Join us for an expert-led overview of the tools and concepts you'll need to pass exam PL-300. The first session starts on June 11th. See you there!
Get registeredPower BI is turning 10! Let’s celebrate together with dataviz contests, interactive sessions, and giveaways. Register now.
I have a report with an R visual in it that worked as recently as Feb 2025. Now when anyone opens the page of the report, the visual is blank with no error messages or anything. The visual uses tidyverse, ggrepel and ggpubr libraries, and the output is thru ggplot and various associated functions. It works as expected on Desktop.
Desktop:
Service:
I am not seeing any error messages that would lead me in the right direction. Any assistance appreciated!
Thanks,
David
Solved! Go to Solution.
UPDATE
I have narrowed it down to a single line in my code
... + scale_color_manual(drop = FALSE, values = colors, labels=NULL) + ...
When I comment out that line, the plot renders in the service, but with the "default" colors. I don't get why a snippet of code would stop working all of a sudden, but here we are.
Still exploring Deneb/Vega - looks like a neat solution, but I'll need to figure out if I can duplicate the visual here (scatter plot with arrows connecting the previous time's value to the present time's value)
-David
You can still raise the ticket if you want
Since you have a Pro license you can open a Pro ticket at https://admin.powerplatform.microsoft.com/newsupportticket/powerbi
As a follow-up, I wanted to let the forum know that Power BI now supports R 4.3.3 and associated libraries. When I upgrade my local R version and run the visual in Desktop it errors out so at least I have a place to start.
Documentation on updated R/Python support: https://powerbi.microsoft.com/en-us/blog/important-update-coming-to-r-py-visuals-in-power-bi/?cdn=di...
-David
As a follow-up, I wanted to let the forum know that Power BI now supports R 4.3.3 and associated libraries. When I upgrade my local R version and run the visual in Desktop it errors out so at least I have a place to start.
Documentation on updated R/Python support: https://powerbi.microsoft.com/en-us/blog/important-update-coming-to-r-py-visuals-in-power-bi/?cdn=di...
-David
UPDATE
I have narrowed it down to a single line in my code
... + scale_color_manual(drop = FALSE, values = colors, labels=NULL) + ...
When I comment out that line, the plot renders in the service, but with the "default" colors. I don't get why a snippet of code would stop working all of a sudden, but here we are.
Still exploring Deneb/Vega - looks like a neat solution, but I'll need to figure out if I can duplicate the visual here (scatter plot with arrows connecting the previous time's value to the present time's value)
-David
You can still raise the ticket if you want
Since you have a Pro license you can open a Pro ticket at https://admin.powerplatform.microsoft.com/newsupportticket/powerbi
Hi @dedelman_clng,
Thanks for reaching out to the Microsoft fabric community forum.
In addition to the valuable points mentioned by @TomMartens and @lbendlin , I would like to point out to the licensing requirements of R visuals, as you mentioned these visuals were working fine as recently as FEB 2025.
R Visuals require pro or PPU license to render the r visuals, please check if your license is still active or if you are a Free user of Power BI you can only consume tiles shared with you in Premium workspaces.
Create advanced analytics and visualizations using R scripts - Power BI | Microsoft Learn
Another point that may be helpful to you is the R visuals rendered from the Power BI service (for Premium/ F SKUs) will support payloads of < 30Mb.
Source: Learn which R packages are supported - Power BI | Microsoft Learn
If you find this post helpful, please mark it as an "Accept as Solution" and consider giving a KUDOS. Feel free to reach out if you need further assistance.
Thanks and Regards
Thank you for your response. I have a valid Pro license. My payload is very small, certainly less than 30 MB.
Nothing changed in the report (aside from data refreshes) since the last time it worked. Is there a console or debug that I can access (I have tenant admin rights) to see if it is throwing an error or something, or at this point, should I put in a support ticket?
David
You must use the default renderer and you must use a Personal Gateway.
Hi @lbendlin - could you expand on this a little more? I don't see options for a "default renderer" in PBI desktop. This visual was working as late as Feb 2025, so I don't know what could have changed since then unless it was something in the service?
Thanks
David
Hey @dedelman_clng ,
I stopped using R-script visuals or Python-script visuals some time ago. Instead, I'm using the Deneb custom visual that allows me to use the Vega-Lite and Vega visuals and comes with cross-filtering capabilities and other interactive features.
However, I recommend comparing the versions of the R packages supported by the Service and that you use locally. I had some issues with different versions in the past.
R-packages supported by the service: https://learn.microsoft.com/en-us/power-bi/connect-data/service-r-packages-support
Hopefully, this will help tackle your issue.
Regards,
Tom
Thanks @TomMartens - I will check out the custom visuals you mentioned.
In the interim, does the "supported packages" list show the minimum version for supported packages, or the only version that it supports?
Thanks
David
Hey @dedelman_clng ,
It shows the "supported" package, NOT the minimal supported package, the package versions used when the Power BI Service executes the R-script.
Of course, this does not mean that "newer" or "older" packages will break the visual, but I have encountered instances more than once where my newer local packages introduced "breaking changes."
Another reason I stopped using R- or Python-script visuals because I think local package management to comply with the Power BI Service package versions is not worth the effort, at least not on my site.
Regards,
Tom
User | Count |
---|---|
47 | |
32 | |
30 | |
27 | |
25 |
User | Count |
---|---|
56 | |
55 | |
36 | |
33 | |
28 |