Skip to main content

AWS ECS Service Status Check

Last updated on

AWS ECS Service Status Check is a built-in Command Probe template that validates the status of one or more Amazon ECS services during a chaos experiment. It confirms that each service has reached its desired state, where the running task count matches the desired task count, within the configured timeout. Use it to assert that ECS services stay healthy and self-heal while a fault disrupts the cluster.

The probe runs the healthchecks utility bundled in the chaos probe image, queries the Amazon ECS API, and prints [Pass] when every targeted service has reached its desired state. The comparator marks the probe as passed when the output contains [Pass].

Built-in probe template

This is a built-in Command Probe template that runs on Kubernetes chaos infrastructure. Add it to an experiment from the probe library and customize its inputs. Go to Built-in probe templates to browse the full library, or go to Command probe to understand how command probes work.

Use cases​

Use this probe template to:

  • Confirm that ECS services maintain the desired task count during failures.
  • Validate service auto-scaling behavior during load changes.
  • Monitor service health during container chaos experiments.
  • Verify service deployment and rollback operations.

How the probe works​

The template configures a Command Probe that runs healthchecks -name aws-ecs. The utility resolves the services named in SERVICE_NAMES within the CLUSTER_NAME cluster and the supplied REGION, calls the Amazon ECS API, and prints [Pass] when every service reaches its desired state within STATUS_CHECK_TIMEOUT. The comparator passes the probe when the output contains [Pass], and fails it otherwise.

Prerequisites​

  • Chaos infrastructure: A Kubernetes chaos infrastructure with network access to the Amazon ECS API endpoints.
  • AWS credentials: Cloud credentials available to the chaos infrastructure, with the permissions listed below.
  • Target services exist: The cluster named in CLUSTER_NAME and the services in SERVICE_NAMES exist in REGION.

Permissions required​

The credentials used by the probe need the following AWS actions:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "ecs:DescribeServices"
      ],
      "Resource": "*"
    }
  ]
}

The probe uses the AWS credentials available to your chaos infrastructure. Go to AWS IAM integration to set up access through IAM Roles for Service Accounts (IRSA), or go to common policy for all AWS faults to apply a single superset policy.

Probe properties​

Command​

healthchecks -name aws-ecs

Comparator​

TypeCriteriaValue
stringcontains[Pass]

The probe passes when the command output contains [Pass], which indicates that every targeted ECS service has reached its desired state.

Environment variables​

VariableDescriptionRequiredDefault
CLUSTER_NAMEName of the ECS cluster that contains the service (for example, my-ecs-cluster).Yes-
SERVICE_NAMESComma-separated list of ECS service names to check (for example, web-service,api-service).Yes-
REGIONAWS region where the ECS cluster is located (for example, us-east-1, eu-west-2).Yes-
STATUS_CHECK_TIMEOUTMaximum time in seconds to wait for the service to reach the desired state.No180
STATUS_CHECK_DELAYDelay in seconds between status checks.No2

Run properties​

PropertyDescriptionTypeDefault
timeoutMaximum time to wait for the probe to complete (for example, 30s, 1m, 5m).String300s
intervalTime between probe executions (for example, 5s, 30s, 1m).String10s
attemptNumber of retry attempts before the probe is marked as failed.Integer1
pollingIntervalTime between retry attempts (for example, 1s, 5s, 10s).String-
initialDelayInitial delay before the probe starts (for example, 0s, 10s, 30s).String-
stopOnFailureStop the experiment if the probe fails.Booleanfalse
verbosityLog verbosity level (info, debug, trace).String-

Troubleshooting​

AWS ECS Service Status Check probe fails with an authorization error

The credentials available to the chaos infrastructure do not have the required ECS permissions in the target region. Confirm that the IAM policy attached to the role or IRSA service account includes ecs:DescribeServices, and that any policy condition allows the region passed in REGION.

AWS ECS Service Status Check probe cannot find the cluster or service

The cluster or service did not resolve in the supplied REGION. Verify that CLUSTER_NAME matches the cluster exactly, that every name in SERVICE_NAMES exists in that cluster, and that REGION matches the cluster region.

AWS ECS Service Status Check probe times out before the service reaches the desired state

The running task count did not match the desired count within STATUS_CHECK_TIMEOUT. Increase STATUS_CHECK_TIMEOUT and the run-property timeout, confirm that the service has enough capacity to launch replacement tasks, and check the ECS service events for task placement failures.