Unassigned Devices - Managing Disk Drives and Remote Shares Outside of The Unraid Array

[dlandon]

1 minute ago, trurl said:

It does mean you get the choice to do either /mnt/user0 or /mnt/user, but it is confusing to call /mnt/user0 "user shares".

That's what it is.  User shares only and not Pool shares.  What would be a better choice of naming it?

[trurl]

1 minute ago, trurl said:

It does mean you get the choice to do either /mnt/user0 or /mnt/user, but it is confusing to call /mnt/user0 "user shares".

And it's not immediately obvious that the other choice in the dropdown will give you what is normally meant by user shares, so I had never even tried it.

[trurl]

1 minute ago, dlandon said:

That's what it is.  User shares only and not Pool shares.  What would be a better choice of naming it?

No, it is not User Shares only, instead it is /mnt/user0.

 

Does anybody really need /mnt/user0 for this?

[dlandon]

1 minute ago, trurl said:

No, it is not User Shares only, instead it is /mnt/user0.

What's a better description?  /mnt/user0 doesn't make any sense to the average Unraid user.

 

3 minutes ago, trurl said:

Does anybody really need /mnt/user0 for this?

Yes, if they don't want all their Pool devices to be exposed at the Root Share - appdata, system, etc.

 

Maybe user shares and pools would be a better default choice.

 

I do think the explanation on the pool add dialog is pretty clear:

 

 

[dlandon]

17 minutes ago, trurl said:

I don't understand.

Change your Root share to the "User and Pool Shares" selection and it will work as you expect.

[dlandon]

10 minutes ago, trurl said:

And it's not immediately obvious that the other choice in the dropdown will give you what is normally meant by user shares, so I had never even tried it.

I'll change the "User and Pool Shares"  choice to the default selection.  Thats more in line with what users will be doing.

[trurl]

 

14 minutes ago, dlandon said:

I'll change the "User and Pool Shares"  choice to the default selection.  Thats more in line with what users will be doing.

I think if you let the second choice still be called User Shares it is still confusing.

 

Another way to "explain" it in the dropdown would be:

(default) User Shares

User Shares excluding Pools

[dlandon]

25 minutes ago, trurl said:

 

I think if you let the second choice still be called User Shares it is still confusing.

 

Another way to "explain" it in the dropdown would be:

(default) User Shares

User Shares excluding Pools

The whole concept of /mnt/user0 shares is a bit strnge to me, but I have no conrol over that.  There was actually a point where it was considered that it be removed, but some users wanted that mount point to stay.

 

Let me take that under advisement.  To us old timers, we understand the /mnt/user0, but new users could be quite confused.

 

I put this together to see if it works any better:

Still palying with ideas to see what might work best.

[trurl]

I expect Exclusive shares to be included in User Shares, as would anybody else since they appear on the User Shares page.

 

Seems to me "User Shares" should mean exactly what appears on the User Shares page, and if you want to also provide some way to leave pool files out of the root share, that shouldn't be called User Shares.

[trurl]

Why would anyone want a root share that doesn't include pool files?

[dlandon]

31 minutes ago, trurl said:

I expect Exclusive shares to be included in User Shares, as would anybody else since they appear on the User Shares page.

I would also, but I have no control over that.  The Pool Share is a bind mount and the symlinks for exclusive shares don't work on a bind mount.

 

34 minutes ago, trurl said:

Seems to me "User Shares" should mean exactly what appears on the User Shares page, and if you want to also provide some way to leave pool files out of the root share, that shouldn't be called User Shares.

Ok, I get it.  This is what I'm thinking now:

 

[dlandon]

36 minutes ago, trurl said:

Why would anyone want a root share that doesn't include pool files?

Not sure, but there are people that use /mnt/user0/.

[Kilrah]

Never seen "user share" refer to anything other than user anywhere in the documentation.

https://docs.unraid.net/unraid-os/manual/shares/user-shares/

 

user0 is just an additional option that doesn't include pools but doesn't have a name of its own, certainly not default and has been marked as deprecated and possibly being removed for at least a year and a half when I started using Unraid...

Edited December 22, 2023 by Kilrah

[dlandon]

15 minutes ago, Kilrah said:

user0 is just an additional option that doesn't include pools but doesn't have a name of its own, certainly not default and has been marked as deprecated and possibly being removed for at least a year and a half when I started using Unraid...

It was actually removed in a test or beta version (can't remember which), and then put back.

