[faangguyindia]

The easiest I’ve done is in MongoDB replication, sharding, failover, and all that is super easy.

Recently, I did it in PostgreSQL using pg_auto_failover. I have 1 monitor node, 1 primary, and 1 replica.

Surprisingly, once you get the hang of PostgreSQL configuration and its gotchas, it’s also very easy to replicate.

I’m guessing MySQL is even easier than PostgreSQL for this.

I also achieved zero downtime migration.

[acdha]

Replication is not a backup. It helps for migrations or clean single node failures but not human error, corruption, or an attack.