How do I troubleshoot cross-Region switchover and failover issues to a secondary cluster because of a minor version mismatch in my Amazon Aurora global database?

Short description

Even when your primary and secondary DB clusters run on the same Aurora PostgreSQL-Compatible version, you might experience version mismatch. Because the primary and secondary clusters are running on different patches of the same version, you can't perform managed cross-Region switchovers or failovers.

You get an error that's similar to the following:

"Target cluster db-cluster-id must be on the same engine version as the current primary cluster"

Note: Some Aurora PostgreSQL-Compatible engine versions can perform cross-Region switchovers and failovers when the primary and secondary clusters run on different patch levels. 

Resolution

Note: If you receive errors when you run AWS Command Line Interface (AWS CLI) commands, then see Troubleshooting errors for the AWS CLI. Also, make sure that you're using the most recent AWS CLI version.

To check what versions your primary and secondary Aurora clusters are running on, run the following query with the aurora_version function:

SELECT aurora_version();

Run the describe-pending-maintenance-actions AWS CLI command to check for pending maintenance updates on the clusters:

aws rds describe-pending-maintenance-actions --resource-identifier cluster-ARN

Note: Replace cluster-ARN with your cluster's ARN.

If pending maintenance updates are available, then apply the updates to the DB cluster that you want to upgrade.

Note: The Aurora version upgrade process causes downtime. It's a best practice to perform the upgrade during your planned maintenance window.

To confirm that your primary and secondary Aurora clusters are running on the same version, run the following query:

SELECT aurora_version();

You can perform a managed cross-Region switchover or failover in your global database when the engine and patch versions match for both clusters.