(Please do not tell me not to spin down my drives, I have been doing it for decades without issue, and my usage patterns are well suited to spinning them down.)
I have four 14TB shucked WD140EDGZ drives. They appear to be identical in every way that I have been able to observe (other than smart statistics, serial numbers, etc). Yet for some reason, only one of the four will actually spin down properly on a timer or via APM.
I set a 5 minute timer on all of them with "hdparm -S60 /dev/sd?" and all my drives spin down as expected except these three. I would assume it was a problem with this model, except one of the four of this model does spin down.
I can manually spin them down with hdparm -y, so they are certainly capable.
I am currently running them on a bare debian linux server. But they had the same behavior in another machine. I haven't installed anything on it yet because I want to get these drives working properly first. The drives have no data on them, and exhibited the same behavior with or without a partition/filesystem created on them.
I suspect that there is some setting in the firmware on 3/4 of these drives that is preventing the spin down. but so far I have had no luck figuring it out.
I already compared the output of smartctl -a and the values in /sys/block/sdX and didn't see anything different (that was relevant). Anyone have any ideas of other things to look at?