Hi @imarimas
Would you like to share more about how this work was set up to run?
The limit inside of orthofinder is specific to the tool, and it is with respect to the number of temporary working files created, not dataset files, correct? I’m pretty sure this would be a limit no matter where the original algorithm was run: inside of Galaxy or anywhere else. The warning is likely indicating either a missing data preparation step or that there is potentially a parameter tuning issue to address.
However I may be misunderstanding! We can confirm!
As a reference, the protocol is from this source and would be the best resource to follow as we work through this.
The current computational resource allocation at UseGalaxy servers is this.
Please let us know if you would like more help! I think others had trouble helping because the job inputs, logs, and full error messages were not shared. You can do that now for some feedback! If you would rather share in a direct message, please let me know and I can start one up. 