What’s Not in the IBM Redbook: Apache Users and IFS Security
A few years ago I worked with the lead developer of RJS Software’s WebDocs-iSeries document management solution to see whether we could use authorization lists to provide a standardized method for securing WebDocs-iSeries objects on the iSeries.
You see, the iSeries is strange in that it has two file systems that are both “native”, and WebDocs-iSeries uses both. The first and the original, is the library/object model that has been around since the early System/36 days. A master library, QSYS, contains system objects and all other libraries, which may themselves contain objects but not libraries. This produces a very flat file structure, with any object being fully specified simply by naming the library it is contained within and its object type.
Later, a UNIX-like file system was added (IFS or Integrated File System) but without removing the old file system; instead, libraries appear as UNIX directories with objects appearing as files and object types their extensions. Native applications were blissfully ignorant of the change, and I’ve actually spoken to administrators who were similarly ignorant. In addition, many iSeries applications do not explicitly work with the IFS.
With V5R3, IBM switched from using its own web server on the iSeries to using the popular Apache web server. Before that WebDocs-iSeries had been storing the actual documents uploaded to it in the IFS, so any comprehensive security solution was going to have to accommodate the library/object model as well as the files and directories in the IFS. Authorization lists fit the bill perfectly.
Or so we thought. Ultimately, they proved too cumbersome for our customers and their real-world utility was minimal. But along the way I stumbled across an odd mismatch between the level of authorization that was theoretically required for Apache in IBM’s Redbook and what was actually required.
Apache on the iSeries has two primary user profiles it uses: QTMHHTTP and QMTHHTP1. The former is used for accessing the web site pages under /www for a given Apache instance, while the latter is used for executing CGI programs. Since the same iSeries job may be used for both operations, Apache switches the user profile of the job when it runs CGI programs.
What this implies is that the only files QTMHHTTP should ever need access to are those in /www. But, in fact, it traverses the directory structure of the IFS to check whether a given CGI program is present before the job switches users. Since our CGI program object is in the RJSIMAGE library, QTMHHTTP needed read access to /QSYS.LIB/RJSIMAGE.LIB/* before it would perform the switch and allow QTMHHTP1 to run the program.
While unexpected, this mismatch between expectations and actuality isn’t necessarily a problem. But paying attention to such mismatches can be educational. The Linux From Scratch project has actually made a sort of perverse game of this form of pedagogy, with their “User Based Package Management” approach. It looks masochistic because it is, but it’s also enlightening: verifying security is all about attention to such details.
Last I checked, V7R1 behaves the same way, and the Redbook still hasn’t been updated. So it goes …
The Insidiousness of Cellphone Malware
Hackers and cyber criminals continue to get more sophisticated, especially when it comes to fooling people via less traditional attack vectors. For example, I got an interesting text on my work cell phone earlier this week …
“heyy liveurpic.com that’s what I was telling you you should join.”
The message came from a local Twin Cities telephone number that I didn’t recognize. So, being the suspicious type that I am, I “Googled” the phone number and found that the same number was spamming local numbers here in the Twin Cities and that there are several complaints about it already.
Taking this a step further, I fired up a clean VM with antivirus on it that goes through our Sophos WS1000 web security appliance and pulled up the URL in Google Chrome. Sure enough, the site was blocked and classified as “High Risk” and that “Mal/HTMLGen-A” has been found on the site.
Sophos classifies malware by its behavior, and Mal/HTMLGen-A happens to be a very commonly used browser payload delivery mechanism that affects Windows, Mac OS X, and Linux.
After doing a little more digging, I found that the URL that was sent resolves to 220.127.116.11, which has some really interesting history to it. This particular IP range was in Volgograd, Russia back in 2010, and is now owned by a Romanian internet hosting company parked in Schiphol, Netherlands.
A few companies I’ve worked at in the past IP block Romania at the firewall since there is a high amount of fraud and scam traffic that originates from there, so it makes sense that they would host from another country to get around geo location-based IP blocking.
And finally, we arrive at the nature of the attack itself.
Cell based text messaging. Both my phones run a variant of Linux, one being Apple iOS (work cell) and the other Android (personal cell). If an attack is highly sophisticated, it will include some sort of detection routine so it knows which payload to push onto the phone such as iOS JailBreak code or Android rooting scripts/programs – something that will allow the attacker to access your dialer without you knowing about it to make premium calls or text messages. Sure, you’ll figure it out once you get your next phone bill, but by then it’ll be too late.
If the attack is low-tech, then it may be as simple as trying to get you to share you URLs (like Chrome does) with a far more vulnerable Windows OS and use some canned Blackhole Exploit Kit scripts that were purchased to farm some bank info.
I’d love to take the time to pull the payload code apart to see what it’s trying to do, but from the poor English used in the text message itself, I’m betting its low tech. If the attacker can’t be bothered to try to make the message look legit, you can more than likely bet they bought their malware off the shelf.
Moral of the story: your cellphone is as at as much risk as your computer. Be wary, be vigilant.
