You can only do one disc at a time that way, BigHat. Any more and you lose data, but the forced fail upgrade swap method is not the recommended way to upsize drives. I have done it before, but your system is vulnerable during that rebuild process and they are not big fans of having data vulnerable. If all your content is handy for a reload via the vault and store if a catastrophe happens during that upgrade, the risk isn't so bad, but if you have moved your original content discs to storage and you are talking about a fully loaded 3U, that is a lot of content to put at risk. Also, a rebuild is much more intensive than a replication as each drive must be polled to calculate the parity bit math, while a replication only pulls data from the drive that contains the information you need, so the chances of failure during a rebuild are higher than during normal operation.
For a 1U server, there isn't a huge amount of data if you have access to a spare chassis to replicate into and then swap disc sets as the process won't take that long. A full 3U replication takes a while, obviously (one and a half months if full of 4TB drives).
For a 1U server, there isn't a huge amount of data if you have access to a spare chassis to replicate into and then swap disc sets as the process won't take that long. A full 3U replication takes a while, obviously (one and a half months if full of 4TB drives).