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


Recommended Posts

34 minutes ago, cferra said:

it will not let me unmount - 

 

I get this error.. 

 

Dec 4 22:28:30 FERRANTEAPP1 unassigned.devices: Mount SMB share '//MEDIAVAULT/Media' using SMB default protocol.
Dec 4 22:28:30 FERRANTEAPP1 unassigned.devices: Mount SMB command: /sbin/mount -t cifs -o rw,nounix,iocharset=utf8,file_mode=0777,dir_mode=0777,uid=99,gid=100,credentials='/tmp/unassigned.devices/credentials_Media' '//MEDIAVAULT/Media' '/mnt/remotes/MEDIAVAULT_Media'
Dec 4 22:28:30 FERRANTEAPP1 kernel: CIFS: Attempting to mount //MEDIAVAULT/Media
Dec 4 22:28:30 FERRANTEAPP1 unassigned.devices: Successfully mounted '//MEDIAVAULT/Media' on '/mnt/remotes/MEDIAVAULT_Media'.
Dec 4 22:28:30 FERRANTEAPP1 unassigned.devices: Adding SMB share 'MEDIAVAULT_Media'.
Dec 4 22:28:38 FERRANTEAPP1 unassigned.devices: Removing Remote SMB/NFS share '//MEDIAVAULT/Media'...
Dec 4 22:28:38 FERRANTEAPP1 unassigned.devices: Unmounting Remote SMB/NFS Share '//MEDIAVAULT/Media'...
Dec 4 22:28:38 FERRANTEAPP1 unassigned.devices: Cannot unmount '//MEDIAVAULT/Media'. UD did not mount the device.

 

I also recreated the mounts inside the ui and it did not work - still get the same thing.

 

 

I just applied a fix that should help with this.  Update and try again.

Link to comment

The way the new mount points work is physical disks are mounted to /mnt/disks/ as before.  Remote shares are mounted to /mnt/remotes/ and a symlink created at /mnt/disks/.  Any existing mappings to /mnt/disks/ for remote share should work with no changes.  I do encourage you to do any new VM or Docker Container mappings of remote shares to /mnt/remotes/, and convert any existing mappings of remote shares to /mnt/remotes/ as you can.

 

Currently this is only applicable to 6.9.  Once I am comfortable that it is working properly, I'll apply this to the older versions of Unraid so rolling back versions of Unraid will not be a problem.

Link to comment
13 hours ago, sauerm said:

I now located the corresponding disk and changed cable and sata controller port. Message persists on reboot. I hope now it could be a controller issue. Should I post this in the forum? 😬

You're using a Marvell controller with a SATA port multiplier, both on their own should be avoided, much worse together.

Link to comment
1 hour ago, JorgeB said:

You're using a Marvell controller with a SATA port multiplier, both on their own should be avoided, much worse together.

Oh yes, just tried something cheap in the beginning. As I see everywhere, it gets more expensive quickly :)
I looked around a bit and found some posts from you and some reddit infos.
I saw you recommended a LSI 9300-8i. Would a LSI 9201-8i do it as well? I found a IT preflashed on ebay for a price I could afford I guess.
How bad is my actual situation - should I switch soon or can I spare money for better equipment? Hope I dont overuse your help :)

Link to comment

I have no idea where to ask this, so I'll try here. 

 

I have SABNZBD downloading to a folder named INCOMPLETE, this folder is on cache SSD set to use only the SSD.

When the download ends it starts unpacking the files and it is set to /mnt/disks/2TB/Downloads as RW Slave. This is the unassigned device.

 

The thing is, whenever a file is unpacked from cache to the unassigned device, all my hard drives start up. 

In file activity I can see only activity on the unassigned device.

 

So, any idea why?

Link to comment
9 hours ago, dlandon said:

The way the new mount points work is physical disks are mounted to /mnt/disks/ as before.  Remote shares are mounted to /mnt/remotes/ and a symlink created at /mnt/disks/.  Any existing mappings to /mnt/disks/ for remote share should work with no changes.  I do encourage you to do any new VM or Docker Container mappings of remote shares to /mnt/remotes/, and convert any existing mappings of remote shares to /mnt/remotes/ as you can.

