I have a distributed Splunk installation with two sites and indexer replication between the two sites such that all data is searchable at both sites. I need to be able to distinguish which site the data came from initially (the original sources) and do not see anywhere where the 'site' designation in the cluster configuration is present in the events. How would I add a static field with the value of the site? I am thinking about a custom field with static values at index time that say 'site1' on the indexers at site1 and 'site2' on the indexers at site2. Has anyone done anything like this? The examples I have see are all based on data by source or host so I am a little unsure of how it would look. Any help is appreciated.
There should be field called 'splunk_server', but don't think that helps with your situation.
You can process data at index time with a transform, there is more info here: http://docs.splunk.com/Documentation/Splunk/6.2.1/Data/Configureindex-timefieldextraction