Question

what to do when monitoring not sending alerts to cpu stress test?

I read through the monitoring tutorial and set up a CPU alert of “above 70%” and 5 minute trigger. I started the stress app and watched with glances to see that the CPU was running at 98%-100%. The alert was not sent via email and the control panel still shows “never” for the CPU alert.

What else do I need to do to make this work?


Submit an answer

This textbox defaults to using Markdown to format your answer.

You can type !ref in this text area to quickly search our full set of tutorials, documentation & marketplace offerings and insert the link!

Sign In or Sign Up to Answer

These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.

Want to learn more? Join the DigitalOcean Community!

Join our DigitalOcean community of over a million developers for free! Get help and share knowledge in Q&A, subscribe to topics of interest, and get courses and tools that will help you grow as a developer and scale your project or business.

@steveloebs

I setup a 2GB / 2 CPU Core Droplet in NY 3, then created a CPU Load alert for usage over 5% for 5 minutes for this test. It didn’t work as I had expected, so this may be something you might want to create a ticket on.

For this test, I used stress which was installed using apt -y install stress and first ran using:

stress --cpu 1

The load shot up to 1.0 and I allowed it to run for 7 minutes – the load increased slightly between the 5 and 7 minute mark – I never received an alert, even though 1.0 is definitely more than 5%.

Since this is a multi-core Droplet, I upped the CPU count for stress and used:

stress --cpu 2

That seems to have worked as the load shot up well past 1.0 and not long after, an alert was send out stating that I was using ~70% of my available CPU.

So when it comes to CPU usage, cores seem to matter in the stress test. The alerts do not seem to fire off if you’re only using 1 CPU core out of multiple, thus you’d have to effectively be using a threaded test or application to trigger the alerts.

I’m not sure if this is by design as this is the first time I’ve used the alerts, but I would recommend putting a ticket in to see if this is the norm.