short version is this:
/dev/md0 is supposed to be mounted on /mnt/backup by /etc/fstab and seen in
lsblk:
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
md0 9:0 0 1.8T 0 raid1 /mnt/backup
md0 9:0 0 1.8T 0 raid1 /mnt/backup
sda 8:0 0 232.9G 0 disk
sda1 8:1 0 100M 0 part /boot/efi
sda2 8:2 0 32G 0 part [SWAP]
sda3 8:3 0 200.8G 0 part /
sdb 8:16 0 1.8T 0 disk
sdb1 8:17 0 400G 0 part /home
sdb2 8:18 0 100G 0 part /opt
sdb3 8:19 0 1.3T 0 part /data1
sdc 8:32 0 1.8T 0 disk /mnt/data2
sdd 8:48 0 1.8T 0 disk /mnt/data3
sde 8:64 0 1.8T 0 disk
sdf 8:80 0 1.8T 0 disk
sr0 11:0 1 1024M 0 rom
But, for some reason not known by me when the dirvish backup software tries
to initialize a vault the data are written to /mnt/backup in / (/dev/sda3)
which is the small SSD holding the OS. (And how I ended up filling / 100%
causing great consternation on my part; fixed thanks to Galen.)
This happened sometime Monday night when one or both of the non-RAID1 disks
in the MediaSonic Probox failed and the OS changed the RAID1 array from
/dev/sd{e,f} to /dev/sd{g,h} when I removed the failed disks Tuesday
morning. I ordered another pair of 1,8 TiB WD Red NAS disks from Amazon and
they were delivered around dinner time Tuesday (probably from the Fairview
or Troutdale wearhouses). Yesterday morning I partitioned and installed ext4
on the new disks and the OS moved /dev/md0 back to /dev/sd{e,f}.
My question is how to fix this mess. Would an mdadm function, such as
manage, allow me to truly have writes to /dev/md0 go to the RAID1 mount
point, /mnt/backup, or do I need to remove /dev/md0 and rebuild the array
from scratch?
TIA,
Rich
