BigFix agent communications

(imported topic written by SystemAdmin)

Hello,

I have a couple of questions with regards to the agent communications on machines with multiple NICs/IP addresses:

  1. How does the BigFix agent determines which NIC/IP address to use for it’s communication? Is it based on the binding order of the connections on the particular machine?

  2. Is that selection dynamic (e.g. if I have a machine with 3 NICs/IP addresses and I unplug the one that is being used will the BigFix agent pick the second one and use it to continue to communicate with the BES server?) and configurable (e.g. can I specify a particular setting/flag during the agent install or post-install that will preset that value?)?

  3. Can you point me to some relevance code that will be useful to simulate the BF agent nic selection routine on a server to produce a retrieved property?

Thanks in advance.

(imported comment written by BenKus)

Hi ageorgiev,

The short answer is that the agent has no control and no idea which adapter/IP it is using… All this functionality is provided by the operating system. The BigFix Agent simply makes the appropriate API calls to say “I want to make a TCP connection to address ‘blah’” and the OS takes care of the interaction… It is the same process if you open your browser and type in the address.

Ben

(imported comment written by SystemAdmin)

Thank you for the feedback, Ben.