I had a few problems with cache drive assignments before, which led to me becoming unavailable to access the drive. This may be related to btrfs format, to the nature of cache drives, or even default "normal" behavior. What happened in two (separate) occasions before:

 

1) I had assigned a second cache disk (because I didn't understand what this was for). After unassigning the second cache disk again, the first one was no longer working

 

2) I had unassigned a cache drive and instead assigned another "new" one (for testing purposes). Once I reversed this (i.e., unassigning the "new" one and assigning the "old" one), the "old" cache drive was no longer accessible. In this example, there is just one cache drive and I never had two assigned in parallel

 

 

Any idea what I am using doing wrong? I would yet again like to change my cache drive, but having the option to go back to the "old" on if the effect is not as wished. Is this an unsupported feature?

Any ideas how the assignment of cache disks work and what to watch out for? Thanks in advance!

There were a lot of fixes to cache pool operations before 6.0 final was released.  Were your issues with cache device assignments before or after that?

Afterwards. I think with 6.1RC2, but could also be with one of the 6.0.1 releases