Traffic log messages record requests that a FortiWeb policy accepted or blocked. If the request was successful, it also includes the reply. Each log message represents its whole HTTP transaction.
Traffic logs do not record non-HTTP/HTTPS traffic such as FTP. This type of traffic is forwarded to your web servers if you have enabled IP-layer forwarding.
Traffic log messages are described below. For descriptions of header fields not mentioned here, see Header & body fields.
Meaning |
---|
Traffic matching and complying with a policy passed through or by FortiWeb. If there is an error in the message, however, and the request/response used HTTPS, FortiWeb could not scan it. Depending on the mode of operation, an attack could have bypassed FortiWeb. |
Solution | |
---|---|
Reponse times can often be improved, for example, by regular expression tuning, offloading SSL/TLS from your back-end server to your FortiWeb (especially if the model supports hardware acceleration), and/or offloading compression. For performance tips, see the FortiWeb Administration Guide. If HTTPS traffic is not flowing as you expect or not being inspected, and you have recently enabled HTTPS, typically this is due to a misconfiguration. The error message in the
If your appliance was operating in reverse proxy or true transparent proxy mode, the traffic was blocked, and no attack could have passed through to your protected web servers. No action is required except to make sure that you have uploaded to FortiWeb the correct certificate for all protected web servers. Otherwise, if your appliance was:
examine the web server to determine whether or not an encrypted attack has passed through. You should also examine your web server’s HTTPS configuration and disable cipher suites and key exchanges that are not supported by FortiWeb so that during negotiation with clients, your web server does not agree to use encryption that FortiWeb cannot scan for attacks. By the nature of log-only actions, detected attack attempts are logged but not blocked. You may also want to determine if the attack is from a single source IP address or distributed: blacklisting an offending client may help you to efficiently prevent further attack attempts, improving performance, until you can take further action. By the nature of the network topology for offline protection mode (which can potentially cause differences in speeds of the separate routing paths), and asynchronous inspection for transparent inspection mode, blocking cannot be guaranteed and some key exchanges are not supported. For details, see the FortiWeb Administration Guide. |
Field name | Description |
ID ( |
All traffic log messages share the same ID ( |
Sub Type ( |
All traffic log messages share the same subtype ( |
Level ( |
See Priority level. |
Message ( |
If the HTTP request triggered the FortiWeb web caching feature, the message begins with The HTTP/HTTPS request’s:
such as:
If the transaction used HTTPS, and there was an error when either decoding it or participating in the handshake, there may be an error message instead of the HTTP method, such as:
|
Source Country ( |
The country that is the source of the traffic. |
HTTP Content Routing ( |
The name of the associated HTTP content routing policy. |
Server Pool Name ( |
The name of the server pool in the associated server policy. |