Counting occurences in aggregation by fields

[RonaldCWWong]

Hi community,

I have a question on counting the number of events per values() value in stats command.
For example having events with src_ip, user (and a couple of more) fields.
I would like to count each of the user occurence in the raw log. Example as below.

 | stats values(user) as values_user by src_ip 

Example:

_time	user	src_ip
2025-08-11	ronald	192.168.2.5
2025-08-11	jasmine	192.168.2.5
2025-08-11	tim	192.168.2.6
2025-08-11	ronald	192.168.2.5

I would like to have result as 

values_user	count_vaules_user	src_ip
ronald jasmine	ronald:2 jasmine:1	192.168.2.5
tim	tim:1	192.168.2.6

[bowesmana]

Just use a couple of stats, first count the user numbers then create a new field with the user and count then re-stats with the values, e.g.

| makeresults format=csv data="_time,user,src_ip
2025-08-11,ronald,192.168.2.5
2025-08-11,jasmine,192.168.2.5
2025-08-11,tim,192.168.2.6
2025-08-11,ronald,192.168.2.5"
``` Like this ```
| stats count by user src_ip
| eval user_count=user.":".count
| stats values(user*) as values_user* by src_ip

[RonaldCWWong]

thanks for help 🙂

[PrewinThomas · Accepted solution]

@RonaldCWWong 

If I understood you correctly, you want to group by src_ip, list all the distinct user values per IP, and also count how many times each user appears for that IP.

try below,

...your base search...
| stats count by src_ip, user
| eventstats sum(count) as user_count by src_ip, user
| eval user_count_pair = user . ":" . user_count
| stats values(user) as values_user values(user_count_pair) as count_values_user by src_ip

Regards,
Prewin
Splunk Enthusiast | Always happy to help! If this answer helped you, please consider marking it as the solution or giving a Karma. Thanks!

[RonaldCWWong]

thanks mate, this is exactly what I am looking for 🙂