View a markdown version of this page

Monitor Amazon ECS container instance health - Amazon Elastic Container Service
Container instance-health issues

Monitor Amazon ECS container instance health

Amazon ECS provides container instance health monitoring. You can quickly determine whether Amazon ECS has detected any problems that might prevent your container instances from running containers. Amazon ECS performs automated checks on every running container instance with agent version 1.57.0 or later to identify issues. For more information on verifying the agent version an a container instance, see Updating the Amazon ECS container agent.

You must be using AWS CLI version 1.22.3 or later or AWS CLI version 2.3.6 or later. For information about how to update the AWS CLI, see Installing or updating the latest version of the AWS CLI in the AWS Command Line Interface User Guide Version 2.

To view the container instance health, run describe-container-instances with the CONTAINER_INSTANCE_HEALTH option.

The overallStatus is determined by the individual health check statuses in the details array. The most severe status takes precedence in the following order: IMPAIRED, INSUFFICIENT_DATA, INITIALIZING, and OK.

The following are the valid values for overallStatus:

  • OK – All health checks are passing.

  • IMPAIRED – One or more health checks have failed.

  • INSUFFICIENT_DATA – Data for one or more health checks is unavailable.

  • INITIALIZING – One or more health checks are being initialized.

The health check type values include:

  • CONTAINER_RUNTIME – Monitors the health of the container runtime (for example, the Docker daemon). Available on EC2 launch type.

  • ACCELERATED_COMPUTE – Monitors the health of accelerated compute devices such as GPUs. When a GPU is impaired, the statusReason contains the NVIDIA Xid error code in the format XID_<number>. Available on Amazon ECS Managed Instances launch type.

  • DAEMON – Monitors the health of required daemon tasks on the container instance. Available on Amazon ECS Managed Instances launch type.

You can also monitor health status changes through events. For more information, see Amazon ECS container instance health change events.

The following is an example of how to run describe-container-instances.

aws ecs describe-container-instances \
     --cluster cluster_name \
     --container-instances 47279cd2cadb41cbaef2dcEXAMPLE \
     --include CONTAINER_INSTANCE_HEALTH

The following is an example of the health status object in the output.

"healthStatus": {
	"overallStatus": "OK",
	"details": [{
		"type": "CONTAINER_RUNTIME",
		"status": "OK",
		"lastUpdated": "2021-11-10T03:30:26+00:00",
		"lastStatusChange": "2021-11-10T03:26:41+00:00"
	}]
}

Container instance-health issues

When the overallStatus any status other than OK, try the following:

  • Wait, and then run describe-container-instances

  • View your container instance health in the EC2 console or by using the CLI.

  • Review the CloudWatch metrics. For more information, see Monitor Amazon ECS using CloudWatch

  • Check the AWS Health Dashboard to see if there are any issues with the service.