Training, outreach, and networking
We decided to leverage existing communications avenues (mailing lists, newsletters, status reports, etc) as well as setup a Wordpress blog. We used these tools to publish information about security news, link to 3rd party security documentation, security guidance, solicit feedback and most importantly, identify individuals throughout the organization that had interest or skills in secure software development. Our goal with the outreach program was not only to make information resources available, but also to ensure that our services were transparent and accessible. Like us, most security groups have to overcome the stigma of being a crazy bunch of paranoid hackers who cost the company money and cause deadlines to slip. As such, the outreach program coupled with our threat modeling and security consulting services were delivered with clarity, transparency, and comprehensiveness.

We also delivered numerous training courses aimed at educating developers and architects in defensive programming, software vulnerabilities, and threat modeling. These courses were typically delivered to smaller audiences and accompanied with hands on activities. We found that these courses were not only being well received, but also that attendees would contact us to request additional training tailored to a specific topic that would be relevant to their products. I feel that we had the distinct benefit of having team members who were very adept at delivering training and realize that not all organizations have this sort of resources. Given a high enough priority in company goals, training can easily be purchased, and employees who are members of professional groups can leverage relationships with professionals outside the organization who would be willing to deliver the training.

For more information on homegrown security teams, checkout my post.

Metrics and defining success
I feel that it would be easier to give you a view of the ultimate goal behind the application security metrics rather than explain how it evolved over time. The ultimate goal was for us to be able to:

• Track vulnerability metrics and relevant meta data across all security assurance services.
• Derive relevant data from vulnerability metrics data to aid in assigning the appropriate resources to address vulnerability trends.
• Demonstrate the effectiveness, reach, and customer feedback of each security assurance service.
• Provide a snap shot view of overall risk to the enterprise based on known application vulnerabilities and demonstrate trends over time.

The rational behind these goals was so that we could demonstrate the effectiveness of our services (and therefore our work), provide reliable vulnerability management data, and give us the data points needed to effectively make use of the resources we’d been allocated. I should note that we never really had an accurate picture of just how many applications were out there, so we had difficulties determining service coverage, so instead we decided to nail a stake in time as the starting point for our date based metrics.

First, we had to build a system that would collect and report on the vulnerability data points we’d identified. I’ve blogged about the system here and here where you can read the details, but to summarize, we wanted to classify the vulnerability, define risk, provide detailed description and any proof of concept data, define the product that is vulnerable, define the development team that created the vulnerable code, define contact information, and track resolution status. With this system, not only did we have each vulnerability we’d identified documented and meta-tagged, but we also had the data that allowed us to determine which development groups had the worst vulnerability rates (most vulnerabilities per product) and should be targeted for training. We also used the vulnerability data and meta-tags to provide executive level views to answer the “how many?” “what type?” “who’s doing it?” and “how bad?” questions from executives.

Findings from the penetration testing, threat modeling, and security consulting services fit very nicely into the vulnerability management system. A simple customer feedback form could have been used to rate the effectiveness of the service delivery and the derived metrics could have been used to refine our processes, we just never got that far in the project. Of course, there were several other activities which didn’t get accounted for such as training and outreach and communications who’s demand grew organically throughout the year and who’s reach in the company could have been measured over time to (hopefully) demonstrate a downward trend in vulnerability metrics for the trained development teams.

I think that the metrics plan was strong and would provide us the data we needed, I just wish we’d started implementing the full plan sooner.

The first 2 chapters of the book deal with installing, configuring, and rule writing basics for ModSecurity 2.5 with Apache 2.x. Magnus Mischel walks the reader through installing and configuring the module in Apache and making sure the installation is working properly. The chapter assumes best case scenarios for defaults in the operating system while in my experience, ModSecurity installs rarely go that easily. That being said, the preface of the book specifies that a good understanding of system administration principles is suggested to get the most out of the book and distribution specific installation guides for ModSecurity are widely available online. Mischel then goes on to introduce the ModSecurity rule writing syntax which he very astutely presents in two sections: rule grammar and structure followed by practical examples of how to write rules to mitigate vulnerabilities. Even after spending 1.5 years running numerous production instances of ModSecurity with custom rules, Mischel’s in depth coverage of the topic still had me hunting around for my highlighter to mark sections of the chapter containing syntax features that I didn’t know existed.

