How to plan shared storage migration of SQL server 2019 on windows failover cluster?

Question

I need migrate a Sql server on a fail-over cluster to a new SAN storage. Currently it is connected to Netapp SAN uning fibre channel and this cluster needs to be migrated to another SAN storage. As this is hosting production database, we prefer zero-downtime migration by performing rolling node replacement. How to plan this migration, how to approach this and what are the precautions to be taken?

Answer 1

To plan the migration of a SQL Server 2019 instance on a Windows Failover Cluster to a new SAN storage while ensuring zero-downtime, you can follow these steps:

1. Inventory and Assessment: Begin by inventorying the current SQL Server environment, including the nodes in the failover cluster, the configuration of the existing SAN storage, and the databases hosted. Assess the new SAN storage capabilities and compatibility with your current setup.
2. Plan for Zero-Downtime Migration: Since you aim for zero-downtime, consider using a rolling node replacement strategy. This involves migrating one node at a time while keeping the other node operational. Ensure that your cluster can handle the load during the migration process.
3. Prepare the New SAN: Configure the new SAN storage according to your requirements. Ensure that it is set up to support the same protocols (e.g., Fibre Channel) and that it meets the performance and availability needs of your SQL Server workloads.
4. Update Cluster Configuration: Before migrating, update the failover cluster configuration to recognize the new SAN. This may involve adding the new storage to the cluster and ensuring that all nodes can access it.
5. Data Migration: Use a data migration strategy that minimizes downtime. You can use tools like the Storage Migration Service to facilitate the transfer of data from the old SAN to the new one. Ensure that the data is synchronized between the old and new storage during the migration process.
6. Testing: After migrating data to the new SAN, conduct thorough testing to ensure that all applications and services are functioning correctly. Validate that the SQL Server instance can access the databases on the new SAN without issues.
7. Failover and Cutover: Once testing is complete, perform a failover to the node that has been migrated to the new SAN. Monitor the performance and stability of the SQL Server instance during this phase.
8. Decommission Old SAN: After confirming that everything is working as expected, you can decommission the old SAN storage. Ensure that all data has been migrated and that there are no dependencies on the old storage.
9. Documentation and Review: Document the migration process, including any issues encountered and resolutions applied. Review the entire migration to identify any areas for improvement in future migrations.

Precautions:

• Ensure that you have a complete backup of all databases before starting the migration.
• Monitor the performance of the cluster during the migration to avoid overloading any single node.
• Maintain clear communication with stakeholders about the migration timeline and any potential impacts.
• Test the migration process in a non-production environment if possible to identify any issues beforehand.

By following these steps and precautions, you can effectively plan and execute the migration of your SQL Server 2019 instance to a new SAN storage with minimal disruption.

---

References:

• Migrate a SQL Server Always On Failover Cluster Instance to Azure VMware Solution
• Use Storage Migration Service to migrate a server