The Home Depot data breach and why hackers love FTP

This is a transcript of the Axway podcast of the same name.

ANDREWS: Security can always be breached. It’s that visibility piece that’s more about detection that really would help people at Home Depot understand that something unusual is happening in their environment. That is, if they were able to get past the security portions of the APIs anyways.

ANNOUNCER: From Phoenix, Arizona, this is The Axway Podcast. Here’s your host, Mike Pallagi.

PALLAGI: In early September, The Home Depot’s banking partners and law enforcement notified them of unusual activity connected to their payment systems. The Home Depot’s IT security team immediately began working with leading IT security firms, their banking partners, and the Secret Service to investigate. That investigation confirmed that a breach of The Home Depot’s payment card systems occurred. Since then, they’ve fixed it, but I had a question: What could they have done to prevent it? Here’s what John Andrews, Axway’s Director of Solution Marketing for Managed File Transfer, had to say.

ANDREWS: FTP is a very old technology. The original specification for it was published in April of 1971. The specification actually pre-dated TCP as a major way for technical communication between computers. And so it was done more as an academic research, being able to share… How do academics share information electronically rather than putting their research into an envelope and then mailing it and hoping it gets there? Got used very quickly in computer science circles because it was a way to share information.

PALLAGI: Andrews said that FTP was never designed with security in mind and because of that, it’s become one of the favorite venues for hackers looking to get into a corporate network.

ANDREWS: They have built security on top of it. However, the secure variance, while they provide protection around being able to log in or protect the data in motion, they don’t provide any audit or log capability. So thinking back to the API example, if you can’t track or have visibility into what’s being done, it becomes harder to detect unusual patterns. And so that kind of lack of traceability, visibility, auditability make FTP a very insecure piece of software. If you think along a CSI type of metaphor, there isn’t a lot of evidence left by FTP to trace back or identify who actually committed the crime. And that’s why hackers love it.

PALLAGI: So when it comes to logging into an FTP, you just need to use your name and password and there’s nothing to authenticate that that username and password belong to the person who’s trying to log in. Also, when a person logs in, it’s not actually recorded, so there’s no audit trail.

ANDREWS: If a file is moved, there’s no audit of it being logged. There’s plenty of examples of hackers using FTP to gain user credentials. If you Google on some of these, you’ll find them. And I have a bunch of examples. Like, in 2001, Yale University had 43,000 people — user IDs — exposed because the database information with all that user information was stored on an FTP server. In the same year, in 2001, 40,000 Acer customers had their details stolen — again, because the information was stored on the company FTP server. More recently, 7,000 FTP sites had their credentials circulated in underground forums. And that was found from a company called Hold Security. That was probably just about a year ago when that happened. If you look at all the other events that have happened in the last little while, while it hasn’t been clearly stated, FTP could be a primary suspect in allowing hackers to get into systems.

PALLAGI: Is there anything a business can do to track that activity? Some set of improvised actions? A best practice?

ANDREWS: Once somebody gets access to an FTP server, there’s no log of that activity. Now, you can write manual scripts to try and track that, but it’s far from foolproof, and often requires a lot of maintenance, so you’re never quite sure if you’re getting all the information. In fact, if you Google for a Python script, you can find a script written in Python that will scroll through a range of IP addresses to tell you if there’s an FTP server on that machine, whether it’s working or not, and whether the anonymous login is available on that server.

PALLAGI: What else is it about FTP that makes it so attractive to hackers?

ANDREWS: There are a number of things around FTP that make it highly suspect to hacking. For example, it can be used in a brute force attack, so just checking every single port that is on an IP address to see if it has an FTP server exposed on it. You can do bounce attacks checking to see whether or not your attempt to log into the system is available. Use a port command and try to just access … use the FTP server as a way to connect to another system. You also have packet capture. So the idea is that if you know which port the FTP server is listening on, you can listen on that port and just analyze the packets as they’re going to that server. And while the secure versions of FTP can address this, you still don’t have traceability. Ultimately, an FTP server sits on top of a file system that is usually connected to your internal network. So once you get access to the FTP server, you then have access to the internal file system. Once you have access to the internal file system, you can access databases, you can access LDAP stores. If you know what you’re doing and know where to look, the FTP is that proverbial back door to get into a network environment, then find almost anything you’re looking for.

