JBOD = Linear?? Can I add additional HDDs in the future?

1. offizieller Beitrag

    Hello friends of OMV :)


    As the title told, i am not sure if OMV supports JBOD?... I'm new at OMV, but as I found in the web, JBOD is called "Linear" in OMV.

    Is this right?


    And if i choose "Linear", and make a huge Volume... Can a add additional HDDs in the future to expand these Volume?


    Also don't know how to change a HDDs in the future to a larger one...


    I know the disadvantages of JBOD. But I use my OMV NAS only as a BackUp NAS and don't need another RAID.

    I want to use the full capacity of all my HDDs one after another... and I want to expand these Space in the future by changing some HDDs or adding new ones.


    Can anyone tell my if this is possible in OMV and if Linear = JBOD :)


    Thank U

    Kind regards

    7.0.4-2 (Sandworm) // ASRock J5040-ITX with Intel(R) Pentium(R) Silver J5040 CPU @ 2.00GHz // be quiet! ATX 500W (BN46) // 2x4GB Samsung M471A5244BB0-CRC

    Fractal Design Node 304 // 4-Port PCIE to SATA 3.0 ControllerCard // 1x 128GB SSD for OS and 2x 4TB, 1x 5TB and 3x 8TB HDD as one big fuse.mergerfs space (no need for raid parity)

    • Offizieller Beitrag
    Zitat von Kamikaze01

    Can anyone tell my if this is possible in OMV

    No

    Zitat von Kamikaze01

    Linear = JBOD

    Yes


    To answer in more depth,

    can you expand the capacity by adding more drives -> No

    can you change drives to increase capacity -> No


    Another option might be to use the unionfilesystem plugin this uses mergerfs and information can be found here various options can be set within the plugin itself.

    Thank u for the answer and informations.


    I just give mergerfs a try and it looks good.

    But do you know what happend if I loose one of my HDDs?

    I am okay with loosing one HDD with its data's. There is my productive Synology NAS and another Backup on external drives (physically separated).

    I just don't want to loos ALL my data (like in RAID 0).


    So with mergerfs I can pool all my HDDs to one big Volumen and I can resize the Volumen afterwards, can add or change some of the HDDs and if one HDD crashes, i will lose the data only of this one HDD.


    But how can I replace the crashed HDD, when I don't know which files are stored on this HDD (because I only see one big pool)?


    Maybe there is a kind of "HowTo" for mergerfs? I am kind of new with this... :/


    Thank u so far and sorry for my many questions...

    7.0.4-2 (Sandworm) // ASRock J5040-ITX with Intel(R) Pentium(R) Silver J5040 CPU @ 2.00GHz // be quiet! ATX 500W (BN46) // 2x4GB Samsung M471A5244BB0-CRC

    Fractal Design Node 304 // 4-Port PCIE to SATA 3.0 ControllerCard // 1x 128GB SSD for OS and 2x 4TB, 1x 5TB and 3x 8TB HDD as one big fuse.mergerfs space (no need for raid parity)

    • Offizieller Beitrag
    Zitat von Kamikaze01

    Maybe there is a kind of "HowTo" for mergerfs? I am kind of new with this

    I don't think there is full blown how to and your questions are above my understanding that's why I referred to github, further searching in relation to the author I found this


    Basically mergerfs creates a single mountpoint for a group/set of drives, if SMART is enabled this should give a warning of a potential drive failure. However, if 1 drive fails without warning then I would assume you lose the data, unlike a Raid setup.

    Thank u a thousand times for your efforts to help me :) I will try to dig a little deeper into this.


    As I understand for now: Linear (JBOD) summarises an amount of HDDs to one big pool. If a loos a HDD I will loose only the data of this HDD. But I can not change the size of the pool or add/change/replace a HDD after first creating...


    Mergerfs is similar.

    It mount some Folders to one big Folder. If I loose one HDD, I will loose the data on this HDD. But if I create one Volume for each HDD, I will be able to add/change/resize my mounted big Folder in the future...


