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
Store

CPU Alerting

I am looking help with CPU Alerting. I wish to create an alert based on the average usage of the CPU and if get a change. We currently set a standard threshold for all servers but have sensitive servers where we need to monitor the baseline CPU and if we get any changes/spikes. Any ideas on how to create an alert on this data.

Find more posts tagged with

Accepted answers

All comments

adatole

Are you familiar with the built-in baselines and alerting based on state (warning, critical) versus a fixed number? Or is that new information for you?

wandie64

No I am not. I have been looking how to do that but not been successful so far. I have looked at Perfomance counters etc.

adatole

On the node details page, you can click "edit properties" and see a screen that lists out CPU, RAM, etc and lets you set a fixed threshold, or use baselines and then set critical/warning based on that.

This episode of SolarWinds lab: All About Alerts - SolarWinds Lab #42 - YouTube

And this Lab Bit: SolarWinds Lab Bits: Using Baseline Data for Alert Trigger Thresholds - YouTube

Should get you moving in the right direction.

jpaluch

I was wondering if there any plans to implement dynamic baseline thresholds that would change in time? What I mean is if we take one server and assume that noone uses it at night (from 10 PM to 6 AM) then Threshold for Critical CPU usage should be set to something around 20%, however during the day (from 6 AM to 10 PM) the same threshold should dynamically change to 90%. Of course granularity should be adjustable here (shorter time periods, etc.)

I know that I could configure manually it using two different alerts and schedule them to run during specified time of the day, but if we would like to change the threshold more frequently then we need to create more alerts and managing them becomes troublesome. It would be nice if the "baseline tool" offered such dynamic approach instead of static one.

wandie64

This has been very helpful and definitely pointing me the right direction. Is there a way to reconfigure all servers to use baseline data or does this need to be done one at a time.

adatole

Once again, I know that aLTeReGo will correct me if I mis-state anything: The baselines calculated from Node data (CPU, RAM, etc) will change over time, but they do use the aggregate data. So if you have a box that is hot during the day and cold at night, you're going to get an amalgamation of that. But as time goes on, so does the baseline calculation.

OTOH, SAM components calculate the baseline when you enable it, and in order to update you have to re-calculate manually. The specific process is:

Thresholds are initially calculated as soon as enough baseline data has been gathered.

Once complete, thresholds are calculated on demand - they are not re-calculated automatically.

Re-calculating thresholds can be done within an Application Monitor. To re-calculate a threshold, perform the following steps:

Navigate to the Edit Application page of the desired Application Monitor.
Find and expand the Component Monitor that have the threshold(s) you want to re-calculate.
Click Use Latest Baseline Thresholds.