PALLAGI: What can organizations do to reduce their reliance on FTP and secure information in motion?

ANDREWS: First and foremost, there are two things that really come to mind. And that is there is a higher level of security, meaning that username and password aren’t always going to be enough to connect to an MFT solution. While we can mimic FTP functionality, the ability to access it may not only require username and password, but authentication, so we can up the level of security needed for people trying to access. We also abstract away the actual physical … the physical file system away from the attackers so they don’t have easy access to the back-end network. Most importantly, we track and audit all of the interactions. So if you log in as Mike Pallagi, we will see that login, exactly what protocol you were using, and what you tried to access. That audit log is hugely beneficial, especially in diagnostic and troubleshooting situations. And that is provided through a level of visibility that FTP doesn’t have. With an MFT solution, not only are you going to log that activity, you’re going to be able to see how frequently, how often, and who’s trying to gain access. That preventative… or that visibility, allows for preventative measures rather than reactive measures.

To download the first two parts of Axway’s three-part MFT Survival Guide series, click here

To view the video blog on YouTube, please click here.

The value that operational intelligence brings to banking and payments

This is a transcript of the Axway video blog post of the same name.

LAURENT VAN HUFFEL: Hi. My name is Laurent Van Huffel. I am responsible for the Axway business unit of operational intelligence in North America. Today I wanted to talk to you about the value that operational intelligence can bring into the banking world and, more specifically, in the world of payment.

What we hear from many of our customers is they often find out issues after the fact, when it’s too late. When they have either missed an SLA or missed their cut-off payment. Whether you’re looking at the high-value payment, SEPA, SCH, check image, or any other form of payment, specifically with the faster payment trends, or what people call the immediate payment — things are moving in a more complex world. And so, because customers are often organized from a business operations standpoint in silos, they only view a partial view of the execution of the business process. Because they often get information when it’s too late, sometimes when their own customers call them to tell them they have a problem that they’re not even aware of. This is obviously not good for the business.

In a nutshell, operational intelligence helps business operations to proactively identify and resolve process issues prior to impacting their business and their customers. Why “proactive” is so critical is because if you take a high-value payment, for instance, that’s an itinerant type of payment — you need to have enough time to act upon information so that you can mitigate your risk and at least make sure you’re going to meet your cut-off all the time.

Our customers are telling us that their concern, for instance, is not being able to identify when a payment is stuck in a business step for too long. Or not being able to identify that they are processing an abnormally high volume of payment at that time of the day. And why this is a concern is because they may run out of capacity.

Or if you have an abnormally high number of payments failing the STP route, going into a non-STP route, this would mean that they will go for manual repair. Maybe you don’t have enough people on the bench to repair those payments fast enough, which means that you may be taking the risk of missing a cut-off.

For multiple reasons, customers want the capability of end-to-end visibility. They want to be able to track their business transactions — their payments — end-to-end, and identify any potential issues during the execution of these payments against the business commitment they have made to their customers. If something is abnormal, they want an exception-driven, actionable-intelligence dashboard or alert that tells them that something is not normal and therefore they should do something about it.

Operational intelligence — the way it works, it collects in real time critical data from multiple sources in order to map out the end-to-end view of the business process — would call out this information using real-time analytics and publish actionable intelligence to each stakeholder that (has) a role in the well-being of their business processes. What is important to understand is that each stakeholder may have a different objective, and therefore, they will have different concerns. The content of this dashboard can be easily customized and tailored to each stakeholder.

If you are an executive, you don’t want to see the same information (that you would see) if you (were) a supervisor in a control room for payment. The Axway Decision Insight operational intelligence solution is a non-development platform. You can configure dashboards very easily, very quickly. You can configure your set of dashboards in less than three days. Then, of course, you would need to move into production and go for the UAT phase and so on.

The way that the solution works is it’s a very iterative and incremental process. We want you to have the ability to easily configure a dashboard, easily modify a dashboard, easily increment the value of the dashboard. Think of it more as a journey as opposed to a destination where you will want to use a dashboard today. As business conditions change, you will have the ability to very easily modify that dashboard to these changing conditions. At the end, the typical business drivers for which banks decide to use operational intelligence are better risk mitigation, better customer experience improvement, and increasing the operation efficiencies which result in less cost.

To learn more about Axway Decision Insight, please click here.

To view the video blog on YouTube, please click here.