topic How to get multiple distinct counts for subset of logs in single query? in Splunk Search

How to get multiple distinct counts for subset of logs in single query?

Damek — Mon, 07 Nov 2022 19:16:36 GMT

Hello, I am currently using the |append method for some queries, but was curious if there is a better way for me to be writing these? We are trying to create a single alert that could be triggered by various conditions such as total number of failures or total number of unique customer failures. The following is a simplified example of what I am currently doing and would like to improve if anyone knows how:

"base query stuff" | stats count as TOTAL count(eval(SEVERITY="INFO")) as SUCCESS count(eval(SEVERITY="SOAP-FAULT")) as FAULT count(eval(SEVERITY!="INFO" AND SEVERITY!="SOAP-FAULT")) as ERROR | append [search "base query stuff" SEVERITY="SOAP-FAULT" | stats dc(userId) as UNIQUE_FAULT] | where UNIQUE_FAULT > 10 OR FAULT > 20 OR ERROR > 30

I would also love to be able to create a table with all of this data (hence the success variable), which contains the totals of each and unique customer impacts of each!

Re: How to get multiple distinct counts for subset of logs in single query?

kamlesh_vaghela — Tue, 08 Nov 2022 06:07:11 GMT

@Damek

If your base search is common for both searches then you can try this.

"base query stuff" | eval SOAP_FAULT_userId = if(SEVERITY=="SOAP-FAULT",userId,null()) | stats count as TOTAL count(eval(SEVERITY="INFO")) as SUCCESS count(eval(SEVERITY="SOAP-FAULT")) as FAULT count(eval(SEVERITY!="INFO" AND SEVERITY!="SOAP-FAULT")) as ERROR dc(SOAP_FAULT_userId) as Splunk

My Sample Search :

| makeresults | eval _raw="SEVERITY,userId INFO,1 INFO,1 SOAP-FAULT,1 SOAP-FAULT,2 INFO1,1 INFO2,1 OTHER_INFO,1 INFO,1 SOAP-FAULT,1 " | multikv forceheader=1 | table SEVERITY userId | rename comment as "upto this is sample data" | eval SOAP_FAULT_userId = if(SEVERITY=="SOAP-FAULT",userId,null()) | stats count as TOTAL count(eval(SEVERITY="INFO")) as SUCCESS count(eval(SEVERITY="SOAP-FAULT")) as FAULT count(eval(SEVERITY!="INFO" AND SEVERITY!="SOAP-FAULT")) as ERROR dc(SOAP_FAULT_userId) as Splunk

I hope this will help you.

Thanks
KV
If any of my replies help you to solve the problem Or gain knowledge, an upvote would be appreciated.

Re: How to get multiple distinct counts for subset of logs in single query?

bowesmana — Tue, 08 Nov 2022 06:46:09 GMT

In your first stats you can do the dc with this

dc(eval(if(SEVERITY="SOAP-FAULT", userId, null))) as UNIQUE_FAULT