Updated JIRA RISK workflow (now with a 'Fixing' State)

As an improvement of the workflow I showed at JIRA Workflows for handing AppSec RISKS here is a version that adds a 'Fixing' state between 'Allocated for Fix' and ‘Test Fix’.

The reason for this change, was to take into account projects (or components) that have a large number of open issues that want to be fixed (vs risks to be accepted).

Since we try to use an Kanban 'Work in Progress' model for the issues to fix (i.e. no more than 3 to 4 active items), this new state helps to keep a nice separation between the issues that:

• need to be 'Risk Accepted' (i.e. there is no intention (or resources) to fix in the next couple months)
• have been reviewed and are 'Allocated for Fix'
• are currently being worked on (i.e. in a 'Fixing' state)

Presenting at OWASP AppSecEU on "Using JIRA to manage Risks and Security Champions activities"

I just received this nice invitation from the OWASP AppSec EU today:

My plan is to use this opportunity to document the JIRA workflows that I have been creating and implementing (when acting as Head of Application Security)

Here are a couple related posts:

• JIRA Workflows for handing AppSec RISKS 
• What are Security Champions and what do they do? 
• Does your team has a Security Champion? If not, get this Mug and Library 
• "New Era of Software with modern Application Security" updated presentation (v.0.6) 

Thinking of writing a book called "Measuring Software Quality using Application Security"

This book will be based on the ideas I've been talking about in my "New Era of Software with modern Application Security" presentation.

The plan is to use my experience with Leanpub (where I have published 7 books), with the content being hosted on GitHub and published early and ofter.

Is Quality is a measure of how successful a product is in what it is SUPPOSED to do?

Here is a question I received on the concept "Application Security can be used to define and measure Quality" (slides here) 

Quality is a measure of how successful a product is in what it is SUPPOSED to do.

AppSec is a measure of how many and what things product does that it is NOT SUPPOSED to.

These two are not related. A startup may have a good quality product full of security holes, and a bank may have a highly secure product that is also of great quality.

I think that measuring Quality by only looking at the success rate of a product is a very narrow definition of Quality.

Video for my LSCC presentation on: New Era of Software with modern Application Security

Skillsmatter has just published the video of the presentation I delivered last week at the LSSC (London Software Craftsmanship Community)

You can see it here:
https://skillsmatter.com/skillscasts/7582-new-era-of-software-with-modern-application-security

"New Era of Software with modern Application Security" updated presentation (v.0.6)

Here is the updated version of the talk I delivered last week at the LSCC (this time around delivered at the OWASP London Chapter)

There are a number of new slides, but it is still far from complete :)

Please take a look at the slides and let me know what you think of them? (and what can be improved for the next version)

I'm delivering an Application Security Training in London on 3rd and 4th of March

This is an 100% customised course (to the participants) with as many practical examples as possible.

Here is the course description:

V0.5 of "New Era of Software with modern Application Security" presentation

Here is my first pass at creating the "New Era of Software with modern Application Security" presentation, which I will deliver as a Keynote at the Codemotion Rome developer conference (March 19th)

This is the version that I presented yesterday at the London Software Craftsmanship Community event and its video is here

Interestingly, one of the concepts that I arrived at (when working on the slides) was that Application Security can be used to define and measure Quality.

This is something that I have been thinking about for a long time, and I'm starting to find a way to explain how I'm able to use Application Security to help developers to create better applications (with not only better security, but with better quality)

Please take a look at the slides and let me know what you think of them? (and what can be improved for the next version)

Published update to my Practical Eclipse book

You can get the latest version from https://leanpub.com/Practical_Eclipse for FREE by choosing the $0 minimum price.

Here is the email I send to my readers:

Speaking at LSCC (18th Feb) on "New Era of Software with modern Application Security"

In preparation to my CodeMotion keynote in March, next week I'm presenting a first version of it at LSCC (London Software Craftsmanship Community) which is also a developed focused audience.

