Hi GauriSplunk,
join is the last resort to solve search problems, not the first choice - see docs http://docs.splunk.com/Documentation/Splunk/6.3.0/SearchReference/Join or this https://answers.splunk.com/answers/129424/how-to-compare-fields-over-multiple-sourcetypes-without-join-append-or-use-of-subsearches.html
Your problem here is the value of time is not a number, its a string and therefore Splunk will not do what you expect because it will compare it differently. First you need to remove the t from the time values and convert it to a numeric value:
eval time=tonumber(trim(time, "t"))
Next you can get the two time values into new field depending on the source :
eval user_time=tonumber(trim(like(source, "%user-info%"), "t")) | eval some_time=tonumber(trim(like(source, "%some-info%"), "t"))
And finally use the new time fields to compare them:
your base search here
| eval user_time=tonumber(trim(like(source, "%user-info%"), "t"))
| eval some_time=tonumber(trim(like(source, "%some-info%"), "t"))
| stats count by ipaddress, name, hits, user_time, some_time
| where some_time < user_time
This is un-tested so you probably need to tweak it, but it should give you some hints how it can be done.
cheers, MuS
... View more