AWS limit exceeded not bubbled up to console

[JonCubed]

I'm currently getting the following in the console

Error applying plan:

1 error(s) occurred:

* aws_kinesis_firehose_delivery_stream.kinesis_stream: timeout while waiting for state to become 'success' (timeout: 1m0s)

Took a while to find the cause but in the log file, it was showing that the request failed due to a bad request, because I have exceeded my firehose stream limit.

plugin: terraform.exe: aws-provider (internal) 2017/01/11 21:55:01 [DEBUG] [aws-sdk-go] DEBUG: Response firehose/CreateDeliveryStream Details:
plugin: terraform.exe: ---[ RESPONSE ]--------------------------------------
plugin: terraform.exe: HTTP/1.1 400 Bad Request
plugin: terraform.exe: Connection: close
plugin: terraform.exe: Content-Length: 1116
plugin: terraform.exe: Content-Type: application/x-amz-json-1.1
plugin: terraform.exe: Date: Wed, 11 Jan 2017 10:55:02 GMT
plugin: terraform.exe: X-Amz-Id-2: Tq1q13oZctAAEJmln37QYtKfUZzYuswX3h/T7EauJur8vX+wgCDCnbny+yYXGsTdueVt91vfZbzoeSkEhxymVw==
plugin: terraform.exe: X-Amzn-Requestid: 674975a9-d7ec-11e6-a147-05a1bd9a3992
plugin: terraform.exe: 
plugin: terraform.exe: {"__type":"LimitExceededException","message":"You have already consumed your firehose quota of 20 hoses. Firehose names: [...]"}

Would be great if this error could bubble up to the console so we have a more meaningful error.

Terraform Version

0.8.2

Affected Resource(s)

aws_kinesis_firehose_delivery_stream (probably others)

Steps to Reproduce

1. Create 20 firehose streams (or what ever your max limit)
2. try to add one more with terraform

[afex]

i am seeing similiar behavior when attempting to create an aws_kinesis_stream which would exceed my shard limit.

[netjunki]

I hit another one of these with an aws_appautoscaling_target:

{"__type":"LimitExceededException","Message":"The maximum limit of 500 scalable targets per service namespace has been reached"}

@grubernaut if you can give me some guidance about where in the code to look to try and deal with this more generally I'd be happy to take a stab at a patch.

[grubernaut]

@netjunki unfortunately with the aws provider we don't have direct access to the API, so we cannot directly capture the 400 returned from EC2 in a general case. However, we can capture and handle the error correctly at the resource level. I can dig in later, but for now there should be quite a few cases of error handling from the aws-sdk in the aws_instance resource

[netjunki]

@grubernaut I know how to do that already. :-) I'll see if I can pull some patches together.

[grubernaut]

This looks related to the dynamodb issue #13339, with the possible bug living inside of the potential throttling happening inside the AWS SDK. Upstream issue: aws/aws-sdk-go#1271

[grubernaut]

Closed via #14571

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.