If you’re experiencing issues with your AWS services, then please refer to the AWS Health Dashboard. You can find the overall status of ongoing outages, the health of AWS services, and the latest updates from AWS engineers.
Why can I view the invocation metrics but not the instance metrics for my SageMaker AI endpoint in CloudWatch?
I can view invocation metrics for my Amazon SageMaker AI endpoint in Amazon CloudWatch. But I can't view instance metrics, such as CPUUtilization, MemoryUtilization, and DiskUtilization.
Resolution
To view instance metrics, you must include cloudwatch:PutMetricData in the AWS Identity and Access Management (IAM) policy that's attached to the execution role. For more information, see How to use SageMaker AI execution roles.
The execution role for the SageMaker AI endpoint must have the PutMetricData permission for CloudWatch.
If you use a custom container, then your algorithm must use stdout and stderr to write the metrics and logs that you want to capture. SageMaker AI sends the same metrics and logs to Amazon CloudWatch Logs that model containers send to stdout or stderr.
Related information
Metrics for monitoring Amazon SageMaker AI with Amazon CloudWatch
AWS Identity and Access Management for Amazon SageMaker AI
Log groups and streams that Amazon SageMaker AI sends to Amazon CloudWatch Logs
- Tags
- Amazon SageMaker
- Language
- English
Relevant content
- asked a year ago
- asked 3 years ago
- asked 3 years ago
