By David Lucky, Director of Product Management, Datapipe
Have you ever gone to see a movie that would have been amazing if not for one person? The plot was engaging, the dialogue was well-written, and there were strong performances from most of the cast. But there was just that one actor who simply didn’t live up to the rest of the film, and it made every scene he was in that much worse? Simply put, that actor was bad, and brought down the whole operation.
That idea of the “bad actor” can be applied to Internet clients, as well. Fortunately, you’re not hurting any feelings by sussing them out: the bad actors are usually automated processes that can harm your systems. The two most common forms are content scrapers, which dig into your content for their own profit, and bad bots, who will misrepresent who they are to get around any restrictions stopping them.
We’d all like to believe that everyone accessing content will use it appropriately. Unfortunately, we can’t always assume the best, and being proactive in dealing with these bad actors will reduce security threats to your infrastructure and apps.
Even better, blocking bad actors will also lower your operating costs. When these bots access your content, you’re serving the traffic to them, whether you want to or not. That adds more to your overall costs. By blocking them, you’re restricting traffic from a number of undesired sources. Luckily, AWS has a pair of tools you can combine to say goodbye to these bad actors: Amazon CloudFront with an AWS web application firewall (WAF).
With AWS WAF, you can define a set of rules known as a web access control list (web ACL). Every single rule contains a set of conditions, plus an action. Any request that’s received by CloudFront gets handed over to AWS WAF for further inspection; if the request matches, the user can access the content as attempted. If the request doesn’t match the conditions in a specified rule, the default action of the web ACL is taken. These conditions will remove quite a bit of unwanted traffic, as you can set filters by source IP address, strings of text, and a whole lot more. As for the web ACL actions, you can count the request for later analysis, allow it, or block it.
Perhaps the best attribute of the WAF is that you can smoothly integrate it within your existing DevOps, and automate workflows to react. Since bad actors are always switching their methods to mask their actions, your proactive detection methods must constantly change, as well. Having those automations in place is immensely helpful in finding bad actors and restricting their access.
There’s a great walkthrough of how to set up this solution on the AWS Security Blog, step-by-step. Feel free to check it out for more information, or get in touch with us if you have any additional questions. And for AWS customers that need even more than what the AWS WAF has to offer, there are services that are complimentary to the AWS WAF that provide enhanced protection for business critical applications on AWS. You won’t even need to thank the Academy when all of those bad actors are removed.