Please excuse any mistakes. I tried to make this as clear as I could considering how convoluted this process is.
Lets walk through the API communication pathway to understand this better.
Arr sends download request to RDT.
RDT gets file.
RDT tells Arr that file is complete and look in this mapped folder to find it. (This folder mapping is defined by RDT and not Arr || No remote mapping is needed in ARR itself, only RDT)
Arr looks in that folder and completes post download process.
The key to getting this to work is to map a download folder on both docker instances through the unraid interface.
In this example map I want to make sure both containers have access to the following by clicking "Add another Path, Port, Variable, Label or Device":
Container Path: /data/foldernameyouwant/
Host Path: /mnt/user/data/foldernameyouwant/
Make sure that "/data/foldernameyouwant/" or even the upstream folder of "/data/" accessible.
Now it is time to set this up inside RDT
Complete step above to Map path in all docker instances /data/foldernameyouwant/
Go to Settings at top of RDT
Click Download Client
Change Mapped path & Download Path to the same download folder that Arr & RDT have access to. Example: /data/foldernameyouwant/
Check System->Status in Arr program to see if the error is resolved
Note: "Download path" is the download folder for manually added downloads/blackhole and not for Arr support. That is where this gets a bit confusing. Mapped path is for Arr. Set both to the same directory to take any guess work out of the equation.
Edit: You must add RDT-Client details in the Arr program under download clients. Use "qBittorrent" since the RDT emulates that. Make sure there is no category assigned. Check under settings and look to make sure Arr can see the /data/foldernameyouwant
Good luck!
------
Now if I could just figure out Gluetun setup with docker instances but still maintain local pass throughs I would be over the moon.