Can I have a universal forwarder collect data from other universal forwarders and then send that off to the indexer?
So, it will work as an relay for things like syslog, and probably for very simple file monitoring use cases similar to syslog type configurations. It is not intended to work for Splunk2Splunk communication, and you will run into problems with attempting this type of a configuration. Thanks to Steve in Splunk doc's for dropping me an email with the clarifications.
Yes, you can.
There was some controversy about this a while ago, so pardon the confusion, but, yes, that topology should work.
If you're running into issues getting it to work, I suggest you pose a new question in Answers, with the specifics of your configuration.
Hi Steve G,
I am confused by the answer. Do you mean the below topology will not work? I have the latest version 4.2.3.
Program -> Universal forwarder -> Universal Forwarder (Intermediate) -> Main Indexer
'->' the flow of TCP message originated from the Program.
The table says 'Yes' against 'Serves as intermediate forwarder?' for Universal Forwarder.
I am asking this as this setup is not working for me right now. I am not sure if I didn't configure it properly or it does not work.
Here's a chart with the capabilities of all the types of forwarders:
http://www.splunk.com/base/Documentation/latest/Deploy/Typesofforwarders#Forwarder_comparison