You can register at https://skillsmatter.com/meetups/7845-lscc-talks-feb-2016

Here are the talk details:

Title: New Era of Software with modern Application Security

Description: This presentation will start with an overview of the current state of Application Insecurity (with practical examples). This will make the attendees think twice about what is about to happen to their applications. The solution is to leverage a new generation of application security thinking such as: TDD, Docker, Test Automation, Static Analysis, cleaver Fuzzing, JIRA Risk workflows, Kanban, micro web services visualization, and ELK. These practices will not only make applications/software more secure/resilient, but it allow them to be developed in a much more efficient, cheaper and productive way.

Bio:Dinis is focused on creating Application Security teams and providing Application Security assurance across the SDL (from development, to operations, to business processes, to board-level decisions). His focus is in the alignment of the business’s risk appetite with the reality created by Applications developed internally, outsourced or purchased. He is also an active Developer and Application Security Engineer focused on how to develop secure applications. A key drive is on 'Automating Application Security Knowledge and Workflows' which is the main concept behind the OWASP O2 Platform.

Is Google a geopolitical threat to the UK? (i.e. what would happen if it pulled the plug on UK's traffic)

During one of the recent Application Security training courses I delivered recently, one interesting example I gave during a section on "Our dependencies on Technologies and Frameworks that we don't fully understand" was the concept of how much of a threat to the UK economy is Google?

For example if Twitter or Facebook were not available from the UK, I don't think the impact would be significant.

But if Google and all its services (search, mail, calendar, maps, geolocation, docs, spreadsheets,  contacts, Google ID) was suddenly not available, I bet that there would be a significant disruption to a LOT of individuals, business and government agencies.

There is a lot of talk in the UK about the Geopolitical threat of Russia (and its control on natural resources used by the UK), but I'm pretty sure Google can do more damage.

Of course that it would be economical/business suicidal for Google to do such a thing, but that doesn't make it less real or dangerous.

Speaking at Codemotion Rome on "New Era of Software with modern Application Security"

Next march I'm going to be delivering the "New Era of Software with modern Application Security" keynote at Rome's Codemotion (17-19 March),

This is very exciting, since Codemotion is a developer focused conference, which is exactly the audience that we (AppSec) need to be talking to (and learning from).

The speaker line up is also pretty impressive (see more details here), so if you are around, this is a good conference to go to this year.

I still have quite a bit of work to do on my presentation and slides, but the key idea is to cover how a new generation of application security thinking (using TDD, Docker, Test Automation, Static Analysis, Fuzzing, JIRA Risk workflows, Kanban, micro web services visualization, ELK) not only makes apps more secure/resilient but it allows them to be developed in a much more efficient and productive way

Job post on "Application Security Manager" for The Hut Group (in Northwich, UK)

Here is a real cool opportunity to work for a company that is focused on Application Security and developing innovative solutions to embed Application Security into the SDL (disclamer: I'm currently contracting for them as interim 'Head of Application Security')

You can see full details at https://www.linkedin.com/jobs2/view/102336625 and here is the main description

We are looking for an individual to who can take a hands-on approach to build and run an industry leading application security team. The Application Security Manager will develop, implement and run a secure application development program, with supporting standards and processes, and formal methodologies where relevant. 

Securing our applications and customer data is critical to the success of our business. The Application Security Manager will be a security evangelist who can translate security concepts to technical and non-technical audiences, and will approach application security from the perspective of business risk. This person will be the leading authority for Application Security within the group.

In addition to being an AppSec expert, the key for this role is to have significant development experience/knowledge.

A large part of the work is in supporting the existing network of Security Champions and working with devs/architects on figuring out how to secure the wide variety of apps they are developing (see here and here for more details on what these Security Champions do)

You can apply for the job at that LinkedIn page, and let them know that you saw this on my blog :)

First-Party-Only Cookies - nice solution to mitigate CSRF

Just saw https://tools.ietf.org/html/draft-west-first-party-cookies-01 which proposes

   This document updates RFC6265 by defining a "First-Party-Only"
   attribute which allows servers to assert that a cookie ought to be
   sent only in a "first-party" context.  This assertion allows user
   agents to mitigate the risk of cross-site request forgery attacks,
   and other related paths to cross-origin information leakage.

