Splunk Search

Where and or statement for multiple values

codedtech
Path Finder

Hello,

I'm building a search that tracks the use of memory allocated(mem_alloc), memory in use(mem_used), CPU in use(CPU_used) CPU allocated (CPU_alloc) along with the Cluster type (CT) for our vmware envrioment. My base query works perfect but when I try to build the logic behind it using where, and , or. I keep getting false positives, like the example below.

where CT=Tier_1 or CT=Tier_2 and CPU_used>=50 or mem_used>=50  or mem_alloc>=0.9 or CPU_alloc>=0.9 

I end up getting results that have different cluster tiers in them

|CT | |CPU_used| |mem_used| |mem_alloc| |CPU_alloc|
|Tier_2 |50.01| |25.35| |.82 | |.82 |
|Tier_1 |62.23| |72.33| |.90 | |.65 |
|Tier_2 |45.53| |32.55| |.97 | |.55 |
|Tier_4 |23.25| |36.58| |1.01| |3.25|
|Tier_5 |40.32| |85.15| |3.25| |1.11|
I need to find a way to be able to narrow it down to Tier_1 or Tier_2 clusters only, and alert if any of the values break the thresholds in the where statements.

0 Karma

mayurr98
Super Champion

try this?

where (CT=Tier_1 or CT=Tier_2) AND (CPU_used>=50 or mem_used>=50  or mem_alloc>=0.9 or CPU_alloc>=0.9) 
0 Karma

codedtech
Path Finder

That worked, how would I repeat the process for Tier_3 where the values are different from Tier_1 and Tier 2 in the same query?

0 Karma

mayurr98
Super Champion

try:

((CT=Tier_1 or CT=Tier_2) AND (CPU_used>=50 or mem_used>=50  or mem_alloc>=0.9 or CPU_alloc>=0.9)) OR (CT=Tier_3 AND ( cpu_used>50 OR mem_used>60...))

Please upvote/accept the answer if it works for you.

0 Karma
Get Updates on the Splunk Community!

Developer Spotlight with Brett Adams

In our third Spotlight feature, we're excited to shine a light on Brett—a Splunk consultant, innovative ...

Index This | What can you do to make 55,555 equal 500?

April 2025 Edition Hayyy Splunk Education Enthusiasts and the Eternally Curious!  We’re back with this ...

Say goodbye to manually analyzing phishing and malware threats with Splunk Attack ...

In today’s evolving threat landscape, we understand you’re constantly bombarded with phishing and malware ...