@cvwarren @Dougfir the Mastodon software, if you have a second database server up, will use it to offload and distribute requests from users, theoretically making everything run faster and smoother. The big issue is the replication between those servers has to be fast and continuous, and there can't be any glitches or the secondary server can't catch up with the synchronization (which has happened now a few times).

Anyway, the only reason I am doing the replication is because I am hoping to switch EVERYTHING over to the new server with that secondary database, and it's a lot smoother to replicate and then fail over the database... otherwise it's about 5 hours or more of downtime to transfer the database over (shut down Mastodon, transfer database, reconfigure, bring it back up... assuming no issues). The replication means it's a few minutes to do the changeover because the data is already and sync and moved and only minimal time needed to shut down the server... in theory, lol. Plus it acts as a continuous backup. I hate to have the server COMPLETELY offline for 5 hours, not only because users won't be able to access it, but 5 or 6 hours of a server trying to catch up with the Fediverse 😬