Hello and welcome to this lecture regarding monitoring your WAF service. As a prerequisite of this lecture you should have a basic understanding of AWS Simple Notification Service, and AWS CloudWatch. For more information on CloudWatch we have a lab which will guide you through an introduction to the service which can be found here. If you decide to monitor the activity of your WAF service and how it's performing then are a rang of features that allow you to do this. However as with any monitoring you want to conduct, you need to make sure you know what you are monitoring for, what is the purpose of the monitoring, and also how often you intend to monitor and what elements of the service you want to monitor. Are you looking to perform monitoring to maintain the liability and operational performance? Or are you looking to understand trends allowing you to implement additional controls and making your infrastructure even more secure? Once you know what you're monitoring for and why it becomes easier to select the best approach. And this is true for all monitoring. Let's start with the service itself. From within the AWS WAF service dashboard in the management console you are able to view certain statistical information for the Web ACLs you have created. You can't generate reports from here however the service dashboard does provide a graphical view of the requests that match each of your rules within each of your Web ACLS along with the total number of requests. For additional monitoring functionality and features you can use AWS CloudWatch. 

As I mentioned in a previous lecture WAF integrates well with AWS CloudWatch allowing you to monitor set metrics for the service. WAF CloudWatch metrics are reported in one minute intervals by default and are kept for a two week period. The metrics monitored are AllowedRequests, BlockedRequests, CountedRequests, and PassedRequests. These are very much self explanatory, however these metrics provide a SUM count of web requests that hit a specific rule or Web ACL. The past request metric might throw you off but essentially this is a metric that lets you know how many requests didn't match any rules within your Web ACL. You may have noticed during the demonstration I gave earlier that on the first step of the configuration of the WAF service you are asked to enter a name for the Web ACL. At the same time the service is automatically generating a CloudWatch metric name with the same name to allow you to report statistics against this Web ACL. For each Web ACL you have there will be an associated CloudWatch metric, and the same applies to WAF rules. From AWS CloudWatch you can perform all the same functions as with other services that you monitor such as alarm creation and viewing the history from the graphical interface. If you are creating a reactive policy within your Web ACL instead of a whole host of static pre-configurables that can get a little difficult to manage, then you could set a count action for a number of rules you have configured. From within CloudWatch you could then set an SNS notification to alert the security team to change the relevant rule action to either Allow or Block as required. By setting reactive rules it could help reduce a number of rules within your Web ACL and reduce the number of false positives that occur more often with larger rule sets. That brings me to the end of this lecture. Coming up next I'll look at the limitations associated with AWS Web Application Firewall.