host name extraction during data upload

[ugoetzen_splunk]

Just trying to manually add data with different host names in the logs. (with the "add data wizard")
What is the best way to extract the host names during the import? (nobud02, appplusf13,...)

Sample data looks like this:
{"host":"nobud02","ident":"dockerd-current","message":"time=\"2017-06-12T18:00:00.745384875+02:00\" level=error msg=\"Handler for POST /containers/weaveproxy/exec returned error: No such container: weaveproxy\"","log_time":"2017-06-12T16:00:00Z"}
{"host":"appplusf13","ident":"systemd","message":"Created slice user-0.slice.","log_time":"2017-06-12T16:00:01Z"}
{"host":"appngu51","ident":"systemd","message":"Created slice user-0.slice.","log_time":"2017-06-12T16:00:01Z"}
{"host":"delme-20170530-rs1","ident":"systemd","message":"Started Session 2594 of user root.","log_time":"2017-06-12T16:00:01Z"}

Thanks in advance.

[somesoni2 · Accepted solution]

YOu'd need to setup props/transforms to override host from raw data. See this for reference.

https://answers.splunk.com/answers/409734/unable-to-override-host-value-using-regex.html

[skoelpin]

You're looking for host_segment

If set to N, the Nth "/"-separated segment of the path is set as host. If
  host_segment=3, for example, the third segment is used.

https://docs.splunk.com/Documentation/Splunk/6.6.2/Admin/Inputsconf

[richgalloway]

I believe host_segment refers to the file path of the data source, not the data itself.