Upgrade from OMV 4 to 7 / RAID1 VS RAID5 / Data Migration?

Quote from Krisbee

Your BTRFS filesystem look OK. The "details" concatenates output from three separate BTRFS commands:

btrfs fi sh mountpoint

btrfs fi df mountpoint

btrfs dev stats mountpoint

The output of the last command should show dev stats for all four individual devices, but is curtailed.

Display More

1.) Yes, you're right I tested it and those 3 commands you mentioned including the mount point show exactly the same data:

btrfs fi sh /srv/dev-disk-by-uuid-73180add-31d0-4cac-b6ce-8a218144cd63

btrfs fi df /srv/dev-disk-by-uuid-73180add-31d0-4cac-b6ce-8a218144cd63

btrfs dev stats /srv/dev-disk-by-uuid-73180add-31d0-4cac-b6ce-8a218144cd63

Quote from Krisbee

If you want a fully picture of the way data and metadata have been allocated. etc use: btrfs fi us mountpoint.

2.) Yes, the command you mentioned including the mount point shows more data which is as follows,
including that the metadata is also running in RAID1:

root@nas:~#
btrfs fi us /srv/dev-disk-by-uuid-73180add-31d0-4cac-b6ce-8a218144cd63
Overall:
    Device size:                  10.92TiB
    Device allocated:              6.34TiB
    Device unallocated:            4.57TiB
    Device missing:                  0.00B
    Device slack:                    0.00B
    Used:                          6.27TiB
    Free (estimated):              2.32TiB      (min: 2.32TiB)
    Free (statfs, df):             1.91TiB
    Data ratio:                       2.00
    Metadata ratio:                   2.00
    Global reserve:              512.00MiB      (used: 0.00B)
    Multiple profiles:                  no

Data,RAID1: Size:3.17TiB, Used:3.13TiB (98.89%)
   /dev/sda        2.21TiB
   /dev/sdb        2.21TiB
   /dev/sde        1.36TiB
   /dev/sdc      553.00GiB

Metadata,RAID1: Size:5.00GiB, Used:4.41GiB (88.17%)
   /dev/sda        3.00GiB
   /dev/sdb        4.00GiB
   /dev/sde        2.00GiB
   /dev/sdc        1.00GiB

System,RAID1: Size:8.00MiB, Used:480.00KiB (5.86%)
   /dev/sda        8.00MiB
   /dev/sdb        8.00MiB

Unallocated:
   /dev/sda      523.51GiB
   /dev/sdb      522.51GiB
   /dev/sde        1.36TiB
   /dev/sdc        2.19TiB

Quote from Krisbee

As you built and copied data to your 4 device BRTFS RAID1 filesystem ins stages, I'd do one full balance now. It will take a couple of hours to complete. You should not need a full balance in the future unless disks are replaced. A regular "filtered" balance of data only at say 5%, in step with your backup schedule would be useful. See: https://wiki.tnonline.net/w/Btrfs/Balance.

Stick with RAID1 for data and metadata.

Running into a "no space left" error (ENOSPC) on BTRFS can be a pita. A kind of catch-22, when you can't delete stuff because there's not enough unallocated space to increase the number of metadata chunks needed for the deletions. A consequence of BTRFS first allocating space in 1GB chunks, before filling those with extents. You can be bitten by ENOSPC during a balance too, because of the "working space" it needs. See: https://wiki.tnonline.net/w/Btrfs/ENOSPC

So, enable email notifications and "edit" your filesystem in the WEBUI to set a usage warning threshold at say 85%.

Display More

3.) Okay, I will stick with RAID1 for data and metadata and not switch to RAID1 for data and RAID1C3 for metadata.

I have enabled email notifications via the OMV GUI (="System" -> "Notification" -> "Settings" by adding the respective data from my email provider accordingly and sending the test email worked fine).

I have also checked via the OMV GUI that such a usage warning is set for my filesystem (="Storage" -> "File Systems" -> Button "Edit" -> "Usage Warning Threshold *" set to "85%").

However, this was selected and activated by default already.

Quote from Krisbee

As you built and copied data to your 4 device BRTFS RAID1 filesystem ins stages, I'd do one full balance now. It will take a couple of hours to complete. You should not need a full balance in the future unless disks are replaced. A regular "filtered" balance of data only at say 5%, in step with your backup schedule would be useful. See: https://wiki.tnonline.net/w/Btrfs/Balance.

4.) With "full balance" you mean the following command according to the link you've posted, right?:

btrfs balance start

Okay, I will try to schedule a regular "filtered" balance of data only at 5% with my backup schedule.

Is there a way to see the progress during running such a balance or when it has been successfully completed?

To be honest, the last time I just left it running for a couple of hours until I could also hear no read/write activity from the HDDs anymore.