I don't know if it is fair to call it a disaster. I don't know enough from the inside, but I believe in retrospect the goal was maybe to ambitious or plain wrong.
They were attempting to port huge amounts of decades old Office macros to OpenOffice. That failed, but before the LiMux project they had already failed to migrate the same to a modern version of MS Office.
The goal for LiMux was to be a better Windows than the best Windows Microsoft would offer at the time. Literally impossible.
That combined with strong lobbying and users confused with a different UI and probably a lot of small day-to-day issues (which happens with any software, but can make an IT department look bad) made it politically hard to sustain an 'experiment'.
The current IT lead of Munich, hired after migrating back to Microsoft, does not seem to be a Microsoft fan.