The book also does a great job of covering ModSecurity performance and logging. Mischel uses httpperf to benchmark server response time, memory usage, and processor load and presents the results for Apache with/without ModSecurity and with/without the default ruleset. Readers will find that this data will be invaluable if they ever need to sell the security benefit vs. system performance degradation to system administrators and executives. The chapter on logging and auditing offers an in depth look at the features available in ModSecurity offering insight on real world log and audit needs balanced with the processing overhead associated with excessive data collection. Mischel also provides a detailed implementation guide for deploying the ModSecurity Console with mlogc giving administrators and analysts a dashboard for monitoring alerts in a environment using multiple instances of ModSecurity.

One of the chapters I was most looking forward to reading was “Blocking Common Attacks” where Mischel presents a number of common web application vulnerabilities, offers an overview of the causes and impacts of said vulnerabilities and presents a virtual patch using a series of ModSecurity rules. To be perfectly honest, I was left a little disappointed with lack of depth in some of the content. While I understand that in depth coverage of web application vulnerabilities is probably beyond the scope of the book, the additional knowledge required to fully understand the causes and exploits of the issues that are presented in this book is likely beyond the average skill set of a system administrator. I was also left a little disappointed with the section on cross site request forgeries (CSRF) which never offered any virtual patching examples to mitigate the issue. Negatives aside, Mischel does a great job at presenting ModSecurity rules for mitigating 14 types of web application vulnerabilities such as cross site scripting (XSS), sql injection (SQLi), and shell code execution.

Mischel spends an entire chapter discussing configuring ModSecurity using a positive security model tailored to a YABB (Yet Another Bulletin Board) deployment. For a veteran ModSecurity user, this was by far the best chapter in the book, even though Mischel suggest you use Ethereal (Wireshark) as your local web proxy (you should be using Burp Suite by Portswigger). I can only hope that readers who are rolling out new ModSecurity installations or who are looking to improve existing deployments read this chapter and realize the value of this approach and choose to do the same to protect their applications.

By including the full ModSecurity directive and variable reference and a detailed guide on Regular Expressions, Magnus Mischel has created a complete guide to ModSecurity that is not only a great introduction to the WAF technology, but also a great desk reference for the experienced administrator.

You can find this book on from the publisher PACKT Publishing or Amazon.com.

Title: ModSecurity 2.5
Author: Magnus Mischel
Publisher: PACKT Publishing
Front page taglines:
* Securing your Apache installation and web applications
* Prevent web application hacking with this easy-to-use guide

Threat modeling, architecture reviews, and security consulting
Our application security program always offered a design review service. Initially this service wasn’t very well defined and simply involved a security engineer performing a review of available documentation and interviews with the development team. The outcome of this engagement was a series of recommendation and requirements for the project to implement during the development phase. As you can imagine, this approach was very subjective and lacked documentation, but formalization of the offering was put on the back burner due to the successes with the penetration testing offering.

A little over a year and a half after the application security program was defined, we started focusing on assimilating some of the industry proven approaches to threat modeling and security architecture best practices into our existing design review service. We leveraged the opportunities presented by the local OWASP chapter and were able to obtain the training in formal threat modeling approaches that we were easily able to adapt and offer as a fully documented and repeatable process. Unfortunately, my time with the team was up before we could really determine the effectiveness of this service offering. My hope was that we could use this service to not only identify deficiencies in secure design early in the lifecycle, but also to setup a plan for providing training, code review / SCA services, and penetration testing at the appropriate times. This would effectively setup a consulting relationship between the individual security engineers and the project team where a single person / team could be leveraged for all aspects of application security.

