Request: Cache-enabled user shares shouldn't report free space of cache devices

[JustinChase]

I just installed a new cache drive after my last one died

 

In setting everything back up, I looked at the shares list and noticed it's not quite right, see screenshot.

 

the documents folder has a file now sitting on the cache drive, and it's free space shows the free space of the array AND the cache drive.  I suspect this isn't by design

[archedraft]

I've noticed this as well. It only happens when there are files on the cache drive waiting to be moved by the mover. Once the mover is done then it shows that space that you would expect.

[JustinChase]

This is still a defect on the new beta 12, fyi

[stewartwb]

Given the complexity of unRAID storage configurations, there are differing opinions on the correct way to compute free space. 

See this thread http://lime-technology.com/forum/index.php?topic=28382.msg252487#msg252487

 

I think free space for a user share should be shown as the total available space remaining on all included drives.  I don't think that cache free space should be included in that total.  However, I can continue writing to a user share that is full, and the new data will sit in the cache drive, unable to be moved to the array. 

 

If a program queries free space before writing, it might abort due to lack of free space, even though there's plenty of space on the cache drive to complete the write operation.

 

So, it's possible that the amount of Free Space should be reported differently depending on how that value will be utilized.

[jonp]

Given the complexity of unRAID storage configurations, there are differing opinions on the correct way to compute free space. 

See this thread http://lime-technology.com/forum/index.php?topic=28382.msg252487#msg252487

 

I think free space for a user share should be shown as the total available space remaining on all included drives.  I don't think that cache free space should be included in that total.  However, I can continue writing to a user share that is full, and the new data will sit in the cache drive, unable to be moved to the array. 

 

If a program queries free space before writing, it might abort due to lack of free space, even though there's plenty of space on the cache drive to complete the write operation.

 

So, it's possible that the amount of Free Space should be reported differently depending on how that value will be utilized.

 

The OP is actually not a bug.  That said, it may be that we should consider changing how we report free space similar to how stewartwb is mentioning.  I'm going to move this to the roadmap as this isn't really a defect.