Splunk Search

coalesce count

dpolochefm
Explorer

The goal is to get a count when a specific value exists 'by id'.  This is not working on a coalesced search.

The search below works, it looks at two source types with different field names that have the same type of values.  I used this because appendcols is very computation costly and I try to avoid it as much as possible. 

One alternative I tried was:

count(eval(if(isnotnull(calcValue),1,null())))

Lastly I tried, however that one gives an error:  Error in 'stats' command: The eval expression for dynamic field 'if("total-calcValue">0,1,null())' is invalid. Error='Type checking failed. The '>' operator received different types.'.

 

count(eval(if("total-calcValue">0,1,null())))

 

Below is the full search

 

index=someindex (sourcetype1)  OR (sourcetype2) 
| rex field="Some Amt" "-(?<Amount>[\d\.]+)"
| convert num(Amount)
| rename Amount as total-calcValue, total as total-charges, "Some ID" as miq, "Other Source Company Name" as "other_source_company_name"
| eval mid=coalesce(mid,miq)
| eval Company=coalesce(other_source_company_name,company_name)
| stats count first(Company) AS "Company" sum(total-*) AS * count(eval(if(isnotnull("total-calcValue"),1,null()))) as "calcValue Count" by mid

 

Labels (3)
0 Karma
1 Solution

dpolochefm
Explorer

The issue was that 

"total-calcValue"

needed to be changed to

'total-calcValue'

 It was taking the double quote version as a literal text value instead of the field..

View solution in original post

dpolochefm
Explorer

The issue was that 

"total-calcValue"

needed to be changed to

'total-calcValue'

 It was taking the double quote version as a literal text value instead of the field..

richgalloway
SplunkTrust
SplunkTrust

Using hyphens in field names is not a great idea.  Stick with underscores or use camelCase.  Introducing other characters may be valid, but can lead to errors (as you've seen) or confusion (is "total-calcValue" a string, an expression, or something else?).  Often, putting single quotes around the field will help, but not always.

Try this modification of your query.

index=someindex (sourcetype1)  OR (sourcetype2) 
| rex field="Some Amt" "-(?<Amount>[\d\.]+)"
| convert num(Amount)
| rename Amount as total_calcValue, total as total_charges, "Some ID" as miq, "Other Source Company Name" as "other_source_company_name"
| eval mid=coalesce(mid,miq)
| eval Company=coalesce(other_source_company_name,company_name)
| stats count first(Company) AS "Company" sum(total_*) AS * sum(eval(isnotnull(total_calcValue))) as "calcValue Count" by mid

 

---
If this reply helps you, Karma would be appreciated.
0 Karma
Get Updates on the Splunk Community!

Splunk Enterprise Security: Your Command Center for PCI DSS Compliance

Every security professional knows the drill. The PCI DSS audit is approaching, and suddenly everyone's asking ...

Developer Spotlight with Guilhem Marchand

From Splunk Engineer to Founder: The Journey Behind TrackMe    After spending over 12 years working full time ...

Cisco Catalyst Center Meets Splunk ITSI: From 'Payments Are Down' to Root Cause in ...

The Problem: When Networks and Services Don't Talk Payment systems fail at a retail location. Customers are ...