlook inbox file tied to your computer (in the days of POP3 before IMAP was around)?  As our web browser became our interface to the online world, and as we demanded anywhere-accessibility to those services and they data they create or consume, those bits moved off our hard drives and into the nebulous service provider cloud, where data security cannot be guarenteed.

This is meaningful to consider in the context of today’s problem because:

  1. Governments and corporate enterprises were historically unable to sufficiently regulate, censor, or monitor the internet because they lacked the tools and knowledge to do so.  Thus, the Internet had security through obscurity.
  2. Due to the solutions to general problems around identity and encryption relying on central authorities,  malefactors (unscrupulous governments and hackers alike) have fewer targets to influence or assert control over to tap into the nature of trust, identity, and communications.
  3. With the collapse of service providers into a handful of powerful actors on a scale of inequity on par with a collapse of wealth distribution in America, there exist now fewer providers to surveille to gather data, and those providers host more data on each person or business that can be interrelated in a more meaningful way.
  4. As information infrastructure technology has matured to provide virtual servers and IaaS offerings on a massive scale, fewer users and companies deploy controlled devices and servers, opting instead to lease services from cloud providers or use devices, like smartphones, that wholly depend upon them.
  5. Because data has migrated off our local storage devices to the cloud, end-users have lost control over their data’s security.  Users have to choose between an outmoded device-specific way to access their data, or give up the control to cloud service providers.

There Is A Better Way

Over the next few blog posts, I am going to delve into a number of proposals and thoughts around giving control and security assurances of data back to end-users.  These will address points #2 and #4 above as solutions that layer over existing web technologies, not proposals to upend our fundamental usage of the Internet by introducing opaque configuration barriers or whole-new paradigms.  End-users should have choice whether their service providers have access to their data in a way that does not require Freenet’s darknets or Tor’s game-of-telephone style of anonymous but slow onion-routing answer to web browsing.  Rather, users should be able to positively identify themselves to the world and be able to access and receive data and access it in a cloud-based application without ever having to give up their data security, not have to trust of the service provider, be independent to access the data on any devices (access the same service securely anywhere), and not have to establish shared secrets (swap passwords or certificates).

As a good example, if you want to send a secure e-mail message today, you have three categorical options to do so:

  1. Implicitly trust a regular service provider:  Ensure both the sender and the receiver use the same server.  By sending a message, it is only at risk while the sender connects to the provider to store it and while the receiver connects the provider to retrieve it.  Both parties trust the service provider will not access or share the information.  Of course, many actors, like Gmail, still do.
  2. Use a secure webmail provider:  These providers, like, encrypt the sender’s connection to the service to protect the message as it is sent, and send notifications to receivers to come to a secure HTTPS site to view the message.  While better than the first option, the message is still stored in a way that can be demanded by subpoena or snooped inside the company while it sits on their servers.
  3. Use S/MIME certificates and an offline mail client:  While the most secure option for end-to-end message encryption, this cumbersome method is machine-dependent and requires senders and receivers to first share a certificate with each other – something the average user is flatly incapable of understanding or configuring.

Stay tuned to my next post, where I propose a method by which anyone could send me a message securely, without knowing anything else about me other than my e-mail address, in a way I could read online or my mobile device, in a way that no one can subpoena or snoop on in between.



Tags: ,

Doing Your Due Diligence on Security Scanning and Penetration Testing Vendors

All too often, development shops and IT professionals become complacent with depending on packaged scanning solutions or a utility belt of tools to provide security assurance testing of a hosted software solution.  In the past five years, a number of new entrants to the security evaluation and penetration testing market have created some compelling cloud-based solutions to perimeter testing.  These tools, while exceptionally useful for a sanity check of firewall rules, load balancer configurations, and even certain industry best practices in web application development, are starting to create a false sense of security in a number of ways.  As these tools proliferate, infrastructure professionals are becoming increasingly dependent upon their handsomely-crafted reporting about PCI, GLBA, SOX, HIPPA, and all the other regulatory buzzwords that apply to certain industries.  If you’re using these tools, have you considered:

Do you use more than one tool?  If not, and you should, is there any actual overlap between their testing criteria?

There is a certain incestuous phenomenon that develops in any SaaS industry that sees high profit margins: entrepreneurs perceive cloud-based solutions as having a low barrier to entry.  This perception drives new market entrants to cobble together solutions to compete for share in the space.  But are these fly-by-night competitors competitively differentiated from their peers?

