PostgreSQL is an advanced open source database that is completely community driven. Continuous development and performance improvements, while maintaining a secondary focus on Oracle compatibility, gave PostgreSQL a great market penetration. When a database server is deployed in production, we often wish to achieve several 9’s of availability. Is that even possible with PostgreSQL? What is the combination of tools that you could combine and implement to achieve High Availability (HA) and automatic failover in PostgreSQL? How can we avoid data loss during such failovers? We’ll address these questions and then some more in this talk.
We are going to discuss **
1. How the implementation of HA differs for each type of replication available in PostgreSQL.
2. How to combine HAproxy with etcd plus a detailed explanation of the RAFT Algorithm and HA using Patroni.
3. How to combine repmgr with keepalived to achieve HA.
4. HA solution built for PostgreSQL on kubernetes.
5. What are the tools and solutions that help you achieve automatic failover in AWS and other cloud-based environments.
6. How to avoid huge data loss during failover.
