Failed to run Fargate tasks.

[guyShindel]

When running distributed load tests using the solution, executions fail with a “Failed to run Fargate tasks. ” error. The CloudWatch logs show an error during a Lambda “invoke” call with a message:
"errorType":"LimitExceededException","errorMessage":"Resource limit exceeded."
This suggests that one of the service quotas—likely Lambda’s concurrent executions or invoke API rate limit—is being exceeded.

Screenshots

Expected behavior
The load test should either complete successfully or fail with a clear indication of which service quota is exceeded. Ideally, the solution would provide more detailed diagnostics or mitigation advice when such limits are approached.

Despite reviewing CloudWatch metrics, I am unsure which quota is the problem for me and would appreciate any help specifying which quotas it could be.

[kamyarz-aws]

Can you provide more detail on how I can recreate the issue on my end. Maybe script that you can share.

What version of DLT are you using ?

"errorType":"LimitExceededException","errorMessage":"Resource limit exceeded." where do you see this in CW logs for ECS tasks?

Can you check the step function associated with the solution to see what lambda invocation is actually failing?

[guyShindel]

Hi,
Thank you for the fast response.

I’m currently using DLT version 3.2.5
the error only appears in the Lambda invocation logs within the Step Functions execution.

here is the full cause field:

"cause": "{\"errorType\":\"LimitExceededException\",\"errorMessage\":\"Resource limit exceeded.\",\"trace\":[\"LimitExceededException: Resource limit exceeded.\",\"    at Request.extractError (/var/task/node_modules/aws-sdk/lib/protocol/json.js:80:27)\",\"    at Request.callListeners (/var/task/node_modules/aws-sdk/lib/sequential_executor.js:106:20)\",\"    at Request.emit (/var/task/node_modules/aws-sdk/lib/sequential_executor.js:78:10)\",\"    at Request.emit (/var/task/node_modules/aws-sdk/lib/request.js:686:14)\",\"    at Request.transition (/var/task/node_modules/aws-sdk/lib/request.js:22:10)\",\"    at AcceptorStateMachine.runTo (/var/task/node_modules/aws-sdk/lib/state_machine.js:14:12)\",\"    at /var/task/node_modules/aws-sdk/lib/state_machine.js:26:10\",\"    at Request.<anonymous> (/var/task/node_modules/aws-sdk/lib/request.js:38:9)\",\"    at Request.<anonymous> (/var/task/node_modules/aws-sdk/lib/request.js:688:12)\",\"    at Request.callListeners (/var/task/node_modules/aws-sdk/lib/sequential_executor.js:116:18)\"]}",
        "error": "LimitExceededException"
    },

[kamyarz-aws]

that is a relatively old version, have you started seeing this error recently ? what is the concurrent lambda invocation quota in your account? Which lambda is getting this error

Your test is not something that solution would fail for, I have seen this kind of tests daily running from other clients with no problem.

I might need more info most probably about your account to give you more guidance. Right now this doesnt strike me as a solution bug, and it is more of an account limitation.

[kamyarz-aws]

Can you go to your cloudtrail, and set the event history to the past 30 minutes, then run a dlt test. Then if you monitor your cloudtrail api actions, we can see what resource is hitting the limit.

I know it was confusing but that is the error that AWS generates, it was really out of DLT hands.

[guyShindel]

Got it
I’ll investigate and update here once I have more info.

[kamyarz-aws]

It is probably metric filters that are hitting the limit. If you delete older tests, metric filters will be removed as well and your issue would be resolved. We update our documentation to mention this. We discuss it internally and see if we can resolve this or increase the limit and address it in one of the future releases.

[guyShindel]

It looks like it’s working now after deleting all the previous tests—thanks!

Just to confirm, the current versions still don’t address this issue, right?
Besides deleting older tests, how can I ensure I don’t hit this limit again?