Sadly, I have found in practical experience this not to be the case.  Too many times have I have enrolled in a free trial of a tool or actually shelled out for some hot new cloud-based scanning solution to find at best only existing known vulnerabilities are duplicatively reported by this new ‘solution’, with only false positives appearing as the ‘net new’ items to bring to my attention.  Here in lies the rub — when new entrants to this market create competing products, there is an iterative reverse engineering that goes on — they run existing scanning products on the market against websites, check to see those results, and make sure they develop a solution that at least identifies the same issues.

That’s not good at all.  In any given security scan, you may see, perhaps, 20% of the total vulnerabilities a product is capable of finding show up as a problem in a scan target.  Even if you were to scan multiple targets, you may only be seeing mostly the same kinds of issues in each subsequent scan.  Those using this as a methodology to build quick-to-market security scanning solutions are delivering sub-par offerings that may only identify 70% of the vulnerabilities other scanning solutions do.  eEye has put together similar findings in an intriguing report I highly recommend reading.  Investigating the research and development activities of a security scanning provider is an important due diligence step to make sure when you get an “all clear” clean report from a scanning tool, that report actually means something.

How do you judge your security vendor in this regard?  Ask for a listing of all specific vulnerabilities they scan for.  Excellent players in this market will not flinch at giving you this kind of data for two reasons: (1) a list of what they check for isn’t as important as how well and how thoroughly they actually assess each item, and (2) worthwhile vendors are constantly adding new items to the list, so it doesn’t represent any static master blueprint for their product.

Does your tool test more than OWASP vulnerabilities?

The problem with developing security testing tools is in part the over-reliance on the standardization of vulnerability definition and classifications.  While it is helpful to categorize vulnerabilities into conceptually similar groups to create common mitigation strategies and mitigation techniques, too often security vendors focus on OWASP attack classifications as the definitive scope for probative activities.  Don’t get me wrong, these are excellent guides for ensuring the most common types of attacks are covered, but they do not provide a comprehensive test of application security.  Too often the types of testing such as incremental information disclosure, where various pieces of the system provide information that can be used to discern how to attack the system further, are relegated to manual penetration testing instead of codified into scanning criteria.  Path disclosure and path traversal vulnerabilities are a class of incremental information disclosures that are routinely tested for by scanning tools, but they represent only a file-system basis test for this kind of security problem instead of part of a larger approach to the problem through systematic scanning.

Moreover, SaaS providers should consider DoS/DDoS weaknesses as security problems, not just customer relationship or business continuity problems.  These types of attacks can cripple a provider and draw their technical talent to the problem at hand, mitigating the denial of service attack.  During those periods, this can and has recently been used in high-profile fake-outs to either generate so much trash traffic that other attacks and penetrations are difficult to perceive or react to, or to create opportunities for social engineering attacks to succeed with less sophisticated personnel while the big-guns are trying to tackle the bigger attacks.  Until weaknesses that can allow for high-load to easily take down a SaaS application are included as part of vulnerability scanning, this will remain a serious hole in the testing methodology of a security scanning vendor.

So, seeing CVE identifiers and OWASP classifications for reported items is nice from a reporting perspective, and it gives a certain credence to mitigation reports to auditors, but don’t let those lull you into a false sense of security coverage.  Ask your vendor what other types of weaknesses and application vulnerabilities they test for outside of the prescribed standard vulnerability classifications.  Otherwise, you will potentially shield yourself from “script kiddies”, but leave yourself open to targeted attacks and advanced persistent threats that have created embarrassing situations for a number of large institutions in the past year.

What is your mobile strategy?

Native mobile applications are the hot-stuff right now.  Purists tout the HTML5-only route to mobile application development, but mobile web development alone isn’t enough to satisfy Apple to get access to the iOS platform, (since 2008) and consumers still can detect a web app that is merely a browser window and prefer the feature set that comes from native applications, including camera access, accelerometer data, and usage of the physical phone buttons into application navigation.  The native experience is still too nice to pass up to be at the head-of-the-class in your industry.

If you’re a serious player in the SaaS market, you have or will soon have a native mobile application or hybrid-native deliverable. If you’re like most other software development shops, mobile isn’t your forte, but you’ve probably hired specific talent with a mobile skill set to realize whatever your native strategy is.  Are your architects and in-house security professionals giving the same critical eye to native architecture, development, and code review as they are to your web offering?  If you’re honest, the answer is: probably not.