What you need to know about HIPAA Compliance
The Health Insurance Portability and Accountability Act of 1996 (HIPAA) and the Health Information Technology for Economic and Clinical Health Act of 2009 (HITECH) are national standards put into place to: 1) protect the privacy of personally identifiable heath information, 2) secure personally identifiable health information stored and/or transmitted electronically, and 3) promote the meaningful use of health information technology. As of September 23, 2013, compliance with these standards will be mandatory for all applicable healthcare entities AND their third-party vendors. Failure to comply with these standards can result in both civil and criminal penalties.
Do you know if you and your third-party vendors are HIPAA compliant? Do you need some help finding out?
Thankfully, RJS Smart Security provides healthcare organizations and their business associates with a comprehensive evaluation of their protected health information and data environment with a HIPAA assessment. RJS follows the emerging Penetration Test Execution Standard (PTES) as the model for these assessments.
HIPAA Compliance with RJS Smart Security
Our HIPAA assessment focuses on HIPAA regulations for companies who may or may not be facing an audit. During this engagement, we look at the following:
Compliance is not the same as security. Compliance defines the bare minimum needed to protect specific data types or industries. This engagement identifies lean ways to meet HIPAA requirements so that your business still has resources for security.
Strategy is perhaps the most critical security task, as well as the least used. A strategy engagement identifies what HIPAA compliance pieces you already have, what you need and how to get there. Several risk assessment methodologies are available to guide us in crafting a security plan that ideally fits your health care business.
Policies and Procedures review the body of written controls that are currently in use and assess how well they are being followed. Policies often age as technology improves and procedures are followed poorly as they lose applicability.
And, depending on the scope of your engagement, we also examine:
Network Vulnerability involves scanning your network to identify the operating systems and applications in use. Older applications are a common vector in successful attacks, but these flaws can only be addressed if you know they exist.
Web Vulnerability focuses on the many common problems found in web applications, such as SQL injections looking to steal or alter data, scripting to exploit users or weak configurations.
Data Analysis identifies the documents and databases an organization is storing and the risks resulting from possible data leakage. This analysis helps you determine ways to centralize storage and eliminate the unnecessary.
A Success Story
Healthcare Services Provider*
A medical services company had grown by acquisition and upon examination of their network infrastructure, an increasing number of legacy applications were cause for mounting concern. With fewer people on staff who knew how they functioned, it was increasingly difficult to maintain their security and comply with the regulations of HIPAA and HITECH.
To help simplify the process of application management while working within a tight budget, the company hired RJS to review one legacy application each quarter. In the short term, this approach helps the firm meet their Business Associate requirements. In the long term, the newly-built applications can be maintained at a lower cost.
* The company name has been kept anonymous due to the sensitivity of the work performed.
Sophos: Pushing the Boundaries
Several people have been asking me lately if I still prefer Sophos technology. After all, they recently released a bad update and Tavis Ormandy’s recent paper illustrated some design flaws in the product.
There’s spin on both sides. Unsurprisingly, Sophos is downplaying the issue and Tavis Ormandy’s tone in his second paper is much like that of his first. So I thought it would be good to explore the issues more fully.
I am disappointed in Sophos’s recent fumbles, but not disheartened. I still think that, for a great many companies, they have the best solution available.
There are two core problems here. The first is that of scaling. As companies grow bigger, they often become slower to react. Sometimes, they fail to adjust to their new reality. Sometimes, however, they get through it and become incredible companies. There are indications that Sophos is beginning the turnaround.
The second problem is industry-wide and has to do with market-splintering. Today, we are facing a splintering security vendor space and reports like the ones from Tavis are a symptom. Reports we’ve been getting for the past few years about AV and IDS being “dead” is yet another symptom. There’s nothing wrong with these reports and it’s good that people are thinking about the issues. But unfortunately, they’re missing the big picture.
In a splintering space, there is an increasing deviation between what a product actually does and how it is branded. This continues until the vendors wake up and pivot their branding to better match what they actually do. This always takes longer than we’d like, because employees of a company are driven by their brand even more than their customers.
As I see it, the AV vendor space is breaking apart into four chunks:
- Traditional AV – Focused on being lightweight and supplementing the protections built into an operating system.
- Anti-Malware – Focused on monitoring and responding to bad and potentially bad things that can happen to an operating system.
- Application Whitelisting – Focused on locking down an operating system to only allow known applications to run.
- Malware Analytics – Focused on providing detailed data about events so human analysts can make appropriate decisions.
We are also seeing the attack space splintering as well. Specifically, we’re seeing a tiered structure emerging:
- Background Radiation – A constant stream of trivial attacks, legacy viruses and worms that float about the internet.
- Industry-focused Asset Attack – Attacks that focus on specific industries aiming to steal monetary assets. These often rotate between industries, “campaign” style. At present, Western banks are in vogue. Indications are that these attacks are run by organized criminal groups. If you have above average protection in this space, you gain significant competitive advantage as attacks are driven to those that do not.
- State-sponsored IP Attack – Attacks that also focus on specific industries, but are run by better-organized groups suspected of being funded by state agencies. They aim to steal intellectual property. You get significant benefit from being above-average here too.
