Fair enough. Im in devops and the first thing I thought about was Jenkins, where “server” and “agent” fit quite well.

I dont think master/slave is that good of a naming scheme for fault tolerance either, since the “slave” doesnt do work so that the master doesnt have to, but it’s rather an active/reserve kind of thing.

But I also admit that using different terms that fit best for every usecase would only cause more confusion than good.