Is there anyway to analyze trans data in SplunkStorm?
Here is what I have:
transaction is defined by beginTour and EndTour by user_id
Within a transaction, there could be any number of activties (events) taken by user_id
I want to be able to average distinct number of activities by user_id when taking tours ( a user_id may have many transactions/tours, so each transaction having distinct number of activities, then averaging that dc(activites) number accross transactions by user_id).
Also, is it possible to calculate the avg time spent on each event by user_id for each tour?
Is there anyway to define transaction within transaction and be able to add a field to the outer trans for avg inner trans duration?
even though the extract caused the fields to be recognized by Splunk and the user_id under interesting fields shows up with 14 values, still when I do the following, it only comes back with user NULL and one avg value. I was hoping to get avg(duration) calculated for each user_id based on the trans duration values:
source=xxxx | transaction user_id keeporphans=f maxspan=-1 maxpause=-1 startswith="BeginTour" endswith="EndTour" mvraw=t delim="," mvlist=user_id |
extract pairdelim=",", kvdelim=":" | stats avg(duration) AS avg_dur by user_id
Thanks
Found out about mvlist and all the eval functions for mvlist and got it to work.
Found out about mvlist and all the eval functions for mvlist and got it to work.
What do you get back from your query? Does just source=xxxx | transaction user_id keeporphans=f maxspan=-1 maxpause=-1 startswith="BeginTour" endswith="EndTour"
work?