Expanding my storage. Need suggestions.

  • Hello,


    Just want some feedback please.
    I am running out of space quick-fast in my micro server and have purchased larger hard drives. I have 1x 8tb drive for media (including plex database), 1x 8tb for downloads, 1x 6tb for backup (photos, home videos, documents e.t.c.) and lastly 1x 6b drive (a backup of my other 6tb drive via rsync). I was going to shuffle my drives around (purchase 2 10tb drives, move the 2x 8tb drives to them and then use my 8tb drives as my backup drives (copy data from the 6tb drives).
    I want to avoid setting up my shares again, setup plex again etc. Can I simply clone the data from one disk to another (e.g. clonezilla)?
    Would this keep OMV happy?


    Regards


    George

  • Assuming your microserver has 4 bays, keeping 2 backups of the same data in the same device isn't much of a backup. If that 6TB backup drive is already a backup of other data, you might want to consider putting the second backup drive in an external USB enclosure that you could move to a different location (attached to a router for example, ideally at a friend's house).


    You could also consider mergerfs drive pooling, with the Union Filesystems plugin. This combines your larger drives into a single virtual drive, which also combines the empty space together. You can add and remove drives to the pool as you wish. You can leave the Plex database where it is, and just point the libraries to the mergerfs drive.

  • Thanks for the reply Nibb.
    My microserver has 4 bays plus an icy dock (2 drives, one for the os and 1 I have used for offsite backup, swap this out).
    I can’t remember the last time I rotated the offsite disk (mainly don’t use it by it won’t fit all the data, it is a 3tb drive).
    I will take a look at union fs.... but I would still need to copy the data ?


    Can you easily keep track of what folders are on what drive with union file systems? I’m worried it might confuse me

  • Mergerfs is a layer on top of the actual filesystem. It simply creates another drive on your system that merges together the contents of all the drives in the pool. Those drives are left intact and can still be accessed individually, so there is no need to copy or move the data.


    Once you have a pool like that, you shouldn't have to worry about what data is on what drive, that's the whole point. It acts as a single big disk. But you can still look at the contents of each drive if you really want to.

  • Ok Nibb I think I’m getting my head around this. Just a couple more questions.
    If the os drive fails, will the pool work with a clean reinstall?
    If one disk fills up eg. Plex drive, will data going on next drive maintain the same folder structure (if I where to look at individual drives)?
    Lastly, if everything appears as one single disk what about backup copies of files .
    E.g I have my Aperture photo library backed up on my two 6tb (using rsync ) from my Mac.


    Cheers again

  • If the OS drive fails, your individual drives remain accessible as individual drives. You recreate the pool and your back to business.
    For where new files are written, you define the policy: it can be in existing folders, or it can go to the drive with the most space, etc...)
    Duplicate files (same folder, same filename) cause ambiguity, so you have to avoid them as much as possible. There are tools for deduping, but it's best to just rename folders before you create the pool and always go through the mergerfs drive when writing files.
    I wouldn't pool together two backups in the same pool. That would be pointless. You should keep your backup on another system (rsync to an SMB share on your router for example, if your router supports a USB drive).

Jetzt mitmachen!

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