I’m running RoonServer 1.3 build 200 under a LXC container (Ubuntu 16.04 host). My music folder is bind mounted into the container and happens to come from a ZFS data store from the host.
Using a completely fresh container and install of RoonServer (empty database) if I add the folder to it, it goes through the analysis and works fine. The problem is that whenever I restart the RoonServer or the entire container the folder goes to the DriveNotReady state. Disable/Enable/Edit does not fix it. The only way I’ve found to fix it is to remove and re-add the folder and then it works fine until the next restart.
Having to remove and re-add folders every restart is obviously no good. What’s the deal with this? The storage is obviously there and working fine.
Can you get use the whole contents of the logs folder from your RoonServer to us in some sort of zip file? By default they’re located in /var/roon/RoonServer/Logs. If you don’t have a convenient place to upload them let me know and I’ll send you a PM with some instructions.
One question: When you add these folders, are you doing it by navigating from the “/” drive at the filesystem root, or from some other drive? Do you have any USB drives or similar connected?
I have vague ideas of why this might be happening, but we honestly thought we’d fixed this class of problems with build 196 or so. The fundamental problem is probably something related to our tracking of drives by device/partition uuid rather than path, which seemed like a great idea at the time but is looking worse over time.
I worked with @ben to solve this bug… there was an issue with how things started up in Roon, and your container made the initial mount stuff relatively slow, and this case never came up before.