A blog post that appeared last Friday, observed that Facebook is signing their mail with DomainKeys Identified Mail (DKIM) using a 512-bit RSA key. The author went on to analyze the security of doing so as compared with a longer key, and concluded that a determined attacker could probably factor the public key quickly enough to be useful in sending falsified messages purporting to be from Facebook. The blogger, John Graham-Cumming, said:
Some months ago I started an 8 core Mac Pro machine at work on breaking this key. It ran for 70 days non-stop and was close to a break when I had to use the machine for something else.
If I can do that, pretty much anyone can. And those people will be able to forge mail from Facebook. Facebook has a simple solution, of course, just change the key length. And if you are using 512-bit RSA keys in your DKIM implementation, please stop.
PS The owner of a spam botnet could factor keys like that very quickly. Imagine having a few thousand machines that can be used for key factoring.
One question that comes to mind is how many other domains are using 512-bit keys? It’s hard to answer this question directly because one needs to know the “selector” (key name which is included in the signature) to look up the key, but some of the data Cisco has collected on DKIM metrics gives an approximation. The methodology is a bit indirect because we don’t collect the selector name for successful verifications (only for failures), but since we usually get a smattering of verification failures for domains sending us messages, we can use that data to infer the selector names they use.
Read More »
If you’ve ever taken a look at the (now deprecated) RFC-1700 (a.k.a. “Assigned Numbers”), or at its replacement, IANA’s maintained PORT NUMBERS database, you may have been as puzzled as I was about these two lines:
tcp-id-port 1999/tcp cisco identification port
tcp-id-port 1999/udp cisco identification port
What is that supposed to mean? Does Cisco IOS devices have some kind of custom IDENT server running on ports 1999/tcp and 1999/udp? Well… no. This is yet another instance of “gather around the campfire to hear a story.”
Read More »
The case of the compromise of a video to Wikileaks and unconfirmed claims of compromised U.S. State Department cables by an Army Intelligence analyst stationed in Iraq from classified government networks has been widely reported and commented upon, highlighting numerous security, ethical, moral, and legal lapses. There is no doubt that the military and government organizations involved have been conducting similar, less public reviews and official investigations are continuing. As a case study for security risks, this incident could easily generate a laundry list of issues to be examined as well as an equally long list of lessons learned. Although many of the details may never be fully disclosed due to the sensitivity of national security, many of the issues are fairly obvious and well known to security professionals and have been highlighted in numerous case studies. Similarly, most of the issues should have been addressed in policies, procedures, and controls in most business and government environments. The elephant in the room that many would prefer not to discuss and that is often overshadowed by discussion of technologies and policies are the people: the most complex of security risks.
Read More »
Over the last few years, there has been a push to severely limit entire categories of attacks (such as buffer overflows) by incorporating specific hardware functionality with various compiler options to produce more secure code. When developing software, it is easy to mandate that these compiler options be utilized during software development, but how does the end user determine if the options were actually used? Before we can determine what compiler options have been enabled, we must first examine some of the functionality that has been developed to help protect code. Some of the options include:
Read More »
With the continuous flow of varying government regulations surrounding IPv6, I’ve been wondering about the impact on security. Just having addressing support isn’t enough. Lucky for us, today Cisco announced the early availability of cloud-based IPv6 support for the Cisco IronPort Email Security portfolio. Cisco email security customers of all form factors — appliance, cloud and hybrid — are able to send and receive IPv6 emails through the Cisco infrastructure. Customers so far are very pleased.
The continuous growth of the Internet requires that its overall architecture evolve to accommodate new technologies to support the growing numbers of users, applications, appliances, and services. As per Cisco and industry estimates, the IPv4 address space will be exhausted in the next two years. This will cause every organization to face the inevitable transition from IPv4 to IPv6.
In recent months, Cisco Security Intelligence Operations (SIO) has witnessed a rise in criminal activity on IPv6, particularly as sources of email threat messages and in channels used by botnet command-and-control infrastructure.
Read More »