After enabling Amazon integrations with New Relic Infrastructure, you encounter a rate limit for service-specific APIs. You might see this message in your AWS monitoring software, often with a
AWS::EC2::Errors::RequestLimitExceeded Request limit exceeded.
- Verify your Infrastructure account's ARN
- Ensure that you are not collecting inventory information for the wrong ARN account. Verify that the ARN associated with your Infrastructure account is correct.
- Change the polling frequency
- The polling frequency determines how often New Relic gathers data from your cloud provider. By default, the polling frequency is set to the maximum frequency that is available for each service. If you reach your API rate limit, you may want to decrease the polling frequency.
- Filter your data
- You can set filters for each integration in order to specify which information you want captured. If you reach your API rate limit, you may want to filter your data.
- Review API usage
To review the API usage for New Relic Infrastructure integrations with Amazon AWS:
- Go to one.newrelic.com > Infrastructure > AWS > Account status dashboard.
- Review the New Relic Insights dashboard, which appears automatically.
The Insights dashboard includes a chart with your account's Amazon AWS API call count for the last month as well as the CloudWatch API calls (per AWS resource) for the last day. This information is the API usage for New Relic only. It does not include other AWS API or CloudWatch usage that may occur.
For assistance determining which services may cause an increase in billing, get support at support.newrelic.com, or contact your New Relic account representative.
Infrastructure Amazon integrations leverage the AWS monitoring APIs to gather inventory data. AWS imposes hard rate limits on many of the AWS service-specific APIs consumed by New Relic Infrastructure integrations. Adding New Relic Amazon integrations will increase usage of the service-specific APIs and could impact how quickly you reach your rate limit.
This may be caused by either of the following:
- Enabling Amazon integrations on several plugins for the same service
- Adding the incorrect Role ARN to your AWS integrations