I am currently working on a process monitor for the cloud VM that runs the server these days. Moving to the cloud had decreased downtime DRAMATICALLY but there are still instances where the ultima server crashes and the VM does not relaunch the service. Further more since the server does not run locally I dont know about it unless its reported or I cant log in.
In the coming days I will deploy a new program to monitor the ultima six online server health and automatically relaunch the server in cases of fatal error or close. I hope this will bring down time to near 0. I will report back when it's deployed
I let the first implementation run over the last week while I was out of town. It appears to have detected three server crashes and automatically rebooted the server. I will continue to monitor this process over the next few months but right now it seems solid.