Schedule a recurring ECS RunTask invocation with Amazon EventBridge Scheduler including IAM setup, retries, DLQ, and timezone-aware cron

Documented steps

1. Create an IAM execution role for EventBridge Scheduler with a trust policy allowing Principal Service scheduler.amazonaws.com and action sts:AssumeRole (add aws:SourceArn/aws:SourceAccount conditions in production to avoid the confused-deputy problem).
2. Attach a permissions policy granting ecs:RunTask on the task definition ARN, iam:PassRole for the task's execution role and task role scoped with condition StringLike iam:PassedToService=ecs-tasks.amazonaws.com, and ecs:TagResource if propagating tags.
3. Choose a ScheduleExpression: rate(value unit) with unit minutes|hours|days (e.g. rate(15 minutes)); cron(minutes hours day-of-month month day-of-week year) (e.g. cron(0 8 1 * ? *)); or at(yyyy-mm-ddThh:mm:ss) for one-time.
4. Set ScheduleExpressionTimezone (IANA zone, e.g. America/New_York) to evaluate cron/at in local time instead of UTC; EventBridge Scheduler handles DST — a time skipped by spring-forward is not run, and a time repeated by fall-back runs only once.
5. Set FlexibleTimeWindow: {"Mode":"OFF"} for exact-time invocation, or {"Mode":"FLEXIBLE","MaximumWindowInMinutes":N} (1-1440) to randomize invocation within a window and smooth load.
6. Build the Target: Arn = the ECS cluster ARN, RoleArn = the execution role, and EcsParameters with required TaskDefinitionArn plus optional TaskCount (1-10, default 1), LaunchType, NetworkConfiguration (required for awsvpc/Fargate), PlatformVersion, Group, CapacityProviderStrategy, EnableExecuteCommand, PropagateTags.
7. Add a RetryPolicy: MaximumRetryAttempts (0-185) and MaximumEventAgeInSeconds (60-86400); Scheduler retries with exponential backoff until either limit is hit.
8. Add a DeadLetterConfig pointing Arn at a standard SQS queue ARN to capture invocations that exhaust retries; the queue's resource policy must allow the Scheduler role to send messages.
9. Create it: aws scheduler create-schedule --name my-ecs-schedule --schedule-expression 'cron(0 6 * * ? *)' --schedule-expression-timezone 'America/New_York' --flexible-time-window '{"Mode":"OFF"}' --target '{"Arn":"arn:aws:ecs:us-east-1:111122223333:cluster/my-cluster","RoleArn":"arn:aws:iam::111122223333:role/SchedulerEcsRole","EcsParameters":{"TaskDefinitionArn":"arn:aws:ecs:us-east-1:111122223333:task-definition/my-task","LaunchType":"FARGATE","NetworkConfiguration":{"awsvpcConfiguration":{"Subnets":["subnet-abc"],"SecurityGroups":["sg-abc"],"AssignPublicIp":"ENABLED"}}},"RetryPolicy":{"MaximumRetryAttempts":2,"MaximumEventAgeInSeconds":3600},"DeadLetterConfig":{"Arn":"arn:aws:sqs:us-east-1:111122223333:my-scheduler-dlq"}}'
10. Optionally set GroupName to organize schedules and State DISABLED to create without activating; toggle with aws scheduler update-schedule --state ENABLED.
11. Official docs: https://docs.aws.amazon.com/scheduler/latest/APIReference/API_CreateSchedule.html | https://docs.aws.amazon.com/scheduler/latest/APIReference/API_EcsParameters.html | https://docs.aws.amazon.com/scheduler/latest/UserGuide/setting-up.html | https://docs.aws.amazon.com/scheduler/latest/APIReference/API_RetryPolicy.html