We have Nimsoft in our environment along with relay/Proxy hubs. Discovery server probe is installed on our primary Hub.
We would like to know how Discovery probe discovers and updates the Robot information in CM_* tables in the back end database..
We have a few robots on a secondary hub that are not showing up in USM and the cm_computer_system table. Why is this happening?
Below is how the discovery_server collects information and populates the CM_computer_system and other tables for about 98% of probes.
1) The discovery_server makes a call to the primary hub and asks for a list of robots.
2) as these robots are attached to the primary the discovery_server assumes it should be able to connect to all of these robots on directly on port 48000 to collect the information stored in the robot niscache.
3) next it asks the primary hub for a list of all hubs and how to contact the. The hub responds with either a direct connection method to the hub on port 48002 or tunnel connection information.
4) if the connection type to a secondary hub is direct connection from the primary hub on port 48002 then the discovery_server assumes that all communication to the robots attached to that hub will be direct connection on port 48000 to each robot.
5) if there is a tunnel to a secondary hub the discovery_server will use the tunnel port and then the secondary hub will make connection to the robots on port 48000 to collect the information on the niscache.
Note: If there is direct connection AND a tunnel to a secondary hub the tunnel will not be used.
If you set the discovery_server loglevel to 5 and deactivate the discovery_server and then activate it you should be able to see the communication attempt between the discovery_server and each robot.
If there is a problem with collecting information you will see a communication error.
If this is seen you will need to correct the underlying connection problem to allow this process to happen.