To figure out what went wrong it is best that you create and save diagnostic snap shots for EACH node in the cluster.
To actually save a copy of these files on each node you can create the diagnostic snap shot from the SZ CLI: en diagnostics show snap (show previous snap shots) execute all (create new snap shot)
The snap shot can also be created from the GUI in the Diagnostics::Application logs using the "Download Snapshot logs" button. This creates the snap shots and downloads it to the browser but will not save it on the system. It may fail if the system is busy as this file can be very large.
If you saved the snapshot using the CLI command - it can be downloaded from the "Diagnostics" tab under Application logs. Click on the number link next to the log name to open the list of contents then click on the snapshot name to download it.
This file is compressed but not encrypted so you could open it up using a decompress tool and dive through all the logs, but I suggest you open a ticket with Ruckus Support to help get the problem analyzed.