Forum Discussion
Using R script and working with multiple developers
- 1 year ago
Hi jaryszek,
Go to Manage connections and gateway, click on the on-prem gateway
after clicking on manage users you can give access to that gateway to another user
The other user can now access your gateway.
Thank you.
Hi jaryszek
You're using a personal gateway to enable Power BI to refresh datasets that rely on R scripts, and you've correctly observed that the gateway must be running on the specific machine where the R environment is configured. When you logged in to Power BI from a second laptop, you couldn't use the gateway because the first laptop—where the personal gateway and R setup exist—was turned off. This behavior is expected because a personal gateway is bound to the specific user profile and machine where it's installed, meaning the R script and its dependencies (libraries, paths, etc.) are tied to that local environment.
Regarding your question on whether the gateway works "per logged-in user"—yes, the personal gateway works only under the context of the specific Power BI user who installed and configured it. However, it's also bound to that user's machine. So even if you log into the same Power BI user account from another device, the R-based refresh won't work unless it's routed through the original machine with the gateway and R environment active.
As for sharing the same personal gateway across multiple users—this is not possible. A personal gateway is single-user, single-machine. It cannot be used by other users or shared across machines. If you have multiple users who need to run R scripts in Power BI, each would need to set up their own R environment and install a personal gateway on their respective machines.
And you're right—the Power BI enterprise (standard) gateway does not support R or Python scripts for scheduled refreshes due to security and sandboxing constraints. So, for scenarios requiring script-based transformations, personal gateway is the only option, but it's inherently limited to the user's local environment. For broader usage across teams, consider moving R logic upstream (e.g., into a dataflow, stored procedure, or external ETL process) and use supported data sources in Power BI.