Problems with disk APM

I'm having a spindown-up problem in the disks of my snapraid array.
I have four disks in the SR array and six in ZFS. As I only use the server at nights, I thought about putting the disks to sleep.
I configured all disks to APM = 1 and 20 minutes wait.

Then I noticed that there was heavy spin up-down, and thought that the drives in the ZFS where the culprit, as they are used by virtualbox, owncloud, mysql... so I deactivated APM and time on them.
But I continued hearing disks waking and sleeping so I started to monitor them. I tried to dump all hdd activity to syslog, using iotop and a custom script that showed processes, files and device, but didn't find anything being accessed on the SR array. All the activity was on the system disk.

So I deactivated all APM options on all disks, but guess what, two disks continue to respawn. Those two disks are the last ones I bought. They are 2tb seagates and are sleeping for 1-2 minutes, then wake and immediately sleep again, in loop.
I'm afraid they will break soon.

Is that a problem with OMV disk management for newer disks? Anyone knows how to stop this?

OMV is failing to set APM to disabled in those two drives:

Zitat

root@DeathShadow:~# hdparm -I /dev/sdg | grep level
Advanced power management level: 1

I've manually deactivated the APM on those disks and they stay online without problems.

Then, I've manually stopped all the disks and setup a script to watch for accesses. They wake after some minutes but nothing is logged.

@ryecoaaron Could that accesses not being logged been at the poolshare and not registered as is by AUFS?