It looks really good, and it seems that Chrome 50 is going to support it https://www.chromestatus.com/features/4672634709082112

The current solution seems to be inspired by the SameDomain Cookie attribute as described at http://people.mozilla.org/~mgoodwin/SameDomain/samedomain-latest.txt

I actually prefer the SameDomain name to First-Party-Cookies :)

Reverse engineering recently patched Wordpress

On the topic of the recent Wordpress update (see https://wordpress.org/news/2016/02/wordpress-4-4-2-security-and-maintenance-release) I was asked an interesting question which was ‘how to test/exploit’ the patched vulnerabilities? (namely the SSRF one)

Since this seems to be an scenario where Wordpress has not released the details, one way to do it is to diff the current release with the previous one

Depending on the technology, this can be really hard (C++ patches requiring IDA Pro foo) or doable (.NET, Java, PHP)

Assuming that Wordpress is not distributed in compiled PHP (http://stackoverflow.com/questions/1408417/can-you-compile-php-code) this could be as simple as doing a file diff (it will depend on how many changes where made in the current release)

And how to perform this diff?

Use Git :)

Just:

1. install previous version
2. commit all files
3. install upgrade (which in Wordpress can be done via the web interface)
4. review changed files (it might be useful to commit files that clearly are not related to the issue)

Come on Amazon, its time for 100% TLS (aka https)

On a thread about moving a site to 100% TLS (ie. SSL), which btw, is the right thing to do in 2016 if one wants to protect users from Man-in-the-middle attacks, I was asked this question:

I notice Amazon is not secure until you authenticate, then all pages become secure. This is an interesting approach. What do you think Dinis?

This really sucks!

Lots of eCommerce companies look at Amazon as the benchmark on what to do (and what risks to accept), so the fact that they don't support 100% TLS (as can see by googling amazon) is not helpful at all.

Here was my reply:

Well shame on Amazon for not also not doing 100% SSL 

That said, amazon has an amazing application security team (with https://firebounty.com/bug-bounty-program/16/amazon) and they have quite a lot of visibility into what is going on in their platform (namely on fraud and account hijack/abuses) 

Also, Amazon is getting there, for example note how if you start your amazon journey on https:// (in most cases) you still stay in SSL if you do some actions and go to checkout

Yes there are users that don't support TLS and in some cases there are a couple performance tweaks that will need to be done. But we shouldn't be downgrading the security of 99% of users due to a couple user's locations or browsers.

The ones to follow on this topic are ETSY (see https://codeascraft.com/2012/10/09/scaling-user-security) who did this change in Oct 2012

Published update to my Practical Git and GitHub book

You can get the latest version from https://leanpub.com/Practical_Git for FREE (by choosing the $0 minimum price).

Here is the email I send to my readers:

Hi, thanks for being reader of my Practical Git and GitHub book. I just released a new version which contains a large number of content and images fixes.

This version is very similar to the previous release, but I'm planning to make big changes in the next months.

I'm going to add a number of new chapters and remove content that might be better in a different book.

I really would like your feedback, so please don't hesitate to contact me at dinis.cruz@owasp.org or directly at the GitHub repo that contains all content and current Issues: https://github.com/DinisCruz/Book_Practical_Git/issues

Published update to my Practical AngularJS book

You can get the latest version from https://leanpub.com/Practical_AngularJS  for FREE by choosing the $0 minimum price.

Here is the email I send to my readers:

Published update to my Practical Jni4Net book

You can get the latest version from https://leanpub.com/Practical_Jni4Net

Here is the email I send to my readers:

JIRA Workflows for handing AppSec RISKS

Recently I have been acting as 'head of Application Security' for a couple UK companies, and one of my most effective actions has been to setup the JIRA workflow that you can see below.

