Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials

Linux script monitor thresholding

We have a perl script that executes on a Linux system and returns 9 pair values. The current set-up of the Linux Script monitor appears to allow only a single Greater Than / Less than / Equal to statement per value (e.g., if value pair 1 is less than 5 it is in Warning, if it is less than 1 it is in Critical.) I need to set it up so that if value pair 1 is greater than 30 AND value pair 9 is less than 1, then it is critical. How can I do this?

Find more posts tagged with

linux_script

Perl

Accepted answers

All comments

aLTeReGo

You have two options available to you. As part of the script you can perform this logic and based upon your exit code the component monitor will return a normal, warning, or critical status. The other option is to create an Advanced Alert that triggers based upon this logic as you've defined it in the trigger condition.

ttl

Perhaps I am not being clear. There are 9 script output pairs in our SAM perl script. Using SAM, you can determine on each returned output what the Warning and Critical thresholds should be. However, there is no way to say "if Output pair 2's value is over 30, and output pair 9's value is < 1, then I want to generate an alert". You can only have the Component in Warning / Critical based on the value of one pair.

Example:

Appserver output pair values

1 = Busy

2 = Available

3 = Sending

4 = Receiving

5 ...

9 = Total application servers

IF pair2 < 1 AND pair9 = 30 THEN (generate an alert).

That's the problem I'm having -- I need dependencies on the two pairs. If one is = 30, and the other is < 1, then I have a problem. It doesn't look like the SAM does that currently.

lukas.belza

Hello,

You are right that SAM can't set the component status based on thresholds for two and more pairs. But you can remove all thresholds and use exit code of the script to set the status of the component.

Exit codes:

0 = Up

1 = Down

2 = Warning

3 = Critical

I mean modify your script to handle the conditions for setting warning/critical status directly in your script. For example:

if ($condition_for_warning)

{ $status_code = 2 }

elseif ($condition_for_critical)

{ $status_code = 3 }

else

{ $status_code = 0 }

exit($status_code)

I hope it helps.

Lukas Belza (SolarWinds development)

familyofcrowes

I end up using the alerting engine to handle this for me. In the alert I say statistic.1 is greater than 5 AND statistic.4 is greater than 15.

Then my application is only pulling data and my alert determines the severity. My havin an action to display on the console, I now get a red event with whatever directions I want the Operations guys to perform.

We have made this the norm for more than 500 monitors.... works great so far.

ttl

Thanks. How would that alert look? I'm looking through the APM:Component properties, is it something like this?

Trigger Alert when all of the following apply:

Statistic Data Message is equal to Total number of appservers

Statistic Data (Numeric) is equal to 30

Statistic Data Message is equal to Number of appservers AVAILABLE

Statistic Data (Numeric) is less than 1

(if so, I'm not certain how it's matching the Data Message in the first requirement to the Numeric value in the second requirement)

aLTeReGo

Based on the logic as listed above all conditions defined must be true before the alert will fire. This is how the relationship between the two properties is made.

ttl

I understand the "all" logic, what I don't understand is how it knows that I mean that "Statistic Data (Numeric)" refers to the line above it and not ANY Statistic Data (Numeric). If it does reference the line above some kind of nesting would be nice to see...