HDD management

  • Hello everyone,
    I just finished to install OMV 5 on a HP proliant gen8 microserver with 3 seagate ironwolf HDD (2*3To and 1To).

    I would want to optimize a little bit the power consumption using the HDD management option.
    As I don't use my NAS everyday, I thought that it would be a good idea to choose "minimal performance with spindown". On my mind that was good because the HDD would have been saved in standby mode most of the time.
    Problem was, in this mode, the delay to spondown is really low. Consequently, when I want to work on the data, after 1 or 2 min of inactivity, the HDD spins down. As a result their are too much spin down / up cycles on my mind.

    Then I tried to set the spindown delay on OMV setting it to 2 hours (idea was not to spin down when NAS is used but spin down during the no activity days).

    My problem is that, whatever the time I set up, and after a reboot, the parameter seems to have absolutely no effect (the HDD spins down after 1 or 2 min, 5 maximum).


    Would you have any idea about that ?



    Also I tried the "minimal power consumption without spin down" option, and with that, the HDD goes in a ghost mode (for me) :

    hdparm -C /dev/sda --> drive state is: unkown

    And whatever this mode is, it behaves like a standby mode (seems to start the HDD when trying to access it).


    Thanks a lot for your help.


    Have a good day

  • First of all sorry but from my experience here not many want to answer the topic Spindown because it is always being moved here.


    The fact is that hdparm does not work well with every HDD's. So it was in my case, for this reason I use the hd-idle instead of hdparm, it works well and reliably.

    [LibreELEC @ 2x RPi3, CoreELEC @ S12 Octa Core]

    [ NAS OMV 5.xx (Usul) @ NanoPI M4 ]

    [ Nextcloud 18.0.4 @ ODROID C2 ]

    [ Motioneye @ RPi4]

  • Try APM level 127.

    If it does not work go with the suggestion from Aux.


    Personally I had the problem, that with hdparm the load cycle count increased dramatically and the drives would have reached the spec limit (300k) within a couple of weeks. I switched to hd-idle and everything was well.

  • Thanks a lot for your information.

    I've just tried do download the hd-idle sources and installed it.

    It seems that it worked great with short time setting to test.


    Now I'm trying with a 7200s delay which seems reasonnable to me in my usecase in order to optimize power consumption and life duration of the HDDs.


    Now I have to correctly understand the SMART part of OMV (I don't really understand at that time why if the HDD are periodically probed OMV proposes forecasted tests in SMART section ... I don't get the difference between automatic probes and forecasted probes. And by the way, I don't succeed to access to the SMART reports for the moment ... but these are other topics :) I have a lot to learn !).


    I'll keep you informed for hd-idle success (or fail).

  • Hello again,
    So I just tried to set the spindown time to 7200s (2 hours) like that :

    hd-idle -a sda -i 7200 -a sdb -i 7200 -a sdc -i 7200

    Surpisingly, after 2 hours and few minutes, the disks were not standbyed ...


    I tryed with 1800s --> HDD in standby after 30 min --> OK

    I'm trying now with 3600s --> Failure / no spindown


    Do you know if there is a maximum value as spindown time ?
    From what I understood in the source code, I would say no, but ... but I checked that 2 hours was a failure ???

    Any idea ?

  • OK I just edited my previous post to indicate that with 3600 that did not work.
    I'm gonna try your solution but what will happen to the last disk which is the system disk, SSD type ?

    I was planning not to standby this one (but I don't really know if standby makes sense for a SSD).

  • Hi everyone,


    thank you all for your solutions, that perfectly works with "-i 7200" without trying to make any difference between my 3 HDD.
    To be honest, I'm a little bit frustrated because I don't understand why, but ... it works !


    Only thing that I have now to do is to make this command automatic after a reboot. That does not seem difficult with debian.


    Thanks a lot for all your help, see you later :)

  • Hey, check out the hd-idle thread. There are many hints for hd-idle configuration in general.


    Please note I have played a lot with hd-idle and you should not use the original version, it's garbage and stops working after some time if you put your server in standby. See last posts made by me, I have linked a re-written version of hd-idle which works flawlessly.


    To set individual spin-down times for HDDs, first reset times to 0 and then can set different times.

    Example: hd-idle -i 0 -a sda -i 1800 -a sdb -i 3600

    I hope it works.

    Your config is correct, should work fine.

    I use lower values but it's just a personal preference.

    OMV BUILD - MY NAS KILLER - OMV 5.x + omvextrasorg


    Core i3-8300 - ASRock H370M-ITX/ac - 8GB RAM - Sandisk Ultra Flair 32GB (OMV), 256GB NVME SSD (Docker), 3x4TB HDD (Data) - Fractal Design Node 304 - Be quiet! Pure Power 11 350W

  • Please note I have played a lot with hd-idle and you should not use the original version, it's garbage and stops working after some time if you put your server in standby.

    I installed it with a simple command: apt install hd-idle

    Would this version be a garbage that should not be used?

    It works perfectly for me so far. I am forced to use it since I installed OMV5, because the spin-down hasn’t worked.

  • I think if everything works as it should then you can be satisfied or?

    Test the thing longer then you show more good or not good.

    [LibreELEC @ 2x RPi3, CoreELEC @ S12 Octa Core]

    [ NAS OMV 5.xx (Usul) @ NanoPI M4 ]

    [ Nextcloud 18.0.4 @ ODROID C2 ]

    [ Motioneye @ RPi4]

  • I installed it with a simple command: apt install hd-idle

    Would this version be a garbage that should not be used?

    I think there are only issues if standby mode of the server (not only HDD) is used. In that case the version in the repo sometimes is not started after the server resumes from standby.


    I don't use standby and it works perfectly for me.

  • The new version also prints in syslog when disks are spindown or waken up. I think is super useful when you want to know what happened. Additionally you can retain the old configuration from the original hd-idle.

    OMV BUILD - MY NAS KILLER - OMV 5.x + omvextrasorg


    Core i3-8300 - ASRock H370M-ITX/ac - 8GB RAM - Sandisk Ultra Flair 32GB (OMV), 256GB NVME SSD (Docker), 3x4TB HDD (Data) - Fractal Design Node 304 - Be quiet! Pure Power 11 350W

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!