The reason your answer is ‘probably not’ is because it is a whole different technology stack, set of development languages, and testing methodology where the tools you invested in to secure your web application do not apply to your native application development.  This doesn’t mean your native applications are not vulnerable, it means they’re vulnerable in different ways that you don’t even know or are testing for yet.  This should be a wake-up call for enterprise software shops: because a vulnerability exists only on a native platform does not mitigate its seriousness.  It is trivial to spin up a mobile emulator to host a native application and use the power of a desktop or server to exploit that vulnerability on a scale that could cripple a business through disclosure or denial of service.

Your native mobile security scanning strategy should minimally cover two important surface areas:

1. Vulnerabilities in the way the application stores data on the device in memory and on any removable media

2. Vulnerabilities in the underlying API serving the native application

If you’re not considering these, then you probably have not selected a native application security scanning tool checking for these either.

In Conclusion

Security is always a moving target, as fluid as the adaptiveness of the techniques of attackers and the rapid pace of change in technologies they attack.  Don’t treat security scanning and penetration testing as a checklist item for RFP’s or to address auditor’s concerns — understand the surface areas, and understanding the failings of security vendors’ products.  Understand your assessments are valid only in the short-term, and re-evaluation of your vendor mix and their offerings on a continual basis is crucial.  Only then will you be informed and able to make the right decisions to be proactive, instead of reactive, regarding the sustainability of your business.

Leave a comment

Posted by on May 29, 2013 in Security


Thwarting SSL Inspection Proxies

A disturbing trend in corporate IT departments everywhere is the introduction of SSL inspection proxies.  This blog post explores some of the ethical concerns about such proxies and proposes a provider-side technology solution to allow clients to detect their presence and alert end-users.  If you’re well-versed in concepts about HTTPS, SSL/TLS, and PKI, please skip down to the section entitled ‘Proposal’.

For starters, e-commerce and many other uses of the public Internet are only possible because the capability for encryption of messages to exist.  The encryption of information across the World Wide Web is possible through a suite of cryptography technologies and practices known as Public Key Infrastructure (PKI).  Using PKI, servers can offer a “secure” variant of the HTTP protocol, abbreviated as HTTPS.  This variant itself encapsulates other application level protocols, like HTTP, using a transport-layer protocol called Secure Socket Layer (SSL), which as since been superseded by a similar, more secure version, Transport Layer Security (TLS).  Most users of the Internet are familiar with the symbolism common with such secure connections: when a user browses a webpage over HTTPS, usually some visual iconography (usually a padlock) as well as a stark change in the presentation of the page’s location (usually a green indicator) show the end-user that the page was transmitted over HTTPS.

SSL/TLS connections are protected in part by a server certificate stored on the web server.  Website operators purchase these server certificates from a small number of competing companies, called Certificate Authorities (CA’s), that can generate them.  The web browsers we all use are preconfigured to trust certificates that are “signed” by a CA.  The way certificates work in PKI allows certain certificates to sign, or vouch for, other certificates.  For example, when you visit, you see your connection is secure, and if you inspect the message, you can see the server certificate Facebook presents is trusted because it is signed by VeriSign, and VeriSign is a CA that your browser trusts to sign certificates.

So… what is an SSL Inspection Proxy?  Well, there is a long history of employers and other entities using technology to do surveillance of the networks they own.  Most workplace Internet Acceptable Use Policies state clearly that the use of the Internet using company-owned machine and company-paid bandwidth is permitted only for business use, and that the company reserves the right to enforce this policy by monitoring this use.  While employers can easily review and log all unencrypted that flows over their networks, that is any request for a webpage and the returned rendered output, the increasing prevalence of HTTPS as a default has frustrated employers in recent years.  Instead of being able to easily monitor the traffic that traverses their networks, they have had to resort to less-specific ways to infer usage of secure sites, such as DNS recording.

(For those unaware and curious, the domain-name system (DNS) allows client computers to resolve a URL’s name, such as, to its IP address,  DNS traffic is not encrypted, so a network operator can review the requests of any computers to translate these names to IP addresses to infer where they are going.  This is a poor way to survey user activity, however, because many applications and web browsers do something called “DNS pre-caching”, where they will look up name-to-number translations in advance to quickly service user requests, even if the user hasn’t visited the site before.  For instance, if I visited a page that had a link to, even if I never click the link, Google Chrome may look up that IP address translation just in case I ever do in order to look up the page faster.)

So, employers and other network operators are turning to technologies that are ethically questionable, such as Deep Packet Inspection (DPI), which looks into all the application traffic you send to determine what you might be doing, to down right unethical practices of using SSL Inspection Proxies.  Now, I concede I have an opinion here, that SSL Inspection Proxies are evil.  I justify that assertion because an SSL Inspection Proxy causes your web browser to lie to it’s end-user, giving them a false assertion of security.

