Auto-Spindown of hdd's does not work on OMV6

    I am new to OMV and so far I have been getting along with OMV6 nicely. The disk spin-up issue, however, is the only thing I cannot solve. I have had experience with other linux distros. I have now dealt with the spin-up issue for a week, checked every blog post (I could find) and also tried other methods. I usually never write forum posts, but here I am. Can you help me?


    Collectively, I have checked various things:

    1. I restarted and reconfigured SMART more than one time.
    2. I installed and configured hd-idle instead of hdparm.
    3. I checked for any errors in various logs.

    I can spin-down my two HDDs (actually in RAID1) using either hd-idle or hdparm. Once they're shutdown, I have used btrace, to find out that smartctl is called, when the disks spin up again, usually 30-180 seconds after spin-down. I even get the same output when SMART is deactivated in OMV (which is confirmed by systemctl status smartmontools). I therefore suspect that another process is triggering smartctl. For completeness, an example output of btrace after spin-down, actually at spin-up:


    But how can I find out, which processes trigger smartctl to call on my disk (which causes it to spin-up)?

    The best idea I had so far was to check fatrace - as also suggested in various forums. fatrace does not allow me to check just the disks, so the following output is a bit messier. Please note that my system runs on a SSD (/dev/nvme0n1p2) and the output looks the same everytime before the disks spin up again. The spin-up only begins when smartctl is being called (lower part of the output). Regardless, I come to the conclusion that smartctl is called by some (ir)regular check-ups that likely also involve php-fpm7.4. Sadly I lack the skills to figure out what process calls smartctl at this point and I also cannot find any hints in the openmediavault github.


    As the output does not contain any "irregularities", error logs and such do not provide additional knowledge. There is another suggestion in the forum, which uses a script to figure out a similar problem (My Guide to Debugging Disk Spin-ups). However "block dumps" have changed and I cannot figure out how to change the script to gain more knowledge (read forum post).


    Have you got any idea how I could proceed to solve my problem? Is this somethingLooking forward to hearing your suggestions!

    I found a thread which had the same problem years ago (Hard Disk spindown), yet, a solution is also missing.


    So far I have figured out that disabling monitoring does not solve the problem, however monitoring is still a suspect of mine. If found out that starting the smartmontools daemon a second time triggers smartctl and the disks spin up. Maybe it's part of a sanity check or something. I like OMV so far and I'm learning with every step. I would still appreciate hints on what else I could check!

    henoobis How did you configure your SMART settings?


    „Power mode: Standby“ in combination with hd-idle is what stopped my disks to spin up:

    • "SMART Check Intervall" is 30 min (1800 sec)
    • "Power mode: Standby" -> Check the device unless it is in SLEEP or STANDBY mode. In these modes most disks are not spinning, so if you want to prevent a disk from spinning up each poll, this is probably what you want.


    These are the default settings:



    The following settings seem to be related to hdparm, so decided to use the following settings:


    • APM: 128 - Minimum power usage without standby (no spindown)
    • AAM: Disabled
    • Spindown time: Disabled



    hd-idle is configured the following way:


    Post
    [How-To] Spin-down hard drives with hd-idle

    INTRODUCTION

    If you like to spin-down your hard drives but you’re having trouble in doing so, you’re not alone.

    Not all hard drives like hdparm, the utility used by Debian and OpenMediaVault. There’s an alternative called hd-idle which does the job but introduces another issue: it will eventually fail if you use standby/hibernation on your server, plus other inconveniences.

    Hd-idle has been abandoned many years ago and was left in a buggy state, but we finally have a solution: a…
    auanasgheps


    My disks spin down after 20 min (1200 sec). My /etc/default/hd-idle looks as follows:

    Code
    START_HD_IDLE=true
HD_IDLE_OPTS="-i 0 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx1 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx2 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx3 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx4 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx5 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx6 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx7 -i 1200 -a /dev/disk/by-id/ata-WDC_WD100EFAX-68LHPN0_xxx8 -i 1200 -l /var/log/hd-idle.log"


    You also want to have a look at this:


    Thread
    Avoid HDD spinning up on Web Login + SSH Login by disabling to show HDD storage info on OMV Dashboard / Armbian Welcome Screen

    Let's login via SSH

    $ ssh username@arbian-omv-machine.local
    username@arbian-omv-machine.local's password:

      ___      _           _     _   _   _  ____ _
    / _ \  __| |_ __ ___ (_) __| | | | | |/ ___/ |
    | | | |/ _` | '__/ _ \| |/ _` | | |_| | |   | |
    | |_| | (_| | | | (_) | | (_| | |  _  | |___| |
    \___/ \__,_|_|  \___/|_|\__,_| |_| |_|\____|_|


    Welcome to Armbian 22.05.3 Bullseye with Linux 5.4.199-odroidxu4


    System load: 8% Up time: 22:44
    Memory usage: 14% of 1.94G IP: …
    porg


    You have to disable some HDD widgets in the Dashboard, if you don't want to spin up your disks, when you launch your dashboard.


    If this doesn’t work for you, I am out of ideas. ;)


    Regards Hoppel

    ----------------------------------------------------------------------------------
    openmediavault 6 | proxmox kernel | zfs | docker | kvm
    supermicro x11ssh-ctf | xeon E3-1240L-v5 | 64gb ecc | 8x10tb wd red | digital devices max s8
    ---------------------------------------------------------------------------------------------------------------------------------------

    Edited 4 times, last by hoppel118 ().

