We updated the passive web application IDS rules from Spiderlabs. These rules should be much more useful. You can set the mode in the sensor configuration as Server or Client. In server mode the Web IDS will alert if we see any suspicious request that caused the server to respond with any HTTP code. In client mode we generate alerts only if a client malformed requests are (1) persistent across multiple servers, (2) trigger different Web IDS rules and (3) elicit error responses from the servers.
Server mode allows you to see how you web servers are being targeted and helps you create better Web application filtering rules. Client mode triggers more rarely and finds persistent attempts that go beyond occasional false positives or probing. If you see a client from your network triggering the We application IDS system, it is a good indication that it is running bad applications or it is accessing questionable services.
If you want to turn off some of the rules, search for the rule ID under Rules->ModSecurity Rules (after you select the appropriate sensor), slide the disable switch and Save and Restart.
To search for Web Application IDS rules only look for modsec using the Search form and set the event type to any. As shows above modsec alerts will also be correlated with your Emerging Threats IDS rules and other external syslog messages and therefore you will also see them if you search for correlated alerts.
Let us know if you have any questions.