I assess the effectiveness of the machine learning-driven threat detection in Barracuda Web Application Firewall as sometimes behaving abnormally, often showing me false positive attacks, so I have to fix these attacks from time to time. From a stability point of view, I would definitely rate Barracuda Web Application Firewall a seven out of ten. There is definitely some room for improvement; nothing is perfect in the world. I am not satisfied with the technical support from Barracuda. I am somewhat disappointed with the technical support that I have received so far. Whenever I generate a ticket for my problem, it goes to the Indian support team, and they all the time start with the most junior team member, consuming all my precious time. At the end, I have to close that ticket without any satisfactory solution. I have complained that they should shift my support to any other region because I don't need Indian support; they are simply pathetic and not up to mark. To improve Barracuda Web Application Firewall, customers should be given ongoing training opportunities regarding the product and its features. I am not familiar with many features that are available, only using those which are necessary for my applications. I believe Barracuda must provide clearer product information or training sessions to make it more user-friendly, as sometimes its interface can be rigid and lacking in helpful resources or user tutorials about its features. For it to get closer to a ten, I think advanced reporting is missing because, as I mentioned earlier, there are many false positive events being recorded. Often, when I analyze these attacks, they turn out to be genuine customers or users interacting with my product, but Barracuda tags them as attackers. Reducing false positives must be a priority.
