Table of Contents

Interface IMultiRegionKeyProvider

Namespace
Excalibur.Compliance
Assembly
Excalibur.Compliance.Abstractions.dll

Extends IKeyManagementProvider with multi-region disaster recovery capabilities.

public interface IMultiRegionKeyProvider : IKeyManagementProvider, IServiceProvider, IDisposable
Inherited Members

Remarks

Multi-region key management provides:

  • Active-passive failover between primary and secondary regions
  • Automatic health monitoring and failover detection
  • Geographic key replication for disaster recovery
  • RPO/RTO optimization through configurable sync strategies

Architecture Decision: Active-passive is chosen over active-active to:

  1. Simplify consistency model (single writer)
  2. Avoid split-brain scenarios
  3. Align with cloud provider capabilities
  4. Maintain clear data residency boundaries

Health monitoring methods are available via IMultiRegionHealthMonitor.

Properties

ActiveRegionId

Gets the identifier of the currently active region.

string ActiveRegionId { get; }

Property Value

string

The region identifier (e.g., "westeurope", "us-east-1") of the region currently handling key management operations.

IsInFailoverMode

Gets a value indicating whether the provider is currently operating in failover mode.

bool IsInFailoverMode { get; }

Property Value

bool

true if the secondary region is active due to primary failure; false if the primary region is active.

Methods

FailbackToPrimaryAsync(string, CancellationToken)

Returns operations to the primary region after failover.

Task FailbackToPrimaryAsync(string reason, CancellationToken cancellationToken)

Parameters

reason string

The reason for failback (for audit purposes).

cancellationToken CancellationToken

A token to cancel the operation.

Returns

Task

A task representing the asynchronous failback operation.

Remarks

Failback is always manual to prevent split-brain scenarios where a recovering primary might conflict with the active secondary.

Before failback, keys created during failover are synchronized from secondary to primary.

This operation is audited via FailbackCompletedEvent.

Exceptions

InvalidOperationException

Thrown when not in failover mode or primary is unhealthy.

ForceFailoverAsync(string, CancellationToken)

Manually triggers failover to the secondary region.

Task ForceFailoverAsync(string reason, CancellationToken cancellationToken)

Parameters

reason string

The reason for manual failover (for audit purposes).

cancellationToken CancellationToken

A token to cancel the operation.

Returns

Task

A task representing the asynchronous failover operation.

Remarks

Use this for planned failover during maintenance or when automatic failover thresholds have not been met but manual intervention is needed.

This operation is audited via FailoverInitiatedEvent.

Exceptions

InvalidOperationException

Thrown when already in failover mode or secondary is unhealthy.

ReplicateKeysAsync(string?, CancellationToken)

Replicates keys from the primary region to the secondary region.

Task ReplicateKeysAsync(string? keyId, CancellationToken cancellationToken)

Parameters

keyId string

Optional specific key to replicate. If null, replicates all keys.

cancellationToken CancellationToken

A token to cancel the operation.

Returns

Task

A task representing the asynchronous replication operation.

Remarks

Most cloud providers handle replication automatically. This method is primarily for manual replication scenarios or after key imports.

For providers with automatic replication (Azure KV Premium, AWS KMS MRKs), this may be a no-op or trigger an immediate sync.