Posts by SerErris

Wow and just as I see it, Raid 0 was not intended. Not sure what I have setup there ...

Okay time to rebuild the raid copy all data over ...

My upgrade went very well, just having one issue:

/etc/cron.daily/openmediavault-mdadm:
mdadm: Value "omv2:0" cannot be set as devname. Reason: Not POSIX compatible. Value ignored.

This is my md output:

/dev/md0:
           Version : 1.2
     Creation Time : Mon Aug  7 19:27:47 2023
        Raid Level : raid0
        Array Size : 23441316864 (21.83 TiB 24.00 TB)
      Raid Devices : 6
     Total Devices : 6
       Persistence : Superblock is persistent

       Update Time : Mon Aug  7 19:27:47 2023
             State : clean
    Active Devices : 6
   Working Devices : 6
    Failed Devices : 0
     Spare Devices : 0

            Layout : -unknown-
        Chunk Size : 512K

Consistency Policy : none

              Name : omv2:0
              UUID : b36f5276:7a5eb43e:cfad32f3:51279d74
            Events : 0

    Number   Major   Minor   RaidDevice State
       0       8        0        0      active sync   /dev/sda
       1       8       16        1      active sync   /dev/sdb
       2       8       32        2      active sync   /dev/sdc
       3       8       80        3      active sync   /dev/sdf
       4       8       64        4      active sync   /dev/sde
       5       8       48        5      active sync   /dev/sdd

I am not sure why it is named omv2:0 and who named it that way, but clearly the name is violating POSIX standards.

It was typically /dev/md/md127 or something alike and now it is /dev/md/omv2:0

I now changed /etc/mdadm/mdadm.conf to this:
# definitions of existing MD arrays
ARRAY /dev/md/omv1 metadata=1.2 UUID=b36f5276:7a5eb43e:cfad32f3:51279d74

and also rebuild the initframfs:
myuser@openmediavault:~$ sudo update-initramfs -u

Now the command executred by anacron daily is executing without an error message.

Great thank you so much.

Did you also upgrade the OS?

In what order ? First OS, then OMV or other way around? First OMV, then OS?

The problem is: If you go static, there is no automatic for none of the parts anymore. It is either fully static or dhcpcd (automatic).

Was not serious about it and never was aware it exists. I just ordered one of the above (your link).

Yep, that would allow you with SystemRescue to dd from one device to another one. Or directly use Clonezilla

Or this

AODUKE 4Bay Fan Cooling M.2 NVME/SATA to USB3.2 10Gbps Docking Station Adapter Reading and Writing,External Hard Drive Enclosure for PCIE NGFF SSD(Cloning Are not Supported) AJM2S4

AODUKE 4Bay Fan Cooling M.2 NVME/SATA to USB3.2 10Gbps Docking Station Adapter Reading and Writing,External Hard Drive Enclosure for PCIE NGFF SSD(Cloning Are…

www.amazon.de

You can use clonezilla to make a copy of the drive and then on recovery you can automatically adjust the new drive size.

So yes, the process actually works directly.

If you do have OMV6 you can even select bootable CloneZilla entries to your boot menu, so no reason to boot from a removable device.

Yes, and you would need to replace your old drives "one after one" to actually utilize the full capacity of your new drive.

For now the new drive will never use more then 3.8 TB.

Where is the configuration file from OMV, where it does save the known file systems? Maybe they are already in there (half baked)?

can you please run the

mount

command and see what the output is?

Also you could just unmount them and they may then appear ...

Can you also please list your /etc/fstab file?

How did you create the file system on it? Via OMV GUI, or manually ?

If you have created it manually I do have the same issue. So I just created a new filesystem and then mounted it.

Is there anything on it?

Quote from golfnut324

Question: (nOOb alert!)

Appears that once a drive has been part of an omv raid, future installs of omv run afoul of latent files or partitioning structure? I noticed that installs on the used drive went to partition 2 whereas on the clean usb install went to partition 1 with successful boot into the operating system. What is best way to totally wipe the old RAID drive?

It looks like you want to reuse a drive that was part of a raid before, correct?

Very simple is to use

dd if=/dev/zero of=/dev/sdX bs=1M count=100

That erases all information in the first 100MB of the drive and therefore eliminates all RAID information. (most likely the first 4MB would be good enough anyway ...

yep ... that is why I now after 10 years rebuild my NAS.

Compared to mine

I only have 6x4TB SSD ... so from power consuption perspective I am at 18Watts max. Your disks are easily eating that up - each.

No idea.

Yes I agree, but if you look at the original post and the current configuration of that massive machine ... Not really makes any difference.

It would be a difference if you having an all SSD and 18W System. But that monster with that many spinning disks ... should be less than the loss because of the power supply.

You cannot mount a raid .. you need to create a filesystem on the raid, and then you can mount that filesystem somewhere.

So first go to Storage->Filesystems and press the + icon.

In the dialog you select the filesystem type (ext4) and then you select the raid device, where you want to create the filesystem on (/dev/md126).

Repeat it for the second raid.

Then you can go to

Storage->Filesystems again and mount the filesystem

And after that go to Storage->Shared Folders and Share that Filesystem

Then I would just go for a small SSD ..

That was the cheapest I could find in a few seconds...

Intenso Interne 2,5" SSD SATA III Top, 128 GB, 550 MB/Sekunden, Schwarz

Die SSD SATA III Top bietet Schnelligkeit und Effizienz für den Alltäglichen Gebrauch zu Hause als auch für ausgiebiges Gaming. Durch die SATA III…

www.amazon.de

This is Amazon expensive 11,xx Euros

You can find it even cheaper at 8,37 Euros ...

A pen drive is pretty much same range now and much slower and less secure (e.g. wear leveling only on very few thumb drives).

I do not think, that you will see the difference of the RAM power on your powerbill.

There is no reason not to do it. The NAS memory bandwidth requirements are so small - every RAM is massively fast enough.

So, go for it.