Disaster Recovery Drill (Synchronous Replication)

Scenarios

Disaster recovery drills are used to simulate fault scenarios, formulate recovery plans, and verify whether the plans are applicable and effective. Services are not affected during disaster recovery drills. When a fault occurs, you can use the plans to quickly recover services, thus improving service continuity.

SDRS allows you to run disaster recovery drills in isolated VPCs (different from the disaster recovery site VPC). During a disaster recovery drill, drill servers can be quickly created based on the disk snapshot data. This way, drill servers will have the same server specifications and disk types as the production site servers.

After drill servers are created, production site servers and drill servers will independently run at the same time, and data will not be synchronized between these servers.

To guarantee that services can be switched to the disaster recovery site when an outage occurs, it is recommended that you run disaster recovery drills regularly to check that:

• Data between the production site and disaster recovery site is consistent at the moment you create a disaster recovery drill.
• Services run properly at the disaster recovery site after a planned failover.
  Figure 1 Disaster recovery drill
  

Precautions

• If the disaster recovery site servers of a protection group are added to an enterprise project, the drill servers created will not be automatically added to the enterprise project. Manually add them to the project as needed.
• If an existing drill VPC is used for a new drill, the subnet ACL rule of the drill VPC will be different from that of the protection group VPC. Manually set them to be the same as needed.
• If a custom route table is configured and associated with a subnet in the protection group VPC, the corresponding route table will not be automatically created in the drill VPC. Manually create one as needed.
• If the disaster recovery site servers run Windows and use key pairs for login, ensure that the key pairs exist when you create the drill. Otherwise, drill servers may fail to create, resulting in the drill creation failure.
  

  If a key pair has been deleted, recreate the key pair with the same name.

• If the disaster recovery site servers run Linux and use key pairs for login, the key pair information will not be displayed on the server details page, but login using the key pairs is not affected.
• After a disaster recovery drill is created and before it is executed, modifications made to Hostname, Name, Agency, ECS Group, Security Group, Tags, and Auto Recovery of disaster recovery site servers will not synchronize to drill servers. Log in to the console and manually make the modifications for the drill servers.
• If the synchronization progress of replication pairs in the protection group is not all 100%, the created drill servers may fail to start. It is recommended that you run disaster recovery drills after all replication pairs are synchronized.

Prerequisites

• The protection group is in the Available, Protecting, Failover complete, Enabling protection failed, Disabling protection failed, Planned failover failed, Re-enabling protection failed, or Failover failed state.
• Do not run disaster recovery drills before the first time data synchronization between the production site servers and disaster recovery site servers completes. Otherwise, drill servers may not start properly.

Procedure

1. Log in to the management console.
2. Click Service List and choose Storage > Storage Disaster Recovery Service.

   The Storage Disaster Recovery Service page is displayed.

3. In the pane of the protection group to which a DR drill is to be added, click DR Drills.

   The protection group details page is displayed.

4. On the DR Drills tab, click Create DR Drill.

   The Create DR Drill dialog box is displayed.

5. Configure Name and Drill VPC.

   Table 1 Parameter description

   Parameter	Description	Example Value
   Name	DR drill name	DR drill servername
   Drill VPC	VPC that used for a DR drill. It cannot be the same as the VPC of the DR site server. The value can be Automatically create or Use existing. Automatically create: The system automatically creates a drill VPC and subnet for the protection group. Use existing: The system uses an existing VPC as the drill VPC. If you select to use an existing VPC, the subnet CIDR block of the drill VPC must be consistent with that of the production group VPC. NOTE: The drill VPC cannot be the same as the VPC of the protection group.	vpc-f9f7

6. Click OK.

   After the disaster recovery drill is created, you can log in to a drill server and check whether services are running properly.