The key to this workflow (and the secret of its success) is the action to get the business owners to click on the 'Accept Risk' button. 

That simple action makes the whole difference, since that is the moment that a particular RISK become REAL.

Now, the responsibility/decision/liability of NOT fixing an issue, is clearly mapped to an individual (which in some cases can even be the CTO).

Note that the definition of 'not fixing' should be 'will not be fixed in the next couple weeks'

Paying OWASP Leaders and some ideas on how OWASP should be supporting its projects

(based on an email to the owasp-leaders list)

The reasons why I believe OWASP  should not be allowed to pay owasp leaders are listed here http://blog.diniscruz.com/2012/04/why-owasp-cant-pay-owasp-leaders.html

And since I have not been on the OWASP board for about 5 years, I think we need to realise that IF it was possible to pay owasp leaders to work on OWASP projects, THAT (paying owasp leaders) would have happened by now (after all, there has been enough budget to make that happen)

The problem is that there is still this 'idea' that "IF ONLY we could do that (pay owasp leaders) amazing stuff would happen". 

Request for OWASP board to approve 100K for a project Summit in 2016

(sent to the OWASP leaders list in early Dec 2015, following the original request made in June 2015)

Bumping this thread, since I believe not much has happened since.

I would like to request again for "OWASP board to approve 100K for a project Summit in 2016. And then ask for a team or OWASP leaders to lead that effort"

Proposed new strategy for OWASP projects - They are Research Projects

(variation of an email send to the owasp leaders list)

I think a key problem is the expectation that OWASP should ever be able to develop professional, best in class and 'secure' apps.

These conversations always tend to have a base on the idea that OWASP 'should not have a lot of projects' and 'only focus on a couple high-value/high-quality ones'. This never gains traction because that goes completely the model and culture of OWASP projects.

The reality is that really good a solid projects at OWASP are the exception and the outliers.

Do you deserialize Java objects? Jenkins zero day and vunls in WebLogic, WebSphere, JBoss, OpenNMS and Appache commons

Last week the What Do WebLogic, WebSphere, JBoss, Jenkins, OpenNMS, and Your Application Have in Common? vulnerability research was published and it included a number of quite worrying exploits on Java apps, more specifically on apps that use the Apache commons library (update: it looks like this is not a vuln in Apache commons, but in how it is used).

This is following up the Java Deserialization research published earlier this year on Marshalling Pickles and Exploiting Deserialization Vulnerabilities in Java (which is a variation of the XStream/XMLDecoder vulns/research I was involved in 2013)

It also looks like the Jenkins issue mentioned in the latest research doc is a zero-day on Jenkins: Mitigating unauthenticated remote code execution 0-day in Jenkins CLI

Since this is a vulnerability that allows RCE (Remote Code Execution), it is really important to understand the internal/external exposure to java deserialization, Jenkins and apache commons usage.

Four amazing years and good luck TM 4.0

After working very hard on multiple versions of TeamMentor (TM 2.0, 3.0, 3.5 and 4.0), the time has finally come for me to let TM go, and move my Application Security efforts in other directions.

The last 4 years at SI have been an amazing experience and I've learned a lot.

Not only I increased my development skills (.NET, Java, Eclipse, Node, Javascript), I finally understood what TDD is all about and where security fits within the SDLC.

I really want to thank Ed and Jason for the opportunity, and the amazing worldwide TM development team (Serge, Michael, Lucy, Roman, Salle) for making TM 4.0 a reality.

I'm sure mine and SI paths will meet again. In fact I'm still contributing a couple bug fixes to TM, so I'm still around :)

Good luck to SI and all the team

PS: In case you are curious, I'm now helping UK companies to set up their Application Security Programmes (i.e. I'm a part-time 'Head of Application Security')