I've been having a think about this, and I suspect it's the result of the console UI not being set up in a way that properly explains what settings it uses for what purpose.
Every machine in the console has a set of configuration settings recorded for it - in the console, this are the "default configuration", and after you've applied a task to a machine and the task has the "Refresh configuration" tickbox set, at the end of the task it will ask the machine for its current settings and record those as the default.
When you create a task, there is a matching tickbox called "Use default settings". When you have a "Configuration" task step, any specific configuration settings you specify for the task take precedence over the saved defaults. Then, for every setting you
didn't override in the task, if the "Use default settings" tickbox is ticked, it picks up the last-saved default setting for the individual machine and tries to reapply that.
Basically, as long as the console has the right machine name stored in the "default configuration", you can use this system to get the name reapplied.
Now, something that may help to make sense of the above explanation is that if you right-click on a task, there is a context menu option called "task scenario". This provides a single document that describes everything the console thinks is going to happen, and includes a list of every configuration setting for every machine that it plans to apply.
If you have a task that isn't giving the results you want, it's a good idea to get that task scenario and perhaps think about posting parts of it here, so we can look at it and perhaps explain why it has made that decision.