I have seen this already in the previous versions, so it definitely is not a new thing.
For us this issue was occurring for a VM which was located in a complete different location compared to the Application Server, to which it was set to connect.
The root cause seemed to be the fact that the connection between the VM and the App Server was relatively slow, and Blue Prism was timing out before the connection was made through the scheduler.
This issue is happening only with scheduled sessions, because when you drag the process on the machine manually, your Interactive Client is connecting straight to the target machine. The direct connections between the Runtime Resources and Interactive Clients have more forgiving connection method compared to the schedulers connection method, which means that for the scheduled sessions the App Server connection needs to be as good as possible.
We made a workaround for this by hosting a local App Server connection from the machine which had the connection issues. So instead of the machine connecting a common App Server, it connects to a BP server instance which it is hosting by itself.
This workaround is not really feasible if a lot of machines are having the same issue. If many machines are affected, you probably should consider relocating the App Server -machine closer to the Runtime Resources in the network. But if only one or two machines are having this issue, then the local App Server option might be something to consider.
