Host lacking an application running on port 0

[Jerry32]

Hi!

We're attempting to map one of our services and I'm getting the following error.

Failed to identify CI due to the host 10.x.x.x lacking an application running on port 0. Make sure that the application is running and run quick discovery on the host.

We are able to map from the entry point through the Load Balancer.  All three hosts running this application fail with the same error.

We are able to discover these hosts with no errors and they are showing the processes and ports properly.

Please help me understand why Service Mapping is looking for something on port 0 and how to fix this.

Any help is greatly appreciated!

[Mohit Malik]

Hi Jerry,

I'm also facing the same issue, were you able to fix that?

Thanks,

Mohit

[Jerry32]

Yes!

I worked with ServiceNow support and we added a step in the pattern to filter out ports with 0 as they are not accurate anyway.  Hopefully this helps.

[Tony DiRienzo]

Hi Jerry,

I don't think that is the best solution to this error. I spoke to my organization's F5 team, and they informed me that port 0 is a valid configuration meaning "all ports". So it will use the port from the incoming connection for the outgoing connection. For example, if your incoming connection to the load balancer is http (80), the load balancer will use 80 for the outgoing connection, but if your incoming connection is https (443) it will use port 443.

Since port 0 is a valid configuration, these connections are valid and absolutely should be included in the service map. The solution that I will be using is to replace port 0 with whatever port value is being used on the incoming connection. Clearing out the port value could potentially work as well, since service mapping can infer the port number in some cases (for example, if the protocol is specified as http and the port is blank, service mapping will use port 80).