AWS DevOps Guru - ASG:EC2 CPU Utilization 99% - No Insight Generated
Hi I was doing a POC on AWS DevOps Guru As per the documentation it support analyzing EC2 hosted by an ASG. So I launched an EC2 instance using an Auto Scaling Group and Enabled AWS DevOps Guru for the instance by resource Tag. As per DevOps Guru, it will detect anomalies automatically and generate insight We don't have access to know how DevOps guru working to analyze the EC2 or to set threshold to generate insights
So to test I put 99% CPU load on the EC2 instance. In Metric it was showing the CPU Utilization is 99%, and it last for 4-6 hours (the manual stress) But no Insight generated by DevOps Guru when the CPU is at 99-100% for 4-6 hours. We tried this for 2 weeks daily but same result, no insight generated. We put the EC2 stress by using 'stress' package. (stress demo - https://www.cyberciti.biz/faq/stress-test-linux-unix-server-with-stress-ng/) So how to detect EC2 anomaly via DevOps Guru when CPU is at 99% ? Its now working as per the DevOps Guru documentation.
Any help will be appreciated.
- Newest
- Most votes
- Most comments
Hello there,
I understand that you are testing the DevOps Guru service and to do that you launched an EC2 instance using AutoScaling Group. On this instance, you ran the stress test to put the CPU Utilisation to 99% for 4-6 hours everyday for 2 weeks daily but were not able to find any insight in DevOps Guru.
DevOps Guru applies machine learning to analyze your operational data and application metrics and events to identify behaviors that deviate from normal operating patterns. It is targeted towards finding "unusual pattern" and explain what is going on with it. Now in your case, the instance launched was under high stress from the beginning. So the DevOps Guru expected this load to be there for 4-6 hours on the instance. It became a pattern which DevOps Guru identified and acknowledged.
If the load varied for example, if it was 5%-10% for most of the day and then suddenly it goes to 90% then an anomaly should be recorded.
I request you to test this scenario and let us know if you still face the issue. If it still exists, please share the steps that you followed.
As per the instruction provided we followed the steps. We keep the EC2 (by ASG) CPU utilization 12% for 11 days constantly. On the 12th day we put a stress load which increases the CPU utilization to 99-100% for 6 hours, but AWS DevOps guru does not generate any insight for the anomaly. Amazon DevOps Guru not working as per the documentation.
Relevant content
- asked 3 years ago
- asked 2 months ago
- asked 8 months ago
AWS OFFICIALUpdated a year ago- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated a month ago
- AWS OFFICIALUpdated 3 years ago
Thank you for the suggestion. We have started the analysis as per your suggestion. Now keeping the instance CPU at 25% load. Will keep this for 1 week. After that we will put stress on CPU at 99%. Will revert back to you after the analysis result.
As per the instruction provided we followed the steps. We keep the EC2 (by ASG) CPU utilization 12% for 11 days constantly. On the 12th day we put a stress load which increases the CPU utilization to 99-100% for 6 hours, but AWS DevOps guru does not generate any insight for the anomaly. Amazon DevOps Guru not working as per the documentation.