A new breed of cyberattack is on the rise. Although it was practically unheard of a few years ago, the third-party data breach is rapidly becoming one of the most infamous IT security trends of modern times: Target, Home Depot, Goodwill, Dairy Queen, Jimmy John’s and Lowes are just a few of the US companies to have lost massive amounts of customer records as a result of their contractors’ usernames and passwords falling into the wrong hands.
What went wrong? Hackers have started to see contractors as the easy way into their targets’ networks. Why? Because too many organizations are still using yesterday’s security solutions, which weren’t designed for today’s complex ecosystems and distributed (read cloud-based) applications and data.
Here are four examples of solutions that, in their traditional forms, simply aren’t capable of stopping third-party data breaches. Could your company be at risk?
1. Firewalls and Access Control Lists
Many organizations still control traffic flow between network segments in the same way they’ve done for decades: with firewalls and access control lists (ACLs). Unfortunately, security in the modern age isn’t as simple as just defining which IP addresses and ranges can access which resources.
Let’s say you have a single VPN for all of a department’s workers and contractors, with every authenticated user getting a DHCP-allocated IP address. Your firewall rules are going to have to be wide open to suit the access needs of each user on the IP range, and yet you’re not going to be able to trace suspicious activity back to a particular account and machine.
It’s also a lot of work for your IT department to set up and maintain complex firewall rules across the entire organization, so it’s not unlikely that they’ll make mistakes, respond slowly to employee departures, and leave access wider open than it should be.
2. Authentication and Authorization
Leading on from this, another problem with ACLs is that they generally rely on static rules, which in no way account for the security risks of today’s distributed workforces. A username and password pair will unlock the same resources whether used from a secure workstation at a contractor’s premises or from an unknown device on the other side of the world.
Authentication and authorization rules should be dynamic rather than static, and adjusted on the fly according to the risk profile of the connection. One of your contractors needs remote access to a management network segment? Fine – but only if they use a hardened machine during office hours. If the context of their connection is more suspicious, you might consider two-factor authentication and more limited access.
3. IPsec and SSL VPNs
More than nine in ten organizations (91 percent) still use VPNs – a 20-year-old technology – to provision remote access to their networks. It’s potentially their single greatest risk factor for third-party data breaches, because both IPsec and SSL VPNs are readily exploitable by hackers.
In an IPsec session, remote users are treated as full members of the network. Nothing is invisible – they have direct access to the underlying infrastructure. So, if they’re malicious, they can start digging around and looking for vulnerabilities in seconds.
SSL VPNs, meanwhile, deliver resources via the user’s browser. And what web application has ever been secure? Tricks like SQL injection and remote code execution attacks make it trivial for hackers to start widening their foothold on the network.
4. IDS, IPS and SIEM
Finally, a word on the technologies organizations use to detect data breaches. IDS, IPS and SIEM are generally mature and effective solutions that do the job they’re intended to do: identify suspicious activity on the network.
However, the combination of the antiquated technologies described above means that most networks are rife with false positives: legitimate users and harmless applications causing suspicious traffic in the network layer. Change this model, and IDS, IPS and SIEM systems might start to deliver more value. As it stands, though, they’re often resource-intensive and reactive rather than proactive, so they’re not really equipped to stop hackers in their tracks.
The Alternative to Prevent Third-Party Data Breaches
In the new world of pervasive internal and external threats, distributed organizations and global ecosystems, the perimeter is more porous and less relevant than ever. The old models simply aren’t working. We need to move from perimeter-centric, VLAN and IP-focused security to a model that focuses on securing the entire path from user to application, device to service – on a one-to-one basis.
That’s where solutions like AppGate that enables organizations to adopt a software-defined perimeter approach for granular security control become increasingly a must have security solution. AppGate makes the application/server infrastructure effectively “invisible.” It then delivers access to authorized resources only, creating a ‘segment of one’ and verifying a number of user variables and entitlements each session—including device posture and identity—before granting access to an application. Once the user logs out, the secure tunnel disappears.
Philip Marshall, Director of Product Marketing, Cryptzone
[Cloud Security Alliance Blog]