What exactly are SSL Inspection Proxies?  SSL Inspection Proxies are servers setup to execute a Man-In-The-Middle (MITM) attack on a secure connection, on behalf of your ISP or corporate IT department snoops.  When such a proxy exists on your network, when you make a secure request for, the network redirects your request to the proxy.  The proxy then makes a request to for you, returns the results, and then does something very dirty — it creates a lie in the form of a bogus server certificate.  The proxy will create a false certificate for, sign it with a different CA it has in its software, and hand the response back.  This “lie” happens in two manners:

  1. The proxy presents itself as the server you request, instead of the actual server you requested.
  2. The proxy states the certificate handed back with the page response is a different one than what was actually handed back by that provider, in this case.

This interchange would look like this:

It sounds strange to phrase the activities of your own network as an “attack”, but this type of interaction is precisely that, and it is widely known in the network security industry as a MITM attack.  As you can see, a different certificate is handed back to the end-user’s browser than what in the above image.  Why?  Well, each server certificate that is presented with a response is used to encrypt that data.  Server certificates have what is called a “public key” that everyone knows which unique identifies the certificate, and they also have a “private key”, known only by the web server in this example.  A public key can be used to encrypt information, but only a private key can decrypt it.  Without an SSL Inspection Proxy, that is, what normally happens, when you make a request to, first sends back the public key of the server certificate for its server to your browser.  Your browser uses that public key to encrypt the request for a specific webpage as well as a ‘password’ of sorts, and sends that back to  Then, the server would use its private key to decrypt the request, process it, then use that ‘password’ (called a session key) to send back an encrypted response.  That doesn’t work so well for an inspection proxy, because this SSL/TLS interchange is designed to thwart any interloper from being able to intercept or see the data transmitted back and forth.

The reason an SSL Inspection Proxy sends a different certificate back is so it can see the request the end-user’s browser is making so it knows what to pass on to the actual server as it injects itself as a proxy to this interchange.  Otherwise, once the request came to the proxy, the proxy could not read it, because the proxy wouldn’t have’s private key.  So, instead, it generates a public/private key and makes it appear like it is’s server certificate so it can act on its behalf, and then uses the actual public key of the real server certificate to broker the request on.


The reason an SSL Inspection Proxy can even work is because it signs a fake certificate it creates on-the-fly using a CA certificate trusted by the end user’s browser.  This, sadly, could be a legitimate certificate (called a SubCA certificate), which would allow anyone who purchases a SubCA certificate to create any server certificate they wanted to, and it would appear valid to the end-user’s browser.  Why?  A SubCA certificate is like a regular server certificate, except it can also be used to sign OTHER certificates.  Any system that trusts the CA that created and signed the SubCA certificate would also trust any certificate the SubCA signs.  Because the SubCA certificate is signed by, let’s say, the Diginotar CA, and your web browser is preconfigured to trust that CA, your browser would accept a forged certificate for signed by the SubCA.  Thankfully, SubCA’s are frowned upon and increasingly difficult for any organization to obtain because they do present a real and present danger to the entire certificate-based security ecosystem.

However, as long as the MITM attacker (or, your corporate IT department, in the case of an SSL Inspection Proxy scenario) can coerce your browser to trust the CA used by the proxy, then the proxy can create all the false certificates it wants, sign it with the CA certificate they coerced your computer to trust, and most users would never notice the difference.  All the same visual elements of a secure connection — the green coloration, the padlock icon, and any other indicators made by the browser, would be present.  My proposal to thwart this:

Website operators should publish a hash of the public key of their server certificates (the certificate thumbprint) as a DNS record.  For DNS top-level domains (TLD’s) that are protected with DNSSEC, as long as this DNS record that contains the has for is cryptographically signed, the corporate IT department of local clients nor a network operator could forge a certificate without creating a verifiable breach that clients could check for and then warn to end users.  Of course, browsers would need to be updated to do this kind of verification in the form of a DNS lookup in conjunction with the TLS handshake, but provided their resolvers checked for an additional certificate thumbprint DNS record anyway, this would be a relatively trivial enhancement to make.

EDIT: (April 15, 2013): There is in fact an IETF working group now addressing this proposal, very close to my original proposal! Check out the work of the DNS-based Authentication of Named Entities (DANE) group here: — on February 25, they published a working draft of this proposed resolution as the new “TLSA” record.  Great minds think alike. 🙂


Posted by on September 15, 2012 in Ethical Concerns, Open Standards, Privacy, Security



CNN Lies to Every One of Its Web Viewers

When is it okay to flat out lie to your users?  I would argue: Never.  But the website of one of the world’s most watched sources of news, CNN, does just that.

