Handling Unsupported Metrics for AWS Quota Monitoring

[arthosdevelops]

Hello,
Thanks for providing a simplified version of the official AWS solution for monitoring and alerting on service quotas! It’s a great resource.
I have a question regarding how unsupported metrics are handled in your setup. For example, metrics like “Interface VPC endpoints per VPC” are not available in either Trusted Advisor or CloudWatch. In this case, did you implement custom scripts to track such metrics, or were those metrics excluded from your monitoring altogether? As of now I am considering a custom lambda based approach for metrics needed for our use case which are not supported, but curious to hearing how you approached this scenario!

[max-rocket-internet]

Hi @arthosdevelops

In this case, did you implement custom scripts to track such metrics, or were those metrics excluded from your monitoring altogether?

They are excluded. I am not writing any custom code here for deducing quota usage. The code here just uses the available metrics from AWS, via Cloudwatch 🙂

[max-rocket-internet]

I am considering a custom lambda based approach for metrics needed for our use case which are not supported

It's a lot of work IMO and would only work for quotas where you can see the quota limit value (many you cannot, e.g. API call rates). Some other projects you might want to take a look at:

• https://github.com/aws-solutions/quota-monitor-for-aws
• https://github.com/sebasrp/awslimitchecker

Also note this comment here: aws-solutions/quota-monitor-for-aws#198 (comment)

Good luck 💙