Splunk Search

Sum value from distinct hosts over time

caviman2201
Path Finder

I have a bunch of hosts streaming logs that show metrics like cpu count. The problem is that they all send those logs at different times. I'm trying to calculate perform math on the aggregate metrics, but everything is wrong because I can't figure out how to perform math on the values from distinct hosts. I've searched for deduping on a span and got nowhere. Logs look something like this:

1:00:01 - ip=192.168.1.5 cpu_count=8
1:00:01 - ip=192.168.1.6 cpu_count=8
1:00:02 - ip=192.168.1.7 cpu_count=8
1:00:03 - ip=192.168.1.8 cpu_count=8
1:00:03 - ip=192.168.1.7 cpu_count=8
1:00:03 - ip=192.168.1.5 cpu_count=8
1:00:03 - ip=192.168.1.6 cpu_count=8
1:00:04 - ip=192.168.1.7 cpu_count=8
1:00:04 - ip=192.168.1.8 cpu_count=8

what i want to do is something like | dedup span=1m by ip | eval TotalPercent = cpu_count * 100 | timechart sum(TotalPercent) AS TotalAvailableCPUPercent

hope that makes sense

Tags (1)
0 Karma
1 Solution

Ayn
Legend

You probably want bucket for grouping your events by time and deduping on that.

... | bucket _time span=1m | dedup _time ip | eval TotalPercent = cpu_count * 100 | chart sum(TotalPercent) AS TotalAvailableCPUPercent over _time

View solution in original post

0 Karma

Ayn
Legend

You probably want bucket for grouping your events by time and deduping on that.

... | bucket _time span=1m | dedup _time ip | eval TotalPercent = cpu_count * 100 | chart sum(TotalPercent) AS TotalAvailableCPUPercent over _time
0 Karma
Get Updates on the Splunk Community!

SOC4Kafka - New Kafka Connector Powered by OpenTelemetry

The new SOC4Kafka connector, built on OpenTelemetry, enables the collection of Kafka messages and forwards ...

Your Voice Matters! Help Us Shape the New Splunk Lantern Experience

Splunk Lantern is a Splunk customer success center that provides advice from Splunk experts on valuable data ...

Building Momentum: Splunk Developer Program at .conf25

At Splunk, developers are at the heart of innovation. That’s why this year at .conf25, we officially launched ...