community 
Execution server selectors should be used only if defined on execution servers
The flow today is that if you have a resource with one/some execution server selectors (ESS) - it will be used to find the execution server to be executed on.
However, if you don't define the execution server selector on a resource, it will use any of the available execution servers.
This means, that from the moment you add a selector, you'll have to make sure all of the resources have them (in most cases).
This can happen if the selector represents a new remote lab - in such cases, you add a new execution server located in the remote lab, and you'll want all of that lab resources to use this one. While it's ok to then go and set the ESS on these remote-lab resources, if you won't set something else for the other resources (the ones in the local lab), they'll start using this remote execution server/s as well.
Another use case is when you add a linux execution server, and you want only some of the resources to run on it, while the others should keep using the windows one. Here as well, you'll need to set the ESS on all of the resources, instead of setting it only on the ones that need the linux one.
