keneko

I also had this occur with my disklocation plugin after updating it, however it was not resolved for me after forcing SMART + DB. I'm seeing /dev/sg0 show as UNKNOWN, while the remaining devices are ACTIVE or IDLE...

I’d love to see improved SMB performance with MacOS clients. As far as I’m aware this isn’t an issue with (most) other systems hosting SMB shares, but for UnRaid, performance is absolutely terrible unless you go and make a dozen tweaks to files buried both on the client and server, but even then performance is lackluster. Additionally, I’m not sure if this is already in the pipeline (or even beta) or not, but native ZFS snapshot support with scheduling for snapshots and retention limits. One other feature I’d like to see could be implemented a few different ways, but at a high level, ways to modify pools/arrays without taking the whole system offline. E.g. being able to ‘queue’ changes to these which will then be automatically applied once the array stops, as well as the ability to stop individual pools/arrays without affecting the whole system. One caveat with this is it would also need to check for docker containers and vms which utilize that pool/array and either stop them first or require the user to stop them manually. Related to this is also the ability to add new pools without stopping the array.