Splunk Search

spath outputs wrong percentage for fields in inner jsons

New Member


It look like spath calculates its percentage based on the number of available events instead on the number of occurrences of the checked value. For example if I have the next event (did not include the metadata):

{"test": [{"t": 1}, {"t": 2}, {"t": 2}]}

spath will say that the interesting field test{}.t consists of 2 values and that the value 2 appears in 200% of events (value 1 appears in 100%). This is a very confusing, why not check it against the number of the occurrences of test{}.t (Perhaps there is a way to do it and I missed it).

0 Karma

| makeresults 
| eval _raw="{\"test\": [{\"t\": 1}, {\"t\": 2}, {\"t\": 2}]}"
| spath
| stats count by test{}.t

HI, @electronicsplunk
How about this?

0 Karma