ECS RunTask state machine (consumes ECS RunTask cloud trail events) - reduce scope to handle ECS RunTask requests that failed to place
Parse compute environment and job queue name from the request parameter system created tags (aws:batch:compute-environment and aws:batch:job-queue respectively)
Remove steps after SQS message has been sent as the metrics after a job is successfully placed are now captured by the ECS TaskEvents state machine
Existing RunTaskToCW Lambda will continue to poll from this SQS queue
Remove references to job id as this information is no longer available in the Cloudtrail event
ECS TaskEvents state machine (consumes ECS Task state change events) - add scope to handle ECS RunTask requests that successfully place
Add ECS Task State change event source and SQS queue
Existing RunTaskToCW Lambda will continue to poll from this new SQS queue
Add state machine to parse metadata about the placed job from the ECS Task State change event. This effectively replaces the steps that were removed in ECS RunTask state machine.
Deployed the commit to my personal account (using sam deploy) and ran a series of jobs to verified that the ECSRunTask and ECSTaskEvents state machines execute successfully; and Batch Jobs Placement dashboard renders the API calls after an array job was submitted to an EC2 CE using Batch service linked role. I have used a single JQ associated with a single CE.
Batch jobs placement dashboard
Execution logs from ECSRunTask state machine execution
Issue #, if available: #20
Description of changes:
Docs
Testing
sam deploy
) and ran a series of jobs to verified that the ECSRunTask and ECSTaskEvents state machines execute successfully; and Batch Jobs Placement dashboard renders the API calls after an array job was submitted to an EC2 CE using Batch service linked role. I have used a single JQ associated with a single CE.Batch jobs placement dashboard
Execution logs from ECSRunTask state machine execution
Execution logs from ECSTaskEvents state machine execution
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.