Penetration Testing
Like most application security programs, ours was started by 2 of the team members who had some understanding of the vulnerability space in applications (mostly web based) and knew how to test for them. In a company with a seemingly unlimited number of custom web applications the target space was wide and varied which provided a vast playground for exploration and learning. Over time, the value of our hacking away at these web properties coupled with a few juicy findings quickly became apparent to the leadership and we were asked to work on expanding this capability to the rest of the team while maintaining the high level of thoroughness, accuracy, and professionalism our internal customers had grown accustomed to.

We were faced with the unique challenge of defining and documenting the vulnerability space, provide documented examples of how to identify and exploit the vulnerabilities, provide in person training to explore the complexities of application architectures and the interconnectedness of various components, and provide the development background necessary to work with developers to resolve the issues. Fortunately for us, the Open Web Application Security Project (OWASP) had already published some early versions of the OWASP Testing Guide and the OWASP Development Guide. These two documents provided the written documentation necessary to begin familiarizing the team with the general concepts and jargon associated with web application security. With the help of the OWASP WebGoat tool, we were able to provide a platform on which the engineers to practice what they were learning in the documents. We also conducted some additional internal training which took and in depth look at the vulnerabilities which were most common across our specific platforms and explored the best ways to address these issues in code. After a few years of regular training and refinement of the methodology has yielded a very efficient, accurate, and repeatable security testing service. If I had to go back and do it again, I really wouldn’t change anything. The resources obtained through the OWASP organization served us well and the existing internal knowledge was sufficient to bring the rest of the team up to speed.

I also wanted to address the issue of automated scanning. We did indeed purchase a solution which we worked tirelessly to implement and customize for our needs only to find that our manual approach with a couple open source tools was performing much better than the automated solution. This was partially due to the application we were testing, we had some particularly unique extremes in size of the application to be tested, and the automated tools simply didn’t complete scans in an amount of time that we found acceptable. If I had to do it again, I would explore some more varied options for automated testing that could be better fit for the environment. I realize the importance of an automated solution to validate applications which do have strong security requirements, but that do require some level of security assurance.

Dedicated application security team
When I was first hired on to the team, it was responsible for both security assurance (application security) and incident response. One of the first milestones which lead to the success of the program was the separation of the team into 2 groups with distinctly different areas of responsibility: one team did security assurance, the other did incident response. I credit this move as a success because when the responsibilities were shared, engineers were required to share their focus between the proactive mindset of security assurance, and the reactive mindset of incident response. Inevitably, certain engineers were more skilled in one or the other of the disciplines and the quality of service across the different offerings would be degraded at times. While the teams now had a different and separate structures and managers, the importance of information sharing wasn’t lost as the teams were already so used to collaborating as a single team.

If your company is serious about application security and the workload justifies it, I would suggest that a dedicated application security team be created to provide the expertise needed to address the unique challenges of a complete application security problem space. The charter and team responsibilities must be carefully selected, if the mission is too narrow certain critical of the supporting components in application architectures might be out of scope, while a mission that is too broad will reduce the amount of time engineers will be able to dedicate to appsec will be limited. If I had to do it over again, I would certainly narrow the scope of the team to only include application security and associated disciplines (training, threat modeling, secure design and architecture consulting, security assessments, code reviews and analysis – SCA).

I’m starting the next step in my career on 1/4/2010 with one of the local security consulting firms and am very much looking forward to the challenges and opportunities which the new job presents.

Happy holidays to all and may the new year bring you all the love, good health, and happiness you deserve!

“I understand the risk, potential exposure, and impact of this security issue. With my signature, I indicate that I am making the decision to take responsibility for any adverse impact directly relating to the exploitation of the security issue.”

