SageMaker notebook instance running for 7+ days without modification
- Severity
- Medium
- Service
- SageMaker
- Check ID
- SAGEMAKER_ZOMBIE_NOTEBOOK
What this check finds
A SageMaker notebook instance has been in InService state for at least 7 days without any configuration change. Notebook instances charge per hour for the underlying ml.* instance regardless of activity. Instances left running over weekends or between experiments are a common source of avoidable ML spend.
Passing looks like: No idle SageMaker notebooks.
How to fix it
Stop the instance when not actively in use: aws sagemaker stop-notebook-instance --notebook-instance-name INSTANCE_NAME. Add an auto-stop lifecycle configuration so the notebook shuts down after a period of Jupyter kernel inactivity.
AWS console
SageMaker → Notebook instances → select → Stop.
Compliance
This is a cost check. It flags spend that buys nothing, which no compliance framework asks about, so it is not mapped to a control.
Checked on every scan
KloudLytics runs this check each time it scans a connected AWS account, through a read-only role, and lists every affected resource with its region. On Pro and Business a fix is written for the specific resource rather than the general case above. The exact access it needs
More SageMaker checks
- HighSageMaker notebook instance has direct internet access
- HighSageMaker training job has no checkpoint configuration
- MediumSageMaker endpoint not deployed inside VPC
- MediumSageMaker notebook instance has root access enabled
- MediumSageMaker resource not encrypted with CMK
- MediumSageMaker training job not using managed spot instances