Hopefully this is a case of user error, but I'm not sure what I could be doing wrong.

 

I setup v6 originally with 18 array disks (including 1 parity) and 1 cache drive.  I have 4x 160GB SSDs and I thought I would use them as a cache pool.  Unfortunately unraid doesn't seem to want to keep them in the cache pool and upon reboot, it un-assigns my cache devices:

 

http://i.imgur.com/d6Pq32M.png

 

I attached my tower diagnostics file.

 

EDIT: I utilized the cache pool more (installed some docker containers, utilizing the cache only share for some of the directories that were mapped), stopped the array, rebooted, and unraid automatically mapped the SSDs into the cache devices as they should be.  I changed the Message Icon to "Unexpected Behavior" ... I can consider this solved at my point, I'm not sure if this issue needs to be investigated further or not.

tower-diagnostics-20150619-1915.zip

Unfortunately, there's nothing to see here.  The syslog and the picture agree with what you say, that they aren't assigned.  I waited and waited to the end of the syslog, expecting to see the SSDs get assigned to the cache drive slots, but they never were!  There's evidence of the last drive assigned, a 3TB WD, but it was unassigned, and it's been empty since.  Perhaps I'm not understanding what the problem is.  Would it be possible to try again, and assign them and let's see what happens?  Then grab and attach another diagnostics file.

Unfortunately, there's nothing to see here.  The syslog and the picture agree with what you say, that they aren't assigned.  I waited and waited to the end of the syslog, expecting to see the SSDs get assigned to the cache drive slots, but they never were!  There's evidence of the last drive assigned, a 3TB WD, but it was unassigned, and it's been empty since.  Perhaps I'm not understanding what the problem is.  Would it be possible to try again, and assign them and let's see what happens?  Then grab and attach another diagnostics file.

 

Hi Rob,

 

Sorry, I should have attached a post-array start log.  The syslog is prior to starting the array, and I'm wondering if there is anything showing there indicating why the SSDs were not being assigned to the cache disk.

 

I'll reboot, re-assign the SSDs to the cache pool, start the array and grab the log then.

 

The problem itself is I can assign the SSDs as the cache pool, start the array, put contents into the cache pool, then stop the array, reboot, and when Unraid comes up again, it will have no memory of the SSDs having been previously assigned to the cache pool (and instead it tries to mount the previous cache drive which is this 3TB drive I used to use in its place).

 

Hope that made sense,

Ogi

 

 

Welp, after putting more stuff on the cache drive, I stopped/rebooted and now unraid is recognizing that the cache pool (expected behavior) ... I suppose this should be marked as solved ... but the workaround being that when you assign the cache pool, you actually need to utilize it some?