Throttling Logic for the Amazon ECS Service Scheduler

The Amazon Elastic Container Service (Amazon ECS) service scheduler now includes logic that throttles how often tasks are restarted if they repeatedly fail to launch.

Previously, if tasks had issues in the container image, task definition, network configuration, they could fail to become healthy when started by the Amazon ECS service scheduler. Continuous restart attempts by the service scheduler could lead to overall application performance degradation and incur cost.

Now, the Amazon ECS service scheduler includes logic that checks for tasks that continuously fail to become healthy after instantiation, increases the duration between restart attempts, eventually increasing the duration to 15 minutes, and adds an event to the Amazon ECS service event messages so you can take corrective action. This logic works for tasks using both the AWS Fargate and EC2 launch types.

To learn more, visit the Amazon ECS documentation. For more information about Amazon ECS see the product page.

Amazon ECS is available in US East (Ohio), US East (N. Virginia), US West (N. California), US West (Oregon), Canada (Central), South America (São Paulo), Asia Pacific (Mumbai), Asia Pacific (Seoul), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Tokyo), EU (Frankfurt), EU (Ireland), EU (London), EU (Paris), and China (Beijing) regions. For more information on AWS regions and services, please visit here.