Splunk Dev

Issue getting multi-value field to correlate properly

burras
Communicator

I have an existing data set that provides a dump of multiple data points in a single event. The data set looks something like this:

HOSTNAME = "NODE1"
License Type: MAIN INSTALLED.Count: 1 Mainboard: 0 I/O 1: 0 MFP 2: 0 Total Licenses in Use: 0
License Type: TRANSRATE INSTALLED.Count: 12 Mainboard: 0 I/O 1: 0 MFP 2: 0 Total Licenses in Use: 9
License Type: EXTENSION is NOT INSTALLED

And the data set continues with about 50 more license types per node. I've been able to set up extraction of various fields (node, license_type, license_avail, license_used) using props.conf and transforms.conf:

props.conf
[license]
EXTRACT-hostname = HOSTNAME=\"(?<node>\w+)\"
REPORT-license_type = license_type
REPORT-license_available = license_available
REPORT-license_used = license_used

transforms.conf
[license_type]
REGEX = License\sType:\s(?<license_type>\w+)\sINSTALLED\.Count
FORMAT = license_type::$1
MV_ADD = true

[license_available]
REGEX = INSTALLED\.Count\s(?<license_avail>\d+)\s+
FORMAT = license_avail::$1
MV_ADD = true

[license_used]
REGEX = \s+Total\sLicenses\sin\sUse:\s(?<license_used>\d+)
FORMAT = license_used::$1
MV_ADD = true

These seem to be working fine - I'm able to extract and see a list of all of the available licenses on the system (i.e. MAIN, TRANSRATE); the count of all available licenses (i.e. 1, 12), and the count of licenses actually in use (i.e. 0, 9). The problem I'm having is correlating this data together in an actual report for the users. I can't seem to make the numbers line up properly. I've tried numerous stats, charts, and other commands, but can't seem to make it line up into what should be a fairly simple report - For each node, list installed license types with a count of licenses available and licenses in use:

Host Type Available Used
NODE1 MAIN 1 0
NODE1 TRANSRATE 12 9

What I've ended up with is usually the same values for each license type or a completely unordered list with no correlation between the counts and the license type. Some of the searches I've used are:

|stats values(license_type) values(license_avail) values(license_used) by node    - returns uncorrelated list of numbers

|stats latest(license_type) latest(license_avail) latest(license_used) by node        - returns visually what I'm looking for but only 1 license type per node

I'm sure there's something pretty simple that I'm missing here but I appreciate any help...

Tags (1)
0 Karma
1 Solution

somesoni2
Revered Legend

Try these

Use list function of stats instead of values. The values function gives sorted , unique values of the field hence the non-correlated ordering

your base search
|stats list(license_type) list(license_avail) list(license_used) by node 

OR, make the report linear by expanding multivalued fields.

your base search
| fields node license_type license_avail license_used
| eval temp=mvzip(mvzip(license_type,license_avail,"##"),license_used,"##")
| mvexpand temp 
| rex field=temp "(?<license_type>[^#]+)##(?<license_avail>[^#]+)##(?<license_used>.+)"
| table node license_type license_avail license_used

View solution in original post

0 Karma

somesoni2
Revered Legend

Try these

Use list function of stats instead of values. The values function gives sorted , unique values of the field hence the non-correlated ordering

your base search
|stats list(license_type) list(license_avail) list(license_used) by node 

OR, make the report linear by expanding multivalued fields.

your base search
| fields node license_type license_avail license_used
| eval temp=mvzip(mvzip(license_type,license_avail,"##"),license_used,"##")
| mvexpand temp 
| rex field=temp "(?<license_type>[^#]+)##(?<license_avail>[^#]+)##(?<license_used>.+)"
| table node license_type license_avail license_used
0 Karma

burras
Communicator

Thanks - using list fixed up the issue. Appreciate the help!

0 Karma
Got questions? Get answers!

Join the Splunk Community Slack to learn, troubleshoot, and make connections with fellow Splunk practitioners in real time!

Meet up IRL or virtually!

Join Splunk User Groups to connect and learn in-person by region or remotely by topic or industry.

Get Updates on the Splunk Community!

Cisco Data Fabric from Architecture to Investigation, Better SOC Visibility, and More ...

Splunk Lantern is Splunk’s customer success center that provides practical guidance from Splunk experts on key ...

The Trust Gap: Why a Data Foundation is Fundamental to an Agentic Enterprise

The Trust Gap: Why a data foundation is fundamental to an  Agentic Enterprise.   Agentic AI is transforming ...

Data Management Digest – September 2026

    Welcome to the September 2026 edition of Data Management Digest! September brought a fresh wave of ...