Near the bottom of every article is a section called “We recommend” and “From around the web”.  These sections list about six links to other articles either on CNN itself, other Turner properties, or simply as a paid referral service for selected partners.  So what’s my beef with this?  It’s not the targeted marketing, it’s the outright lie I noticed they make when you hover over any of those links with your mouse.

For some background, I’m a huge dissident against outbound link tracking.  It’s fundamentally the same as gluing a GPS tracking device to your forehead and giving a a tracking device to the website you’re visiting.  I have a problem with it because I think there is a fundamental freedom that is eroded by this technology – the freedom to consume information without being tracked for doing so.  Do I have the right to pick up a magazine and browse through it without giving someone my telephone number?  I would say yes — I think it is a natural right to be able to consume information without having your consumption observed.

But my belief here isn’t realistic — tracking basic visitor behavior and consumer preferences is the basic monetization and sustainability model for most of the Web as we know it.  So, this world doesn’t mesh with my perfect world, but at least I should know if someone is observing my behavior, right?  Observing CNN’s privacy policy one can clearly see the word “link” is referenced twice, once in relation to third-party sites that may cookie you, and once for integration to social media or other partner sites that may have differing privacy policies.

Okay, fair enough, therefore I should expect that if I am surfing just CNN’s website, if I disable cookies, and if I turn on my do not track header, I should expect not to be tracked, right?  No, and the reason is I cannot find out when I’m still on the CNN site to only stay within it.  The reason is CNN has specifically coded it’s site to lie to me about when I’m staying within it or navigating away.  For an example, if I were to hover over one example link in these two sections, I see the following in my browser status bar:

I right-clicked the link in Chrome and copied the URL.  Then curiously I noticed the link read differently in the browser status bar when hovering over it, this time reading:

Youch, what’s that, and why did it change?  On closer inspection, by viewing the source of the page, I can see the target href of the link is exactly as reproduced above, going to  I peeked at some other URL’s in the same section that I had not yet left-clicked or right-clicked and noticed this:

<a target=”_self” href=”; onmousedown=”this.href=’;rdid=349349184&amp;type=WMV_d/t1_ch&amp;in-site=false&amp;req_id=968ab83e0a0f44e584d8744520d2aea0&amp;agent=blog_JS_rec&amp;recMode=4&amp;reqType=1&amp;wid=100&amp;imgType=0&amp;refPub=0&amp;prs=true&amp;scp=false&amp;version=59070&amp;idx=4&#8242;;return true;” onclick=”javascript:return(true)”>Knicks’ Jason Kidd arrested on suspicion of DWI</a>

And herein is the deception — this piece of inline JavaScript code changes the target of the link at the moment it is clicked to go to the address.  Because target href originally reads to the final destination of the article, hovering over it gives the false impression that my click will directly take me to it.  Instead, at the moment I click it, the target href is changed to the potentially unscrupulous third-party, and I have been given no browser notification this would happen prior to my click, and upon responding, it redirects me back to the original CNN article I initially wanted to view.  On a broadband connection, you probably wouldn’t even notice the superfluous page load and redirect back to CNN’s site.  Deceptive!

So, sure, why should anyone care?  Isn’t this just plumbing, technology, and toolbox of tricks inherit of the Web?  Maybe, but the problem here is the lie.  You do not lie to your users.  Ever.  Outbound web tracking is not a web beacon.  Web beacons are a different kind of “evil” – usually some JavaScript that opens an IFRAME to a third-party site that issues a cookie to track you; however, web beacons are covered by CNN’s privacy policy, so if they were equivalent, it’s all fair.  Web beacons can be simply disabled by turning off third-party cookies in today’s browsers.  This is precisely why outbound link tracking is becoming popular – it circumvents the privacy management tools most users have available and have knowledge of.  Outbound link tracking is no more insidious than web beacons are, but the implementation of them often lies to the end user about what their action will do (a click in this case).  An honest implementation would be to either clearly state in the privacy policy that any links you click may be link tracked or simply not to deceive the user by rewriting the target href the moment they click it to actually go to the link tracking site so the browser status bar is truthful on hover (Twitter’s strategy).

Well, at least it’s just CNN at fault here.  At least no one else would stoop to such shady tactics.  Surely not Google (/url) or Facebook (l.php).. no, definitely not…