When web requests and RUM beacons are directly sent to an instrumented server, Dynatrace identifies the IP addresses of your end users' devices via the socket connections.
When you use uninstrumented components such as load balancers, CDNs, or proxies, the remote IP address is different from the original IP address. For such cases, Dynatrace also watches certain HTTP headers. These headers are most frequently used to identify the originating IP address when a client connects to a web server through an HTTP proxy, a CDN, or a load balancer.
You can view the list of headers that Dynatrace uses to determine client IP addresses for your applications. To do that, go to Settings > Web and mobile monitoring > IP determination. Dynatrace processes these headers in a specific order, but you can change the processing order and add your own headers.
You can update header names, delete headers you don't need, and change the order in which Dynatrace processes the headers.
If your client IP addresses are located in a different header, create a custom rule so that Dynatrace can correctly identify the IP addresses.
To add a custom client IP header
You can create up to 10 custom client IP headers per environment.
Adding a custom client IP header might be required in one of the following cases.
There might be several reasons why you don't see the correct client IP addresses for your application. Check the sections above for detailed instructions.
The IP detection headers in regular requests of the application are irrelevant for the IP detection of captured RUM data, even when they are captured as traces and those traces are correlated to user actions.