All HDDs are unmountable

[mflotron]

Hey all, I am trying to get my first install of unRAID up and running and have run into a pretty significant bump--none of my hard drives will mount.

 

I assign all the drives appropriately in the array, attempt to Start the array. The array starts, but then shows all drives as unmountable. I attempt to Format and after a few moments the status for every drive is still "Unmountable"

 

Here is the log from 1 of the drives right after the format attempt:

 

Jul  1 21:15:49 Edinburgh root: meta-data=/dev/md8               isize=512    agcount=5, agsize=268435455 blks
Jul  1 21:15:49 Edinburgh root:          =                       sectsz=512   attr=2, projid32bit=1
Jul  1 21:15:49 Edinburgh root:          =                       crc=1        finobt=1, sparse=0
Jul  1 21:15:49 Edinburgh root: data     =                       bsize=4096   blocks=1220942633, imaxpct=5
Jul  1 21:15:49 Edinburgh root:          =                       sunit=0      swidth=0 blks
Jul  1 21:15:49 Edinburgh root: naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
Jul  1 21:15:49 Edinburgh root: log      =internal log           bsize=4096   blocks=521728, version=2
Jul  1 21:15:49 Edinburgh root:          =                       sectsz=512   sunit=0 blks, lazy-count=1
Jul  1 21:15:49 Edinburgh root: realtime =none                   extsz=4096   blocks=0, rtextents=0
Jul  1 21:15:49 Edinburgh emhttp: shcmd (293): mkdir -p /mnt/disk8
Jul  1 21:15:49 Edinburgh emhttp: shcmd (294): set -o pipefail ; mount -t auto -o noatime,nodiratime /dev/md8 /mnt/disk8 |& logger
Jul  1 21:15:50 Edinburgh root: mount: /dev/md8: more filesystems detected. This should not happen,
Jul  1 21:15:50 Edinburgh root:        use -t <type> to explicitly specify the filesystem type or
Jul  1 21:15:50 Edinburgh root:        use wipefs( to clean up the device.
Jul  1 21:15:50 Edinburgh emhttp: shcmd: shcmd (294): exit status: 1
Jul  1 21:15:50 Edinburgh emhttp: mount error: No file system (1)
Jul  1 21:15:50 Edinburgh emhttp: shcmd (295): umount /mnt/disk8 |& logger
Jul  1 21:15:50 Edinburgh root: umount: /mnt/disk8: not mounted
Jul  1 21:15:50 Edinburgh emhttp: shcmd (296): rmdir /mnt/disk8

 

I've tried formatting using DBAN and Preclearing to no avail. FWIW, all of these drives are from a previous FreeNAS installation.

[JonathanM]

Try explicitly setting the file system type in the drive properties instead of leaving it set to auto.

[mflotron]

Curiously enough, that worked...but it does concern me quite a bit as to potentially causing issues for me down the road. The log makes me believe the drives have remnants of their ZFS past, which I definitely want to get rid of.

 

I did figure out how to use wipefs, and I did it on all the drives, but this still didn't allow me to mount the drives while on auto. It DID seem to find and erase some of the so-called "magic strings"... I got a lot of this:

 

/dev/sdd: 8 bytes were erased at offset 0x57541e36000 (zfs_member): 0c b1 ba 00 00 00 00 00

 

While we will obviously have a backup of this data (44TB...), it is crucial to our production and I want to make sure I'm not doing something now that I'm going to regret.

[itimpi]

Not sure why a preclear would not wipe any remnants of the previous usage as it involves writing zeroes to every sector on the disk.

 

I have seen cases where not doing so causes problems, particularly when the the disk file system type is set to 'auto'.  I think this is a by-product of unRAID trying to keep intact the data on disks previously. Used by unRAID and the logic around this nit being complete enough.

[JorgeB]

Not sure why a preclear would not wipe any remnants of the previous usage as it involves writing zeroes to every sector on the disk.

 

Agree, are you sure you did a full preclear?

 

Curiously enough, that worked...but it does concern me quite a bit as to potentially causing issues for me down the road. The log makes me believe the drives have remnants of their ZFS past, which I definitely want to get rid of.

 

It would be best to really clear the disks so you can use auto, if you need to do a new config sometime in the future fs setting will be reset to auto resulting in unmountable disks and you may not remember why.