3.13.39. UseContentsForDuplicateDetection Previous topic Parent topic Child topic Next topic

This optional flag causes the Client to use an alternative duplicate detection algorithm. Normally, Client detects duplicate requests by looking at the source IP address, source IP port, the RADIUS request authenticator and the RADIUS request identifier of incoming requests. Requests with those identical attributes received in the last DupInterval seconds are deemed to be duplicates.
This parameter is not needed in a ServerFarm environment when shared duplicate cache is used by the farm workers. See Section 3.7.50. DupCache.
In a ServerFarm environment, where the server farm is being used to load balance using, say, the EAPBALANCE module, the back end servers will receive requests (and potentially duplicate requests) from any of the front end servers in the server farm. This can defeat the normal duplicate detection algorithm. UseContentsForDuplicateDetection causes Client to look only at the contents of the request that would not be altered by passing through a proxy server, allowing duplicates to be reliably detected, even in a server farm. The packet contents used for duplicate detection when UseContentsForDuplicateDetection is set are:
  • RADIUS authenticator
  • User-Name
  • Called-Station-Id
  • Calling-Station-Id
It is essential that this parameter be defined in the Client clauses of back end servers of an EAPBALANCE Server Farm architecture, otherwise duplicate detection will not be performed correctly.