CDM custom alarm shows wrong free space values in %

Products

DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

CDM probe is giving wrong values in percentage.

We are monitoring servers and created a custom alarm for alarming free space values. In our case, we have added the $free_gb variable to the CpuError default message.

The issue does not occur all the time.

Example 1:

So:

(152.9*100)/1025.9= 14.90 % --> but showing 17.61%

(187.7*100)/1025.9= 18.29 % --> but showing 17.61%

We are seeing a discrepancy between the value of $free_gb in comparison to the actual free space observed in the probe GUI and in the DiskManagement in the server and the same default "Disk Usage (%)" printed in the same message.

Example 2:

On one interval: --> 15.5GB shows 13.63%

But on another interval, --> 15.5GB is also showing 15.50%

Why is this happening and why can we fix this?

Environment

Release: UIM 20.4.x, CDM all versions

Cause

Configuration issue

The alarm of concern is a customized alarm, and it is adding an additional value to the out-of-the-box alarm. The out-of-the-box alarm works fine, while only the customized alarm presents this discrepancy in the below scenario.

Original out-of-the-box alarm for Disk Error:

Average ($value_number samples) disk free on $drive is now $value$unit, which is <= error threshold ($value_limit$unit) out of total size $size_gb GB

Modified Alarm:

Hi Team , Disk free on $drive is now  $free_gb GB ($value$unit), which is <=  threshold ($value_limit$unit) out of total size $size_gb GB. Kindly take an action ASAP

So you have added a value ($free_gb)

If you used the basic alarm you would NOT see this issue.

Why is this happening:

This disk monitor "Disk Usage (%)", is designed to monitor the disk usage, in percentage, issues its value ($value$unit) - where $value = the disk usage $unit=% and not the ($free_gb)

THE VALUE ($value$unit) which is the value that is used for the threshold calculation always corresponds to the disk Management Disk Usage (% Free) you can see in the disk management at the moment of the alarm.

When you add ($free_gb) which is a different value this depends on the Control Properties

So essentially the reason why the values are "wrong" is because there is a discrepancy between the $free_gb (taken every average determined by the controller prproperties VS the free % (taken every poll).