An Amazon Aurora PostgreSQL global database has two secondary Regions. The database parameter group is configured to guarantee an RPO (recovery point objective) of 60 seconds. Occasionally write operations on the primary cluster are blocked because of this RPO configuration. You need to reduce how often write operations are blocked. Which action accomplishes this?

The RPO of 60 seconds means that data changes must be replicated to all secondary clusters within 60 seconds. If replication to any secondary cluster falls behind, write operations on the primary cluster are blocked to maintain the RPO. Removing one secondary cluster reduces the number of targets that must acknowledge replication, making it easier for the remaining secondary clusters to keep up and thus reducing blockages. Adding another secondary cluster would increase the replication burden, making blockages more likely. Enabling write forwarding allows writes to secondary clusters but doesn't address the RPO constraint on the primary. Configuring synchronous replication would worsen the problem by requiring immediate acknowledgment from all secondaries, significantly increasing latency and blockages.