    Anyway: I will research a little more and what is the best way for me to use mergerfs.


    Thanks a lot for your time and help!!!


    :thumbup:8)

    7.0.4-2 (Sandworm) // ASRock J5040-ITX with Intel(R) Pentium(R) Silver J5040 CPU @ 2.00GHz // be quiet! ATX 500W (BN46) // 2x4GB Samsung M471A5244BB0-CRC

    Fractal Design Node 304 // 4-Port PCIE to SATA 3.0 ControllerCard // 1x 128GB SSD for OS and 2x 4TB, 1x 5TB and 3x 8TB HDD as one big fuse.mergerfs space (no need for raid parity)

    Just to muddy the water a bit. If you feel comfortable working from the command line via ssh then you could consider using btrfs. After it is setup, you can use the OMV interface to create shares, etc.


    With btrfs you would get the ability to:

    - create a group of drives, of same or various sizes, similar to JBOD

    - add drives to the group

    - remove drives from the group

    - convert the group to raid1, raid5 or raid6 on the fly

    - convert back to "JBOD", ie. remove raid, on the fly


    If you don't configure a raid setup, then if you lose a disk without warning you lose the data on that disk. If you get a warning you can use a couple of methods to remove/replace that disk from the group and automatically redistribute the data.


    https://btrfs.wiki.kernel.org/…rfs_with_Multiple_Devices

    Thank u for this tip... and 'mudding' the wanter :P


    I thin i am already on btrfs. I just add each single HDD via the "Discs" Menue in the left and mount it in the "File System" Menue.

    I dont need comandline via ssh for this... or do i misunderstood something?


    This is my system on the VM at the moment.

    5 x 40GB Disc and 1 x 45GB Disc for testing purpose (Data1 - 5)


    And this is my "Discs" Menue



    And in the "Union File System" (Plugin) i can put them all together into one big Volume.


    Here i will be able to administrate this in the future (add HDDs, change or replace them...).

    But i will have ONE BIG Shared Folder for my Backups :):thumbup: with the full capacity of my HDDs.

    So the UnionFileSystem Plugin manage that and put all the Data with the policy "Existing path, most free space." on my HDDs.

    Great...


    BUT (sure... there is a BUT...):

    IF i loos one HDD without warning... HOW do i know which data i had lost?!?!?

    I don't know which data exactly are stored on each HDD :/

    so if one HDD is lost, i will have to replace this HDD and run the next backup job (and hopefully only copy the missing files to the new HDD).

    is this right?


    :|

    Bilder

    7.0.4-2 (Sandworm) // ASRock J5040-ITX with Intel(R) Pentium(R) Silver J5040 CPU @ 2.00GHz // be quiet! ATX 500W (BN46) // 2x4GB Samsung M471A5244BB0-CRC

    Fractal Design Node 304 // 4-Port PCIE to SATA 3.0 ControllerCard // 1x 128GB SSD for OS and 2x 4TB, 1x 5TB and 3x 8TB HDD as one big fuse.mergerfs space (no need for raid parity)

    If you lose a disk then you will have to replace the files on your own. Depending on how you do your backup then this may be done relatively quickly (eg. rsync), or you may have to do a comple backup.


    I don't know anything about Union File System. To do this from a pure btrfs setup you would ssh in and type the following commands:


    Note: This will remove all existing data from drives /dev/sdb through /dev/sdf

    Note: "BTRFS1" after -L is the label. You can make it whatever you want.

    Code
    mkfs.btrfs -L BTRFS1 -m raid1 -d raid1 /dev/sdb /dev/sdc /dev/sdd /dev/sde /dev/sdf

    After a reboot, you will have a file system:

    Code
    /srv/dev-disk-by-label-BTRFS1

    which you can then use like any other OMV file system.


    There is more useful information in starting at post #12 of

    Any roadblocks I should be aware of prior to trying to partition the drive(s) & setup a btrfs on seperate partition(s) of individual drives?

Jetzt mitmachen!

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

Benutzerkonto erstellen Anmelden