Azure Storage Account - Disaster Recovery: Planned Failover

Introduction

Disaster recovery (DR) is a crucial aspect of cloud storage, ensuring business continuity during outages. Azure Storage provides built-in geo-redundancy to protect against regional failures.

Microsoft has introduced Planned Failover, currently in preview, to allow seamless failover without data loss, enabling customers to manage the Failover themselves as well as conduct Disaster Recovery drills for testing the resilience.

Overview of Replication Options

Azure Storage offers multiple redundancy options to safeguard data.

Locally Redundant Storage (LRS): Replicates data within a single data center.
Zone-Redundant Storage (ZRS): Distributes data across multiple availability zones within a region.
Geo-Redundant Storage (GRS): Replicates data asynchronously to a secondary region.
Geo-Zone-Redundant Storage (GZRS): Combines ZRS and GRS for enhanced redundancy.

GRS and GZRS provide disaster recovery capabilities by maintaining a secondary copy in a paired region.

Planned Failover in Preview

Planned Failover is a new capability in preview that allows controlled failover from the primary region to the secondary without data loss. Unlike unplanned failover, where data loss might occur due to asynchronous replication, Planned Failover ensures a zero-data-loss transition.

Enabling Planned Failover in Preview

To use Planned Failover, you must register the AllowSoftFailover feature at the subscription level as shown below.

Previews

Once registered, the Planned Failover option will be available for GRS and GZRS storage accounts.

Note. The Planned Failover option is currently in Preview and available only in the regions below.

France Central
France South
India Central
India West
East Asia
Southeast Asia

Official Documentation: https://learn.microsoft.com/en-us/azure/storage/common/storage-disaster-recovery-guidance#customer-managed-planned-failover-preview

Prerequisites

Create a Storage Account in one of the above regions where Planned Failover is available.
Ensure your storage account uses GRS or RA-GRS (recommended).
Register the AllowSoftFailover feature at the subscription level.
Create a Container and upload a few Blobs to the Container.
Download & Configure Azure Storage Explorer

Let’s connect to the Storage Explorer by creating two connections as shown below. Please note that we will connect to the same Storage account but to both locations.

Storage explorer

Primary Region

Connect the Storage account using the Access Key

Secondary Region

Connect the Storage account using SAS Token. For this, create a SAS token as shown below and copy the Blob Service SAS URL.

Secondary Region

While connecting to the Secondary Region using the above SAS token, please make sure that you add -secondary to the storage account name as shown below.

SAS token

Once the connection is created, make sure you validate it as shown below.

Connection

Demo of Planned Failover

Let’s now understand how to perform the Planned Failover by navigating to the Redundancy blade and reviewing the Primary & Secondary regions (after failover, these two regions would swap)

Demo

Once you review the locations you can click on the Prepare for Failover button as shown above which opens up a popup where you can choose the Planned Failover option and then click on Failover

Planned Failover

In a moment you will see the messages below and the regions are also swapped.

Progress

The process would take from a few minutes to an hour. In my case, I have uploaded a fail of size 100+ MB for which the failover took approximately 10 to 12 minutes.

Once the failover process is complete, then without any data loss, you can start using the original secondary as the new Primary Region.

Summary

Planned Failover enhances disaster recovery strategies by enabling a zero-data-loss transition to a secondary region. This feature, currently in preview, allows controlled failover scenarios, ensuring business continuity during planned maintenance or potential outages. By registering AllowSoftFailover, organizations can leverage this capability to enhance their Azure Storage resilience.

Stay tuned for general availability updates and further enhancements to Azure Storage DR strategies!