[akazu]

Hello Community

I was able to successfully complete my backup test with an encrypted USB stick. Now I wanted to do the same with an external SSD (connected via USB) and have the problem that the disk is not mounted (currently it is unencrypted) and it is displayed as follows:

 

ec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] 1953525167 512-byte logical blocks: (1.00 TB/932 GiB)
Dec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] Write Protect is off
Dec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] Mode Sense: 23 00 00 00
Dec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] No Caching mode page found
Dec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] Assuming drive cache: write through
Dec 23 13:12:10 home-net kernel: sdg:
Dec 23 13:12:10 home-net kernel: sd 7:0:0:0: [sdg] Attached SCSI disk
Dec 23 13:12:12 home-net emhttpd: Samsung_SSD_850_EVO_1TB_S2RFNX0J312248L (sdg) 512 1953525167
Dec 23 13:12:12 home-net emhttpd: read SMART /dev/sdg

 

The view under UD

 

 

In some threads I have read that Unraid does not like disks that are connected via USB (or their controllers). Do I have the same problem here? If so, what can I do apart from extending the SATA cables and routing them out of the case?

 

Thanks in advance

 

Edited December 23, 2023 by akazu

[dlandon]

1 minute ago, akazu said:

In some threads I have read that Unraid does not like disks that are connected via USB (or their controllers). Do I have the same problem here? If so, what can I do apart from extending the SATA cables and routing them out of the case?

That disk has a file system, but does not have a partition.  UD cannot mount it.

[akazu]

12 minutes ago, dlandon said:

That disk has a file system, but does not have a partition.  UD cannot mount it.

 

Thank you for your much appreciated help. Gparted was my friend and it works now 

[BoogieSD]

Good evening, does anyone know why my UD NTFS SSD shows incorrect free drive space vs the Windows 10 VM running from it? Thanks in advanced.

 

 

Edited December 24, 2023 by BoogieSD

[dlandon]

3 hours ago, BoogieSD said:

Good evening, does anyone know why my UD NTFS SSD shows incorrect free drive space vs the Windows 10 VM running from it?

Your Windows 10 is running from a vdisk on the UD disk formatted with NTFS?

[Kilrah]

4 hours ago, BoogieSD said:

Good evening, does anyone know why my UD NTFS SSD shows incorrect free drive space vs the Windows 10 VM running from it? Thanks in advanced.

Likely instead of passing through the drive to the VM you created a vdisk file on it and told it to have a 1TB size.

[akazu]

Hey Community

 

I have mounted my encrypted USB SSD via UD and also stored the password. When I go to unmount in Unraid under Disk Devices, the disk is not removed cleanly. A look under /dev/mapper tells me that the entry from backupssd2 is still there. If I mount the disk again, the following message appears 

 

~# tail -f /var/log/syslog
Dec 25 13:17:23 home-net kernel: sd 8:0:0:0: [sdh] Mode Sense: 23 00 00 00
Dec 25 13:17:23 home-net kernel: sd 8:0:0:0: [sdh] No Caching mode page found
Dec 25 13:17:23 home-net kernel: sd 8:0:0:0: [sdh] Assuming drive cache: write through
Dec 25 13:17:23 home-net kernel: sdh: sdh1
Dec 25 13:17:23 home-net kernel: sd 8:0:0:0: [sdh] Attached SCSI disk
Dec 25 13:17:24 home-net unassigned.devices: Mounting partition 'sdh1' at mountpoint '/mnt/disks/backupssd2'...
Dec 25 13:17:24 home-net unassigned.devices: Using disk password to open the 'crypto_LUKS' device.
Dec 25 13:17:24 home-net unassigned.devices: luksOpen result: Device backupssd2 already exists. 
Dec 25 13:17:24 home-net unassigned.devices: Partition 'backupssd2' cannot be mounted.

 

From here on only a reboot helps me to remount the disk or is there another way? 

 

On my Linux client I can mount and unmount the disk cleanly.

 

Thanks in advance

 

[aim60]

I am having trouble renaming folders on exFat formatted UD mounted disks.  This is an example from Windows 10, although MacOS and Ubuntu are having issues as well.  I believe this is a recent problem, since I am often using exFat disks in UD since they are OS agnostic.

Unraid 6.11.5 or 6.12.6

UD 2023.12.15

UD Plus 2023.11.30

The share is public.