- Industrial Espionage – Attacks focus on specific companies and likely come from other specific companies. There is no benefit to be gained from being above-average in defense, as that just creates rapid escalation in an arms-race pattern of growth.
The third splintering effect is familiar to those of you who have heard my talks or read my comic book.
In the defense space, we have two primary trends emerging based on complexity. Both are valid, but they are completely incompatible with one another (at least on the large scale).
- Simplification – This trend involves firms who outsource much of their operations and infrastructure to managed SOC providers, cloud providers, SAAS providers, etc. The idea is that by focusing on their core business and finding trustworthy partners, they can become more nimble and therefore, more profitable.
- Complexification – This trend involves firms who increasingly centralize their infrastructure through virtualization, log management, analytics, etc. The idea is by taking more control, you can better analyze the business and find regions of improvement.
So after this rather long diversion, what does it mean for Sophos? Well, their core strength is offering administrators a solution to rapidly and consistently provide a reasonable level of protection that gives early indicators of attack. This means they’re a great fit for anti-malware up to and including the State-sponsored IP attack tier and work best in simpler environments or in simpler subsets of complex environments.
I do not think that they’re the best solution for highly-targeted complex companies, as they are not an analytics tool. They’re also not the best solution for extremely simple firms that want “set it and forget it” technologies. Security takes work and if you put the work in, you get better security. You should pick a solution that allows you to put in more work than you currently do, but does not require that additional work for the product to still be effective.
Many of the critiques of Sophos tend to be at the edges. And Sophos doesn’t help these concerns by marketing as if they fit everywhere for everyone. I think they’ll eventually reach this goal, but they’re not there today.
If you’re in a highly complex environment that is facing constant incursion from well-funded foreign governments, Sophos better not be your only tool. You need Analytics, too.
If you want the cheapest solution out there that you can install and ignore, Sophos won’t work well for you either. That’s where Traditional AV and the ability to wipe and reload when it fails comes in.
Operationally, you need to figure out what sort of attack and defense space you’re in, so you can select the tool(s) that will provide the best protection for your business. If you overlap spaces, you’re going to need multiple tools. One of the biggest reasons I really like Sophos is because of their flexibility to play in overlapping spaces. They layer quite well with other security products, which is a great benefit for their customers.
Just remember: We live in a complex world. There are no magic bullets. Security requires thought.
Sprinting through Security
We’re all familiar with old school consultants. These are people hired at $20/hr and rented out for $150/hr. It’s good business, if you can get it. All too often, however, the work is neither enjoyable for the consultant nor useful to the client. After years of trying and failing to make the old model work, I decided it was time to throw it out and start over. Thankfully, RJS agreed. As of today, we are one year into the process of reinventing security consulting.
The fundamental difference is that we’ve fully embraced the fact that the idea of 100% security is a trap. You can never be completely secure, so why base a security project around the idea that you can be? Instead, we focus on achieving a measurable improvement over “today.” Different businesses have vastly different security needs, so once you shift the goal away from “find and fix all the problems” to “strike a balance between defense and response,” myriad solutions become available.
Having a large number of solutions is great, as we can select the one that fits your company’s unique situation the best. But remember, it’s not perfect and will need constant attention to avoid “analysis paralysis” and to stay current with new security trends. To combat this, we look at the second key difference: time-bound tasks, or as we call them, security sprints.
With anything you do, there is one resource that completely vanishes — time. Other consulting approaches focus on minimizing either money or an amorphous concept of risk. The catch with those is to clearly pre-identify “risk reduced” or “money saved,” time is required. Since time is billed, it can cost a significant amount of money to identify how much money you’re saving!
Want to know how much more secure a project will make you? Pick a small project that can be done in a week or two, do the project, then measure. There’s no guess-work, no scope-creep and most importantly, no spending more money than required to improve your defenses.
This process affords another advantage we did not anticipate. In many cases, security fails because the people put it in place to manage it are often not those responsible for maintaining it. Since security tends to weaken over time as attackers constantly improve, it is imperative that people explore alerts, identify what they mean to the business and take appropriate action. When the people who must manage the systems are not involved with the initial configuration, they tend to lose a lot of time tracking false alerts or worse, missing legitimate issues.
Since our consulting process is time-bound and focused on helping improve security after we leave, we work on a lot of small projects. These projects are designed so that, when done, they can be absorbed into the business’s existing operations. We then come back for iterative tuning engagements and, over time, help maximize the business’s use of technology. This avoids the common problem of security being “someone else’s” issue, while minimizing the disruption that new technologies can cause.
In the end, after a year’s experimentation, we’ve found that a cyclical short-project consulting model has given our clients a level of security far greater than the traditional defense-only approach. While this didn’t surprise us (after all, that’s why we did it), we were surprised to find that these engagements generally came in 25% to 50% lower in cost than the traditional model. We’ve done security assessments, implementations and strategy planning sessions and, in every case, have achieved better security at a lower cost.
Please contact us if you’d like to learn more about our sprint model and how it can help you achieve a better state of security at a fraction of the price.