Troubleshooting and ErrorsFargate Task Storage

Troubleshooting: Fargate Task Storage Errors

After deploying a fargate or fargate-worker service, you may see the following error:

[
  {
    "type": "FARGATE_DEPLOY_ERROR",
    "message": "Cannotpullcontainererror: ref pull has been retried 1 time(s): failed to extract layer: <layer_id> no space left on device: unknown"
  }
]

Solution

You may need to increase the storage for the ECS Fargate task. By default, AWS assigns 20GB for each task, but can increase this value up to 200GB. For more details about Fargate task storage, see the AWS Fargate task storage page.

In order to change the task storage for the service:

  • If you are using flightcontrol.json for configuration, add storage: 100 property to the configuration file. This will configure the Fargate task with 100GB of storage. You can adjust the number as needed.
  • If you are using the Flightcontrol dashboard (GUI), you can increase the storage under the service’s Instance configuration settings. Instance configuration settings for a service in the Flightcontrol dashboard