    Hey hoppel118,


    thanks for sharing your configuration. It really helped to confirm that I had set up everything the way you did. My hd-idle config looks like this:


    Code
    START_HD_IDLE=true

HD_IDLE_OPTS="-i 0 -a /dev/disk/by-id/ata-WDC_WD40EFPX-68C6CN0_WD-WX32A82FTP4L -i 1200 -a /dev/disk/by-id/ata-WDC_WD40EFPX-68C6CN0_WD-WXV2A82E6HNX -i 1200 -l /var/log/hd-idle.log"


    I noticed that you have WD Red's with EFAX (SMR?), while I use EFPX. Maybe they respond differently to smartcl?


    Also, currently, nothing is on my dashboard and it is not related to me logging in. It is another process that is constantly looking for those values - maybe it is related to the dashboard, but unrelated to how it is configured right now.


    Nevertheless, thank you for your help. I really appreciate those server warriors who offer their help late at night :)


    (Edit) Additional info while I'm at it:


    It is not smartctl alone that spins up the disks. If I spin down the HDDs "manually" via hdparm/hd-idle and run smartctl -i  /dev/sdx, nothing happens. I'm on it and keep you updated in case I solve the issue :)

    Too bad! But maybe my contribution will help someone else. ;)


    Do you have any media servers running that regularly check for new content? I have this completely disabled in Emby, for example.

    ----------------------------------------------------------------------------------
    openmediavault 6 | proxmox kernel | zfs | docker | kvm
    supermicro x11ssh-ctf | xeon E3-1240L-v5 | 64gb ecc | 8x10tb wd red | digital devices max s8
    ---------------------------------------------------------------------------------------------------------------------------------------

    Okay, I believe I found a potential root of the problem. smartctl is the cause of my disk spinning up and I assumed it must be related to OMV. And indeed, after some searching and interpretation, I may have found the problem in smartinformation.inc:

    github - smartinformation.inc (Jan 2022)


    smartinformation calls smartctl with the "--xall" parameter, which, called without "-n standby", causes my disks to spin up again:

    1. hdparm -y /dev/sda >> disk spun down
    2. smartctl -i /dev/sda >> outputs basic information on disk, disk does not spin up
    3. smartctl -i /dev/sda --xall >> outputs full information, but disk spins up again
    4. smartctl -i /dev/sda --xall -n standby >> outputs full information on disk spindown (disk was still spinning)
    5. hdparm -y /dev/sda >> disk spun down (standby)
    6. smartctl -i /dev/sda --xall -n standby >> "device is in STANDBY mode" - hence, no output is provided

    If I understand the OMV code correctly, smartionformation.inc does not take the standby value into account. Hence, despite what has been configured in the OMV settings, my disks will spin up.


    I'm not sure how to proceed right now. hoppel118 - do your disks also spin up/down if you follow the order of my example? Of course, it may be unintended behaviour of my disks to spin up with "--xall", yet, there's a way to prevent it. I'm still wondering why you have a different behavior in OMV6 and why you don't have these smartctl calls interfering with your disks. Let's assume I have these smartctl calls for another unknown reason. Regardless, smartctl should not spin up the disks, unless intended (e.g. a self test every 24h/week).


    Second, it appears the developers intended to implement some caching; however from what I understand the class which provides SMART information is created everytime OMV - for whatever reason - requests SMART information and therefore calls smartctl with the "--xall" parameter, again. I am not really into PHP, so I may be totally wrong. Maybe someone over here can correct me?! Otherwise, I may attempt to file a bug report with these two issues...

    Bad luck, I have the same problem...


    I have one 4 To HDD and one 10To HDD. The 4 To HDD Seagate Iron wolf NAS is spinning down with the official OMV6 way, but not the 10To Seagate drive (ST10000DM005).


    I tried to use hd-idle, but I can't stop this drive :( I used the following parameters :


    No disk monitoring on my dashboard

    SMART disabled


    The disk doesn't spin down :(


    I tried this :


    The sda disk seems to stop few milli-seconds but restart immediatly ! I tried to remove shared folders, but same thing happens...


    Could you help me ?


    Thanks

    My impression is also that SMART monitoring keeps waking up my WD Red Pro even though it being suspended and the setting within OMV for SMART monitoring being set to "Standby".

    I am using a newly install OMV 6.2.0-2 (Shaitan), had install hd-idle, hdpram, both working little weird,

    like: hdpram -C might wake HDD after force HDD spindown by "hdpram -Y", hd-idle works sometimes, but not always, by checked log.


    finally, found out that the "POWER MODE" in smart setting must set to "idle" to make spin down work.

    "standby" might cause many weird thing.


    after set to "idle", no needs to "hd-idle, hdpram" anymore.


    ------one more added-----

    close smart info in dashboard or don't stay at dashboard. it keeps HDD spining.


    ----more observation----

    following is the smart info, this disk needs to set "advence power management" to 64 in disk edit, when set to 127, disk do not spindown.

    smartctl --all /dev/sdb

    smartctl 7.2 2020-12-30 r5155 [x86_64-linux-6.0.0-0.deb11.6-amd64] (local build)

    Copyright (C) 2002-20, Bruce Allen, Christian Franke, http://www.smartmontools.org


    === START OF INFORMATION SECTION ===

    Device Model: HUH728080ALE600

    Firmware Version: A4GNT559

    Hello folks,


    I'm with the opposite situation here.

    I have 2x 3.5 HDD Sata Barracuda 2TB in mirror mode, and I can't make it last awake for more than 10 seconds before it spin down!!!


    I know the weird values of hdparm -S . I've set to 30min, and still...

    If I disable the APM, they get all on again :D

Participate now!

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

Register Yourself Login