 

 

un6126p-diagnostics-20231225-1627.zip

Edited December 25, 2023 by aim60

[jr37]

I recently had to reinstall Unraid on the flash drive I have been using as my boot drive. I did this successfully, but I have no previous backup, and I didn't save any information about the exact configuration of my raid array (i.e. what was/were the parity drive(s) and data drives). I have three 1TB drives, one 2TB drive, and a 256GB SSD. I know that more than likely the 2TB drive was a parity drive, but I am unsure about the others. I followed the instructions on this page https://docs.unraid.net/unraid-os/manual/changing-the-flash-device/ and it suggested I use Unassigned Devices to figure out which drives are not mountable, as those are most likely the parity drives. The 2TB drive is in fact unmountable, but so are two of the 1TB drives (so only one of the 1TB drives and the 256gb SSD are mountable). It doesn't make sense to me that I would have 3 out of 5 drives be parity drives. The instructions on that page suggested I ask in the forums. Here is the result from the syslog file when if I try to mount the unmountable drives:

 

Dec 25 14:25:22 rupert-media unassigned.devices: Error: Device '/dev/sdf1' mount point 'ubuntu-server:0' - name is reserved, used in the array or a pool, or by an unassigned device.
Dec 25 14:25:22 rupert-media unassigned.devices: Disk with serial 'TOSHIBA_HDWD120_30FM34JAS', mountpoint 'ubuntu-server:0' cannot be mounted.
Dec 25 14:26:39 rupert-media unassigned.devices: Error: Device '/dev/sdd1' mount point 'ubuntu-server:0' - name is reserved, used in the array or a pool, or by an unassigned device.
Dec 25 14:26:39 rupert-media unassigned.devices: Disk with serial 'TOSHIBA_HDWD110_27C048VFS', mountpoint 'ubuntu-server:0' cannot be mounted.
Dec 25 14:37:16 rupert-media unassigned.devices: Error: Device '/dev/sdf1' mount point 'ubuntu-server:0' - name is reserved, used in the array or a pool, or by an unassigned device.
Dec 25 14:37:16 rupert-media unassigned.devices: Disk with serial 'TOSHIBA_HDWD120_30FM34JAS', mountpoint 'ubuntu-server:0' cannot be mounted.
Dec 25 14:37:22 rupert-media unassigned.devices: Mounting partition 'sdc1' at mountpoint '/mnt/disks/ubuntu-server:0'...
Dec 25 14:37:22 rupert-media unassigned.devices: Mount cmd: /sbin/mount -t 'xfs' -o rw,relatime '/dev/sdc1' '/mnt/disks/ubuntu-server:0'

 

When the two mountable drives are mounted, a grayed out button will appear next to the unmountable drives that says "Reboot." Any help would be appreciated! (FYI, I have rebooted the server several times, the issue persists). 

[trurl]

4 minutes ago, jr37 said:

three 1TB drives, one 2TB drive, and a 256GB SSD. I know that more than likely the 2TB drive was a parity drive

If the 2TB was in the array then it had to be parity since no data drive can be larger than parity. 

 

 

[je82]

Hi,

 

I am using Unassigned Devices to mount a secondary nvme drive and clone the data from the main nvme drive to this one but i have a question regarding the displayed used data as it does not seem to match.

 

1. I have first nvme mounted to the array in pool /mnt/nvme_appdata

2. I have used Unassigned Devices to mount the secondary nvme at /mnt/disks/Appdata_backup

3. I use rsync to clone the data from /mnt/nvme_appdata to /mnt/disks/Appdata_backup | rsync -avu --no-i-r --delete --stats --exclude=.Recyc* /mnt/nvme_appdata/Appdata/ /mnt/disks/Appdata_backup/Appdata/Appdata/

 

The problem is, the displayed used space it doesn't match! according to unraid there is an additional +20gb of data on the Unassigned Devices nvme device? As you can see the main device that is being backed up only has 164gb , yet the device backing up to has 184gb? Both devices is formated with the same type of filesystem. Rsync command runs and it completes without fail, i can see in the log from rsync that it is deleting files that are not on the main /mnt/nvme_appdata.

 

Is this some kind of weird display bug with Unassigned Devices plugin?

 

 

Actually using df -h it reports the same:

/dev/mapper/nvme0n1p1            745G  153G  593G  21% /mnt/nvme_appdata

/dev/mapper/Appdata_backup       745G  172G  574G  24% /mnt/disks/Appdata_backup

 

So probably not a unassigned devices issue, very strange, never seen anything like it, there are over 7 million objects being synced though, could there be metadata differences that causes a whooping +20gb of data? it seems crazy to me.

 

Edited December 26, 2023 by je82