Amazon ECS service scaling execution block
The ECS service scaling execution block allows you to scale your ECS service in a destination Region as part of your multi-Region recovery process. You can define a percentage of capacity, relative to the Region that Region switch fails over from or deactivates.
Configuration
To configure the ECS service scaling execution block, enter the following values.
Important
Before you configure the execution block, make sure that you have the correct IAM policy in place. For more information, see Sample policy for ECS service scaling execution blocks.
Step name: Enter a name.
Step description (optional): Enter a description of the step.
Resource for Region: For each Region, enter the ECS cluster ARN and the ECS service ARN.
Percentage to match the source Region's task count: Enter the desired percentage of running tasks in the source Region to match in the activated Region.
Capacity monitoring approach: Select one of the following approaches for monitoring capacity for Amazon ECS:
Max running capacity sampled over 24 hours: Choose this option to use the running tasks count value in your Amazon ECS service. This option does not create additional costs, but is potentially less accurate than using the other option, CloudWatch metrics.
For more information, see Automatically scale your Amazon ECS service in the Amazon Elastic Container Service Developer Guide.
Max running capacity sampled over 24 hours via container insights: Choose this option to use Amazon ECS Container Insights metrics. Using the option can be more accurate, but incurs the additional costs of using Container Insights.
To use this option, you must first enable Container Insights. For more information, see Set up Container Insights in the Amazon CloudWatch User Guide.
Timeout: Enter a timeout value.
Then, choose Save step.
How it works
After you configure the execution block in your plan, Region switch confirms that there is only one source ECS service and one destination service. If there are multiple services, Region switch returns a warning for the execution block. Region switch stores this data in all Regions your plan is configured for. The target capacity is defined as the desired count set on your ECS service.
For an active/passive approach, Region switch calculates the new desired capacity for the ECS service
in the destination (activating) Region.
The new desired capacity is compared against the destination ECS service's desired
capacity. The formula that Region switch uses to calculate desired capacity is the following:
ceil(percentToMatch * Source Auto Scaling group capacity), where ceil() is a
function that rounds up any fractional result. If the current desired count for the destination
ECS service is higher than the calculated new desired capacity for the ECS service, the
plan execution proceeds. Note that Region switch does not scale down ECS service capacity.
If the ECS service has Application Autoscaling enabled, Region switch updates the minimum capacity in Application Autoscaling, and also updates the desired count in the ECS service.
When Region switch executes an ECS service block, Region switch attempts to scale up the target Region ECS capacity to match the desired capacity. Then, Region switch waits until the requested ECS service capacity is fulfilled in the target Region's ECS service before Region switch proceeds to the next step in the plan. If you like, you can configure the step to complete before fulfillment is complete by setting a timeout limit for how long Region switch waits for capacity fulfillment.
If you’re using an active/active approach, Region switch uses the other configured Region as the source. That is, if a Region is being deactivated, Region switch uses the other active Region as the source to match for the percent to scale.
What is evaluated as part of plan evaluation
When Region switch evaluates your plan, Region switch performs several checks on your ECS service execution block configuration and permissions. Region switch verifies that ECS services are present in both the source and target Regions, and checks to make sure that the maximum capacity set for the target Region's ECS service is sufficient to handle the specified percentage match of the target Region's capacity. Region switch also validates that the plan's IAM role has the correct permissions for ECS service. For more information about the required permissions for Region switch execution blocks, see Identity-based policy examples for Region switch in ARC.
In addition, Region switch checks that the ResourceMonitor has successfully collected and stored the
necessary monitoring data for the ECS services, and captures a count of the number of running tasks.
If any of the checks fail, Region switch returns warning messages, which you can view in the console. Or, you can receive the validation warnings through EventBridge or by using API operations.
