Document history for the Amazon Application Recovery Controller (ARC) Developer Guide

You can start a zonal shift for an Amazon EKS cluster, or you can allow Amazon to do it for you by enabling zonal autoshift. This shift updates the flow of east-to-west network traffic in your cluster to only consider network endpoints for Pods running on worker nodes in healthy AZs.

For more information, see Support for Amazon Elastic Kubernetes Service.

ARC now supports zonal shift for Network Load Balancers with cross-zone enabled or cross-zone disabled configurations.

For more information, see Support for Network Load Balancers.

With autoshift observer notifications, you can configure zonal autoshift to notify you, through Amazon EventBridge, whenever Amazon starts an autoshift to shift traffic away from a potentially impaired Availability Zone. You do not have to configure any specific resources with zonal autoshift to enable these separate notifications.

For more information, see Using zonal autoshift with Amazon EventBridge.

Reorganizes the developer guide content to be siloed into sub-dev guides. That is, there are now separate sections that contain comprehensive information for each capability in ARC: zonal shift and zonal autoshift for multi-AZ recovery, and routing control and readiness check for multi-Region recovery.

For more information, see What is Amazon Application Recovery Controller (ARC).

Adds a new capability in ARC where you authorize Amazon to shift away resource traffic for an application from an Availability Zone, on your behalf, to help reduce time to recovery during events.

For more information, see Zonal autoshift in Amazon Application Recovery Controller (ARC).

Adds a new service-linked role, AWSServiceRoleForZonalAutoshiftPracticeRun, for zonal autoshift practice runs.

For more information, see Service-linked role permissions for AWSServiceRoleForZonalAutoshiftPracticeRun.

Adds cross-account support for clusters in ARC with Amazon Resource Access Manager, so that you can easily and securely use one cluster to host control panels and routing controls owned by several different Amazon accounts.

For more information, see Support cross-account for clusters in ARC.

Updates the AmazonRoute53RecoveryControlConfigReadOnly managed policy to add permissions for GetResourcePolicy, to support returning details about Amazon Resource Access Manager resource policies for shared resources.

For more information, see Amazon managed policies.

Added new permissions, ec2:DescribeVpnGateways and ec2:DescribeCustomerGateways, to the service-linked role for ARC, to support polling Amazon EC2 instances.

For more information, see Using service-linked roles for ARC.

Supports the GA release of zonal shift for ARC, which includes attribute-based access control (ABAC) for managed resources that are registered in ARC for zonal shift.

For more information, see Attribute-based access control (ABAC) with ARC.

Added content describing a new service in ARC, zonal shift, for multi-AZ applications. You can start a zonal shift to temporarily move traffic for a load balancer resource away from an Availability Zone.

For more information, see Zonal shift in ARC.

Added a new permission, lambda:ListProvisionedConcurrencyConfigs, to the service-linked role for ARC to query information about Lambda functions.

For more information, see Using service-linked roles for ARC.

Updated the AmazonRoute53RecoveryControlConfigFullAccess managed policy to remove Amazon Route 53 permissions and list them as optional.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC).

Updated the AmazonRoute53RecoveryControlConfigFullAccess managed policy to include required Amazon Route 53 permissions.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC).

Added example CLI command and best practices recommendations for the new list routing controls API operation included in the extremely reliable ARC data plane API.

For more information, see List and update routing controls and states.

Added support for overriding safety rules, which allows you to bypass routing control safeguards that are enforced with safety rules that you've configured. Safety rule overrides could be required, for example, in a "break glass" scenario during failover for disaster recovery.

For more information, see Override safety rules to reroute traffic.

Added support for tagging additional resources in ARC, including clusters, control panels, routing controls, and safety rules.

For more information, see Tagging in Amazon Application Recovery Controller (ARC).

Updated the AmazonRoute53RecoveryControlConfigReadOnly managed policy to add permission to list tags for a resource.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC)

Added support for EventBridge, which means that now you can add rules to get alerts and act on ARC readiness check status changes, for example, when a status changes from READY to NOT READY.

For more information, see Using ARC with Amazon EventBridge.

Added code samples to illustrate trying cluster endpoints in sequence when you use API operations to get or update routing control states.

For more information, see API examples for Amazon Application Recovery Controller (ARC).

Added two new permissions to the policy AmazonRoute53RecoveryReadinessReadOnlyAccess: route53-recovery-readiness:GetArchitectureRecommendations and route53-recovery-readiness:GetCellReadinessSummary.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC).

Added a new resource type, Amazon API Gateway, and updated the ARC service-linked role permissions so that ARC can audit API Gateway with readiness checks.

For more information, see Readiness rules and supported resource types and Using service-linked roles for ARC.

Added a new resource type, Lambda functions, and updated the ARC service-linked role permissions so that ARC can audit Lambda functions with readiness checks.

For more information, see Readiness rules and supported resource types and Using service-linked roles for ARC.

Added links to downloadable Amazon CloudFormation and Hashicorp Terraform templates to help you quickly get started with using ARC.For more information, see Recovery readiness with a new application.

Added the following Amazon managed policies for ARC: AmazonRoute53RecoveryReadinessFullAccess, AmazonRoute53RecoveryReadinessReadOnlyAccess, AmazonRoute53RecoveryClusterFullAccess, AmazonRoute53RecoveryClusterReadOnlyAccess, AmazonRoute53RecoveryControlConfigFullAccess, and AmazonRoute53RecoveryControlConfigReadOnlyAccess.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC).

Updates for managed policies will be tracked from the initial release date forward.

For more information, see Amazon managed policies for Amazon Application Recovery Controller (ARC).