Aurora PostgreSQL: How the process work on failover,I have a primary DB-prod for transaction database and read replica for reporting. where reporting will be pointed if primary fail.

similar question on, "If I have a primary database and an Amazon Aurora Replica actively taking read traffic and a failover occurs, what happens?" https://aws.amazon.com/rds/aurora/faqs/

1 Answer 1


The FAQ describes this fairly well. https://aws.amazon.com/rds/aurora/faqs/#If_I_have_a_primary_database_and_an_Amazon_Aurora_Replica_actively_taking_read_traffic_and_a_failover_occurs.2C_what_happens.3F

In your specific scenario:

  1. apg-node-1 - current writer
  2. apg-node-2 - current reader
  3. apg-node-1 fails
  4. AWS detects the failure and of apg-node-1 and initiates a failover to apg-node-2
  5. The reads to apg-node-2 are briefly interrupted (low seconds)
  6. The writer endpoint is re-directed to apg-node-2
  7. If apg-node-1 is unavailable the reader endpoint continues to point to apg-node-2
  8. Once apg-node-1 recovers the reader endpoint is re-directed to apg-node-1

These could be useful as well: https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraPostgreSQL.BestPractices.FastFailover.html https://aws.amazon.com/blogs/database/failover-with-amazon-aurora-postgresql/


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.