AWS AmazonECS documentation change
Summary
Added detailed documentation about Managed Instances termination processes including idle detection, infrastructure refresh policies, and phased termination procedures
Security assessment
The changes document infrastructure refresh termination policies that help maintain security posture through periodic instance updates. While this describes security-related maintenance processes, there's no indication of addressing a specific vulnerability. The documentation improves operational security awareness but doesn't fix an existing flaw.
Diff
diff --git a/AmazonECS/latest/developerguide/container-instance-draining.md b/AmazonECS/latest/developerguide/container-instance-draining.md index a114f64b9..3d397d7c3 100644 --- a//AmazonECS/latest/developerguide/container-instance-draining.md +++ b//AmazonECS/latest/developerguide/container-instance-draining.md @@ -5 +5 @@ -Draining behavior for servicesDraining behavior for standalone tasksProcedure +Draining behavior for servicesDraining behavior for standalone tasksDraining behavior for Amazon ECS Managed InstancesProcedure @@ -35,0 +36,29 @@ Any standalone tasks in the `PENDING` or `RUNNING` state are unaffected; you mus +## Draining behavior for Amazon ECS Managed Instances + +Amazon ECS Managed Instances termination processes ensure graceful workload transitions while optimizing costs and maintaining system health. The termination system provides three distinct decision paths for instance termination, each with different timing characteristics and customer impact profiles. + +Customer-initiated termination + + +Provides direct control over instance removal when you need to remove container instances from service immediately. You run `deregister-container-instance` with the `force` request parameter set to true, This means that immediate termination is required despite any running workloads. + +System-initiated idle termination + + +Implements cost optimization through intelligent idle detection that identifies instances no longer serving workloads. The Elastic Workload Service (EWS) implements sophisticated idle detection algorithms that monitor instance utilization and initiate termination for instances that remain idle for configurable periods. + +Infrastructure refresh termination + + +Implements proactive infrastructure management through Node Manager's natural decay policy, where instances are periodically refreshed to ensure they run on the latest platform versions and maintain security posture. Node Manager implements time-to-live (TTL) policies that initiate graceful termination for instances that have reached their maximum operational lifetime. + +The termination system implements a two-phase approach that balances workload continuity against infrastructure management requirements. + +###### Phase 1: Graceful completion period + +During this phase, the system implements graceful draining strategies that prioritize workload continuity. Service tasks are gracefully drained through normal Amazon ECS scheduling processes. Standalone tasks continue running because they might complete naturally. The system monitors for all tasks to reach stopped state through natural completion processes. + +###### Phase 2: Hard deadline enforcement + +When graceful completion does not achieve termination objectives within acceptable timeframes, the system implements hard deadline enforcement. The hard deadline is typically set to draining initiation time plus seven days, providing substantial time for graceful completion while maintaining operational requirements. The enforcement includes automatic force deregistration procedures and immediate termination of all remaining tasks regardless of the completion status. + @@ -38 +67 @@ A container instance has completed draining when all tasks running on the instan -When you are ready for the container instance to start hosting tasks again, you change the state of the container instance from `DRAINING` to `ACTIVE`. The Amazon ECS service scheduler will then consider the container instance for task placement again. +When you are ready for the container instance to start hosting tasks again, you change the state of the container instance from `DRAINING` to `ACTIVE`. The Amazon ECS service scheduler then considers the container instance for task placement again.