AWS cloudwatch alarms are not triggering

Question

I created a free tier Windows Server EC2 instance on AWS. I created alarms for CPU utilization. When the average CPU utilization exceeds 75% for a period of 5 minutes, the high CPU utilization alarms triggers occasionally and I receive an email from AWS.

But when I setup an alarm for disk reads and disk writes for a sum of >=10MB in a period of 5 minutes, I don't receive an email because the alarm for high disk writes and/or high disk reads doesn't trigger.

Can anybody let me know how to make disk read and disk write alarms trigger? What is a "disk read" and what is a "disk write" in AWS?

I tried to generate an alarm by constantly downloading large files on the EC2 instance, but no alarm was triggered.

When I checked the notification section of the AWS console in AWS dashboard, I noted that disk read and disk write charts were constantly showing 0 values. On the other hand, Network in/out alarms are also not triggering. I have set the triggers when 1500kKB data comes into the network and when 1500KB data goes out of the network. The charts in the Monitoring section of the AWS console show some activity while I download some file on the instance but even if I download GBs of data (both disk write and network in occur), no alarm is generated.

Thanks in advance.

Here is a screenshot of the charts for the time when my download of SQL Server 2008 R2 was in progress on my VPS:

Regards

Answer 1

Disk Read Operations and Disk Write Operations only apply to Instance Store Volumes.

Most likely, your EC2 instance is an EBS backed instance.

EBS has its own Cloudwatch metrics and alarms - so, work out which EBS volume your instance is using, and then look at the metrics for that EBS volume. From there, you can look at adding EBS cloudwatch alarms for your volumes.

Here is an overview of the different types of root volumes - EBS vs Instance Store