I can't repro this in a simpler cell, only in the golden scenario notebook:

https://github.com/claudiaregio/data-science/

• Execute all the cells up to the one that starts #Summarize history of accuracy
• Start RBL on this cell
• Run all the way through it. At the end, immediately click to start another RBL session
• It doesn't start RBL or show the debug highlight
• Click again
• It starts on the second line

I think it did start in the background but didn't get all the wires connected

Haven't looked through this yet
log.txt

A different thing that happens sometimes is that the debug session never ends, clicking RBL starts a second one but we never actually start running and show the debug line. Burke hit this in the golden scenario run.

log2.txt