Obviously, this does nothing to improve the posture of the system or application in question…but the real question I’m left with is: what happens to that executive or government official when the system DOES get hacked through a vulnerability with “accepted risk”? Do they get fired? Demoted? Ordered to pay restitution losses?

Sadly, usually nothing happens. The individual continues to collect a pay check, manage employees, and accept risk for the security vulnerabilities that affect systems under their purview.

I ask you, where has accountability gone?

First – if you’re going to host a conference in DC, there’s really no better venue than the DC Convention Center. This is really a top notch venue in the center of DC that is built for conferences. It’s metro accessible and the conference services (food, beverages, A/V, wireless, etc) are top notch. I’m not saying the venue makes or breaks the con, but it helps.

The speakers, technical content of the presentations, and variety of topics exceeded that of much more expensive conferences I’ve attended. Joe Jarzombek kicked things off with the keynote, David Byrne and Charles Henderson can’t filter the stupid, Jon Rose and Tom Leavey brought the drinking game with a chance of 0-day, Jeff Williams tackled the insider threat, Kevin Johnson and Tom Eston think our friends want to eat our brains, Josh Abraham brought synergy to our pen-testing tools, RSnake gave security experts happy hour chatter for the next year with the 10 least-likely and most dangerous people on the web, John Steven condensed 6 hrs of Threat Modeling training into a 45 minute talk (good thing we had him scheduled at the end of the day), and Chris Weber dazzled with unicode. Not to be outdone, the OWASP projects were equally represented with Pravir Chandra on OpenSAMM, Jeff Williams on ESAPI followed by Arshan Dabirsiaghi on the ESAPI WAF, Sebastien Deleersnyder and Fabio Cerullo brought the OWASP tools together to deploy web applications, Matt Tesauro did his thing with the Live CD, Dr. Boaz Gelbord touched on security spending, and of course, who could forget Dave Wichers at the OWASP Top 10 2010 RC1! The conference also features 2 panels, the Federal CISO panel with Ray Letteer (USMC), Timothy Ruland (US Census), Richard Smith (TSA), and lead by Matt Fisher. The SDLC Panel features Michael Craigue (Dell), Dan Cornell (Denim), Dennis Hurst (HP), Joey Peloquin (FishNet), David Rook (Realex), Keith Turpin (Boeing), and lead by Pravir Chandra. The conference also featured a CTF running the new OWASP CTF project and hosted by Martin Knobloch.

For those of us who have been in the security industry for a few years, these conferences are a great chance to catch up with old friends and make new acquaintances. It was great to see familiar faces like Tom Brennan, @grecs, Ken Van Wyk, Matt Fisher, Dinis Cruz, Jon Rose, John Steven, Lee Anne Hart, Gracie Daniel, Jon McCarty, Jeremy Long, Rob Fuller, Jack Mannino, Rex Booth, Mark Bristow, Doug Wilson and others. At the same time, it was great to make new contacts like Josh Feinblum, Pravir Chandra, Robert Hansen, Matt Tesauro, Arshan Dabirsiaghi, Rafal Los, Jeff Williams, and hell, even the great Dan Kaminski made an appearance!

Just like any good conference, the awards and closing remarks held on the last day were full of thanks, toys, flying vendor squishy balls, foam rockets (courtesy of Tom Brennan), cheers, and clapping. It was truely a great way to wrap up a top notch con!

Finally, and although it’s been done many times already, I want to take a second to recognize all those OWASPers and DCers that came together to make this event what it was. I’m really copying this list verbatim from the last page of the conference booklet:

• Rex Booth, Mark Bristow, Doug Wilson, and Kate Hartmann who provided the leadership without which this conference wouldn’t have come together.
• The OWASP Board – Jeff Williams, Dinis Cruz, Dave Wichers, Tom Brennan, and Sebastien Deleersnyder who gave us “carte blanche” and trusted us to get this conference done.
• The lead volunteers Barry Austin, Angel Contreras, Josh Feinblum, Lee Anne Hart, Martin Knobloch, Jeremy Long, Jon Rose, David Sachdev, Mike Smith, and myself.
• The red shirt people of which there are way too many to name…THANK YOU!
• And all those who spoke at or attended the conference!

