[pipe_connector]

Is simplified failover referring to the currently-in-beta async replicas? Or is there something else on the way that will make it easier to failover?

[mrkurt]

There's nothing that's quite ready to look at for easy failover, but this is a big priority. We absolutely have to figure out failover during deploys. It's doable we just want to reduce the operational complexity as much as we can.

[mtremsal]

I _think_ he's referring to the ease of use of streaming the WAL to S3 and reloading from it upon starting a new write "leader".