HP Hardware RAID status in smart?

  • Is there anyway I can get the storage->s.m.a.r.t GUI to display the smart status of my HP array?


    I've upgraded the smartmontools package to v6.4 from the wheezy-backports repository and I can query the smart status of the individual drives without issue.



    The problem I have is that no matter what changes I make to smartd.conf, the GUI always incorrectly shows /dev/sda (which is the logical disk).


    I've tried adding the following to /etc/default/openmediavault and rebooting, but it doesn't seem to make a difference


    Code
    OMV_SMARTMONTOOLS_DEFAULTDIRECTIVES=-a -d cciss -o on -S on -T permissive


    After adding the -d option I get the following if I attempt to save any changes from the GUI;

    Code
    Failed to execute command 'export LANG=C; omv-mkconf smartmontools 2>&1': /usr/sbin/omv-mkconf: 51: /etc/default/openmediavault: -d: not found



    Can anyone point me in the right direction to be able to get this displayed in the GUI?
    Where does the GUI enumerate the information for the drives it displays and can this be modified?


    I've already added the HP RAID package from OMV-Extras, but this doesn't provide an email mechanism, to let me know when a drive is faulty.


    I've also tried upgrading to stone burner as I see other people with other RAID controllers had their issue resolved by upgrading, but it didn't make a difference.


    Thanks.

  • Use the HP Raid Plugin


    As as I stated, I've already added the HP RAID package from OMV-Extras, but this doesn't provide an email mechanism, to let me know when a drive is faulty. I don't see the point of a hardware status monitor if I have to manually login to get the status of it. It could've been failed for a month before I realise. The whole point of this is to get email notification when a drive is classed as failing or failed.


    i could manually script it to run the hpaculi script via cron, but it would be nice to have it all integrated into openmediavault and managed from the GUI.

  • OMV core does not support that because it only sees the logical volumes, not the physical ones of an HW RAID. This is because this requires specific hardware related implementations and is not possible in a generic way without HW drivers, etc ....


    Thanks for taking the time to reply, but I'm not sure I understand when you talk about hardware related implementations and HW drivers?


    This is an array that's already recognised and used within OMV and I've proved that I can query the smart status of individual drives within the array just by using smartctl. No additional hardware or drivers would be required.
    The problem appears to be in the way that the GUI enumerates the drives, preferring logical over physical.
    I was just wondering if there was a way that I could amend something like a config file so that it would use the physical rather than logical device names or even manually add hardcoded device paths.


    Perhaps you could explain what the following environment variables do exactly and if they are going to bear any relevance on what the GUI displays;


    smartmontools: OMV_SMARTMONTOOLS_CONFIG=/etc/smartd.conf
    smartmontools: OMV_SMARTMONTOOLS_DEFAULT=/etc/default/smartmontools
    smartmontools: OMV_SMARTMONTOOLS_DEFAULTDIRECTIVES=-a -o on -S on -T permissive


    As I said, I've tried hardcoding the devices in smartd.conf which seems to work as far as smartctl is concerned, but I still can't see the info in the GUI.
    I would have thought that using the environment variable OMV_SMARTMONTOOLS_DEFAULT and hardcoding the paths for the physical devices in smartd.conf would have been enough based on what I've been able to ascertain from google searches, but this still doesn't seem to work as expected.


    Thanks.

    • Offizieller Beitrag

    The environment variables are not related to the GUI.


    OMV only detects the logical volumes. To detect the physical volumes behind the HW RAID you have to use CLI commands like tw_cli, sas2ircu or arcconf to detect them. As far as i know there is no way to get the necessary information via sysfs or something like that. This means the manufacturer CLI commands must be used, and here we are at the point that it is not possible to retrieve the information in a generic way and a special implementation in the OMV backend is required.


    Right at the moment only devices that are detected via the kernel are supported. I do not think that this will change in near future.

  • a possibility is a script and a cron Job. Google search tags: hpaculi email script. you'll find a few good scripts.


    Haha, that's the 2nd time in the same thread that you've recommended something that I've already stated in a previous post.
    Are you just here to requote what other people post? :P

  • To detect the physical volumes behind the HW RAID you have to use CLI commands like tw_cli, sas2ircu or arcconf to detect them.


    Maybe to detect them, but not to access them as you can see here:


    smartctl /dev/sg1 -a -d cciss,0


    How about a "manual smart feature" that would allow to choose the /dev/xxx and set the -d vendor,X setting? I know that is not generic... but it would be a solution after all.


    @ryecoaaron ?


    Greetings
    David

    "Well... lately this forum has become support for everything except omv" [...] "And is like someone is banning Google from their browsers"


    Only two things are infinite, the universe and human stupidity, and I'm not sure about the former.

    Upload Logfile via WebGUI/CLI
    #openmediavault on freenode IRC | German & English | GMT+1
    Absolutely no Support via PM!

    Einmal editiert, zuletzt von davidh2k ()


  • That would be awesome if we could manually edit the smart settings. Maybe have an auto or manual setting though, as I'm sure most people will be happy with the auto setting and not want to mess around with it.

  • I would love to see this as well, I have a dell branded raid card which is actually LSI and I can also see the drives using megaraid drivers, but if I add the recommended to smartd.conf it gets removed


    Code
    smartctl -a -d megaraid,4 /dev/sdb
    smartctl -a -d megaraid,5 /dev/sdb
    smartctl -a -d megaraid,6 /dev/sdb
    smartctl -a -d megaraid,7 /dev/sdb



  • Any chance for this "manual smart feature" in OMV3 or OMV4? this will solve all trouble with ARECA, LSI, 3Ware etc which is supported by Smartmontools: https://www.smartmontools.org/…upported_RAID-Controllers


    https://www.smartmontools.org/…montools/smartd.conf.5.in

  • @votdev Maybe Volker can work on something, but I bet he wouldn't like it much.


    Greetings
    David

    "Well... lately this forum has become support for everything except omv" [...] "And is like someone is banning Google from their browsers"


    Only two things are infinite, the universe and human stupidity, and I'm not sure about the former.

    Upload Logfile via WebGUI/CLI
    #openmediavault on freenode IRC | German & English | GMT+1
    Absolutely no Support via PM!

    • Offizieller Beitrag

    @votdev Maybe Volker can work on something, but I bet he wouldn't like it much.


    Greetings
    David

    Hehe, you are right. I don't like this idea because as I have really often said, OMV is no webmin replacement. Everything should be done automagically. Or do you see such selection boxes in Synology or other NAS solutions? If such a feature will be implemented it MUST be done magically without user interaction.

  • Hehe, you are right. I don't like this idea because as I have really often said, OMV is no webmin replacement. Everything should be done automagically. Or do you see such selection boxes in Synology or other NAS solutions? If such a feature will be implemented it MUST be done magically without user interaction.


    Here is an example: https://forum.rockstor.com/t/s-m-a-r-t-for-usb-drives/834/9


    it has a short edit button on the SMART menu where one can add specific "-d "requirements for each drive.


    I see Rockstor as the closest competitor for OMV, unfortunately, it is a bit too buggy and BTRFS is way off being production ready, which is their required filesystem.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!