6

I'm using terraform to manage my infrastructure. I've added new cloudwatch alarms to triggers an ECS (fargate) autoscaling policy. Everything went okay, however, every time any of my cloudwatch alarms is triggered I get the following error in the log:

Failed to execute action arn:$policy_arn Received error: ""

I've checked the IAM policy and everything seems to be fine: I have multiple cloudwatch alarms, triggering the scale-up policy based on the following metrics:

  • requests per target
  • 5xx error count
  • ECS service in the cluster CPU usage

Any idea of what this can be related to?

1 Answer 1

7

I had this error because I was setting the "metric interval upper bound" to 0 instead of setting the lower bound.

2
  • 1
    I had the same except opposite on a scale-down policy.
    – jonnysamps
    Mar 23, 2020 at 15:13
  • +1 If you are copy/pasting terraform their first example has a upper bound of 0 set. Examine details here docs.aws.amazon.com/autoscaling/ec2/APIReference/… to ensure you set upper / lower bounds correctly. Of course, AWS also need to fix the error messages... Dec 2, 2020 at 15:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .