ANALYSIS: While the details regarding the sources and methods of the Oct. 21 DDoS attack that took down vast areas of the internet are still unclear, insecure endpoints are the likely culprit
The massive distributed denial-of-service attack that took out many popular internet services on Oct. 21 was so effective because it targeted part of the core infrastructure of the internet—the Domain Name System.
The DNS service that was targeted, DynDNS, is used by a number of major websites, ranging from Twitter to Spotify. When their access to DNS was interrupted, those services became inaccessible.
The DNS, the service that translates the names you type into your browser address bar to the IP addresses your computer uses, is critical to functioning of the internet.
A DDoS attack on the DNS server that’s used by those major services makes the internet unavailable to the internal networks of those services, which effectively takes the portion of the internet that uses those services offline.
Fortunately, there are other DNS servers available on the internet, including several that are publicly available to anyone who wants to use them. Unfortunately, switching from one major DNS service to another is easier said than done.
The reason it’s difficult is that those major providers frequently have contractual agreements for DNS services that limit their ability to switch on the fly to another service. Secondly, even without such a limitation, it takes the IT staff a while to find and change all of the network assets that might request DNS information.
At first glance, it might not seem all that hard to add secondary or tertiary DNS services. After all, you just need to make the change in your internal DNS server and you’re done, right? But that isn’t necessarily the case. What might work for a smaller company with just a couple of DNS servers grows far more complex in a large organization with many locations and data centers.
Adding to the complexity is the fact that in many companies such network infrastructure components are poorly documented and may have been set up by someone who is long gone. This means that even if you have the information needed to add another DNS provider, you may not be able to do it quickly.
The answer for companies that have been adversely affected by the DNS cyber-attack is to be proactive. Begin the process now to identify the points on your network that request outside DNS data and configure them so that there are multiple layers of DNS services that can found.
When there’s a contract limitation on adding additional DNS targets, it would be wise to renegotiate that contract, pointing to this DNS outage as the reason your provider should allow it. This might cost more money, but the cost of downtime is much higher.
The next question that must be answered is how this attack could get so massive and affect so many internet sites.
Originally published on eWeek