How can one do a point-in-time recovery for Aurora Global Cluster from a replica region?

Question

We have Aurora Globale Cluster with Postgresql. The configurationn is as follows:

* A writer in us-east-1
* A read replica in eu-central-1

Now assume that the writer in us-east-1 crashes and I can also not use the backup there. 
How is it possible to use the backup from eu-central-1 to restore everything from the latest timepoint, such that the setting is like above again (writer in us-east-1 and read replica in eu-central-1), but from the backup in eu-central-1?

Vinay · Answer

Hello,

Benefit of running Global Cluster is that you could perform fail-over to the secondary region during the time primary is unavailable and by Performing **Managed failovers** for Aurora global databases you will make secondary region primary for the meantime, as described in below documentation. Once the old primary region is available you could switchover to it.

(Note: You can only perform a managed cross-Region database failover on an Aurora global database if the primary and secondary DB clusters have the same major, minor, and patch level engine versions.)

[+]https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/aurora-global-database-disaster-recovery.html#aurora-global-database-failover.managed-unplanned

if managed failover is not possible you could perform **Manual failover**

[+]https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/aurora-global-database-disaster-recovery.html#aurora-global-database-failover.manual-unplanned

For more information about various options available with Aurora Global database for disaster recovery please refer to below documentation.

[+]https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/aurora-global-database-disaster-recovery.html

How can one do a point-in-time recovery for Aurora Global Cluster from a replica region?

Relevant content