USB hdd won't sleep on host

oiwruk

Member
Aug 4, 2021
12
0
6
35
I have a USB WD HDD mounted using regular ntfs on the host (proxmox 8.2.2). My goal is to share it with a LXC later and then expose it through nfs and smb to other services. Before I start using the hdd in containers and VMs I want it to be able to sleep for the most of the day (yes, I want it to sleep. I am aware of what that implies for wear & tear).

Here is how I mount it on the host in /etc/fstab/:
Code:
UUID=E432716D32714598 /mnt/BackupDisk ntfs defaults,noatime,nodiratime,lazytime,nofail 0 0

It mounts as sdb and I have excluded it in /etc/lvm/lvm.conf using both id and sdb (using either individually made no difference):
Code:
global_filter=["r|/dev/zd.*|","r|/dev/rbd.*|","r|/dev/sdb.*|","r|/dev/disk/by-id/usb-WD_Elements_25A3_564748354B535547-0:0.*|"]

I have installed hd-idle and the setting in /etc/default/hd-idle is:
Code:
HD_IDLE_OPTS="-a sdb -i 180 -l /var/log/hd-idle.log"

Still, on boot the hdd sleeps but then something wakes it up and it never goes back to sleep (it is always spinning) even after days. The logs show no trace of any service accessing it. If I manually put it to sleep with hd-idle, it sleeps for a little bit and then gets woken up again.

smartd logs show:
Code:
Mar 20 12:09:46 proxmox smartd[637]: Device: /dev/sdb [SAT], is in SLEEP mode, suspending checks
Mar 20 12:39:43 proxmox smartd[637]: Device: /dev/sdb [SAT], is back in IDLE_A mode, resuming checks (1 check skipped)

What could be waking it up and how do I fix this? Many thanks