Hey Guys,

 

I have an unraid system with a ssd cache drive. My cache drive is 240GB. I allocated 120GB of that and created a cache only share and installed a windows VM in it. Then for one of my other user shares (media), i also selected to use cache (so i assume it copies to cache first then moves to the main raid).

 

i have couple of problems:

 

When told my media user share to use the cache, there was no option to say how much of the cache to use. So now it seems like its using up all the cache drive and eating into the space that was allocated for my windows VM. Currently, my windows VM says the C drive has about 60GB free (so only about 60gb used), but the unraid control panel UI says the whole cache drive only has about 10GB free (and im getting warning emails saying that the cache drive is on critical levels due to 95% full-ness).

 

For the last few hours, i dont believe anything was copied to the madia drive anyway, so i dont really understand why the cache drive is not emptying? Is there any reason why? Is there any way for me to manually empty the cache drive (obviously i dont want to empty the VM data, just the daat from the media user share)?

 

Cheers.

If you're copying tons of information over to the array, then that will indeed happen.  Under normal circumstances once a day (or however often you have set the "mover" to run) it will empty out the cache drive and move the appropriate files over to the array

And you can run Mover manually from Main - Array Operation - Run now button

Awesome thanks guys. I didnt know that was called the "mover". Maybe someone should consider renaming it to Cache Mover or just something to do with the emptying of cache. Great if you know what it is, but not intuitive for newbies, and we all start as one.

 

I have set it to run hourly now, so that should solve the problem. But wonder why this is not the default, and wonder why this is not automated instead of scheduled (maybe auto run the mover when the array is not heavily being used for say 5 minutes).

 

Its also confusing why the cache is allowed to write the whole 240gb while 120 of it is allocated to the VM. This could potentially cause issues for the VM if the drive runs out of space (and windows still thinks it has space). Is this a bug?