Service Unavailable. Please try again later. (Service: AmazonECS; Status Code: 500; Error Code: ServerException; Request ID: ab4946dd-0492-4e95-bcc3-35126b76396b; Proxy: null)

The "Service Unavailable" error you're encountering with Amazon ECS (Status Code: 500, Error Code: ServerException) indicates a temporary issue with the ECS service itself. This is different from client-side errors (400-level) and suggests that the AWS service is experiencing temporary unavailability.

For the additional timeout errors related to network interface and EphemeralStorage provisioning, these are likely related to resource creation timing issues during your deployment.

Here are some recommended steps to resolve these issues:

1. For the Service Unavailable error:

• This is a transient server-side issue, so the simplest solution is to wait and retry your operation later.
• If you're using CloudFormation or another infrastructure as code tool, it may automatically retry the operation.

2. For network interface and EphemeralStorage provisioning timeouts:

• These errors typically occur when there are timing issues during resource creation.
• If you're using CloudFormation, consider adding explicit dependencies using the DependsOn attribute to ensure resources are created in the proper sequence.
• For network interfaces specifically, ensure that the instance is fully created before attempting to attach or configure network interfaces.
• You may need to increase timeout settings if available in your deployment tool.

If these issues persist:

• Check the AWS Service Health Dashboard to see if there are any known service disruptions.
• Review your IAM permissions to ensure your role has all necessary permissions for ECS operations.
• If you're creating an ECS cluster, verify that the AWSServiceRoleForECS service-linked role exists or can be created in your account.

These types of errors are often temporary, and retrying the operation after a short wait is frequently the most effective solution.
Sources
How do I troubleshoot CloudFormation error "Error occurred during operation 'CreateCluster SDK error: Service Unavailable. Please try again later."? | AWS re:Post
Resource creation cancelled Resource handler returned message: "The pending-instance-creation instance to which 'eni is attached is not in a valid state for this operation | AWS re:Post
Common Errors - Amazon Lookout for Metrics