So if you get the chance to attend a future OWASP event, or if you haven’t checked out your local chapter, hopefully this blog post and the others I mentioned in the first paragraph will shed the spotlight on the OWASP organization and how WE work to improve application security worldwide.

• Password standard – to make sure that we don’t simply leave the keys to the front door under the welcome mat.
• Acceptable use standard – to make sure that employees aren’t playing WoW all day or hosting warez on the company servers.
• (*)^11 standard – if you want to be PCI/SOX/HIPAA/* compliant.

Standards are meant to define what is acceptable or prohibited in accordance with company rules. Beyond achieving industry/legal compliance and when done right, these standards not only help employees determine right from wrong, they allow the company to operate knowing that everyone is on the same page.

Unfortunately, most organizations either overlook product/software development standards (or coding standards) or do such a poor job at defining the expectations that the development staff are left to code according to what they know best, or what they’ve done at previous employers. For seasoned and experienced developers, this might not seem like a terrible thing, so I’m going to illustrate with an example.

Joe is developing a login user interface for the next big product at the XCompany. This interface will be displayed to users visiting the application using a web browser. Being a seasoned developer, Joe implements the password portion of the login form using the HTML attribute type=”password”.

Mark is developing the iPhone application which will enable mobile users to connect to the same application that Joe is working on. Being an iPhone user himself, Mark thinks that the way the iPhone OS handles password masking is a pain and often results in failed login attempts. Mark then talks with the product manager and together they decide that instead of masking the password according to the iPhone OS standards, they’re going to display the full text of the password field until the user leaves the password field or after 3 seconds have passed without new input from the keyboard.

Jake is developing the Android application for the same product as Mark and Joe. Because Jake is paranoid and a hacker by night, he decides that the Android standard for handling passwords is too insecure and decides to totally mask the password field. Note: Android default behavior is to display the last letter typed in the password field for 2 seconds before masking.

While Mark’s approach might be acceptable and justified from a usability standpoint, and Jake’s approach might be acceptable from a security standpoint, we now have an issue where products from the same company are different in the way they implement the same functionality.

To make matters worse, without software development standards junior developers with little prior experience might end up introducing some major security vulnerabilities into the corporate applications. Lets explore another example.

Joe, Mark, and Jake are all senior developers with years of experience under their belts. In the past, their applications have been torn apart by hackers leveraging input based security flaws. As such, they’ve learned that all input must be validated prior to being consumed by the application.

Martin is the latest hire on the development team. He’s right out of college and this is his first real software development job. Eager to please, he takes on the responsibility of implementing the Search feature in the application. Bill, the project manager, agrees since the task is rather simple, all Martin will have to do is invoke the well documented search appliance API which then searches the back-end database for the results and display them back to the user. Martin quickly implements the search API and not knowing any better, doesn’t escape the user provided inputs before sending them off to the search API – now we’ve got an SQL injection vulnerability introduced into the system. (security folks, humor me, assume the API doesn’t escape input either).

While the first example seems rather benign, end users might interpret these differences as a failure to do proper quality control – which might be disastrous to the brand image if we’re talking about a major financial application. The second example leaves no room for interpretation, the lack of guidance for a new developer to lean on has introduced an avenue for attackers to steal your customers’ information or adversely impact the application by modifying or deleting critical information from the database.

If you don’t have a software development/coding standard in your organization, or if you feel that the one you have now is inadequate, here are some resources to get you moving in the right direction:

• CERT Secure Coding Standards / More here too
• Carnegie Mellon Software Engineering Institute and Team Software Process for Secure Systems Development presentation
• OWASP Secure Coding Principles