 

Currently this is only applicable to 6.9.  Once I am comfortable that it is working properly, I'll apply this to the older versions of Unraid so rolling back versions of Unraid will not be a problem.

Thank you for making this much smoother!! Whole system didn't freak out when a bunch of mount points changed and now there is some migration path.

Link to comment
14 hours ago, pervel said:

Maybe add an option to not create the symlink for those that don't need the backwards compatibility. Not a big deal but it bothers my OCD seeing redundant folders. 🤪

I've added a configuration setting to turn off the symlinks if not needed/wanted.

Link to comment

The latest version of UD maps the remote shares to /mnt/remotes/.  You are encouraged to change your Docker Container and VM mappings to the new mount point.

 

If you can't do that, go the UD settings and set 'Legacy Mount Point Compatibility?' to 'Yes'.  Then unmount and re-mount the remote shares.  There will be symlinks created at /mnt/disks/ to the remote shares mounted at /mnt/remotes/.  You won't have to change any Docker Container or VM mappings.

Edited by dlandon
  • Thanks 2
Link to comment

A few months ago I added a 2.5" 500GB drive with unassigned devices. When naming the mount point I was thinking that it would prevent me from using forbidden characters and HGST 500GB 2.5" was tried. I didn't think that it would be used directly without modification. The name was accepted, but would not mount and it can not be renamed. When clicking on it in the gui nothing happens.

 

I have seen fixes which seem to prevent mistakes like this, but nothing has allowed me to rename the mount point.

 

When removing the mount point and then formatting the drive the same mount point is used.

 

Is there a way to solve this?

Link to comment
46 minutes ago, JorgenK said:

A few months ago I added a 2.5" 500GB drive with unassigned devices. When naming the mount point I was thinking that it would prevent me from using forbidden characters and HGST 500GB 2.5" was tried. I didn't think that it would be used directly without modification. The name was accepted, but would not mount and it can not be renamed. When clicking on it in the gui nothing happens.

 

I have seen fixes which seem to prevent mistakes like this, but nothing has allowed me to rename the mount point.

 

When removing the mount point and then formatting the drive the same mount point is used.

 

Is there a way to solve this?

When you add the drive, the mount point has the quote?

Link to comment

To prevent unnecessary spin-up of my array, and not to wear down my SSD cache any further I decided to add a basic USB disk to my server and use UD to create a share on it, I somehow messed up horribly and have no idea how to fix this.

Dec 6 02:01:32 Tower unassigned.devices: Adding disk '/dev/sdh1'...
Dec 6 02:01:32 Tower unassigned.devices: Mount drive command: /sbin/mount -t xfs -o rw,noatime,nodiratime '/dev/sdh1' '/mnt/disks//mnt/disks/download_cache'
Dec 6 02:01:32 Tower unassigned.devices: Mount of '/dev/sdh1' failed. Error message: mount: /mnt/disks/mnt/disks/download_cache: /dev/sdh1 already mounted or mount point busy.
Dec 6 02:01:32 Tower unassigned.devices: Issue spin down timer for device '/dev/sdh'.
Dec 6 02:01:35 Tower unassigned.devices: Format disk '/dev/sdh' with 'xfs' filesystem result: meta-data=/dev/sdh1 isize=512 agcount=4, agsize=19535700 blks = sectsz=512 attr=2, projid32bit=1 = crc=1 finobt=1, sparse=1, rmapbt=0 = reflink=1 data = bsize=4096 blocks=78142798, imaxpct=25 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0, ftype=1 log =internal log bsize=4096 blocks=38155, version=2 = sectsz=512 sunit=0 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0
Dec 6 02:01:38 Tower unassigned.devices: Reloading disk '/dev/sdh' partition table.
Dec 6 02:01:38 Tower kernel: sdh: sdh1
Dec 6 02:01:38 Tower unassigned.devices: Reload partition table result: /dev/sdh: re-reading partition table
Dec 6 02:01:38 Tower unassigned.devices: Adding disk '/dev/sdh1'...
Dec 6 02:01:38 Tower unassigned.devices: Mount drive command: /sbin/mount -t xfs -o rw,noatime,nodiratime '/dev/sdh1' '/mnt/disks//mnt/disks/download_cache'
Dec 6 02:01:38 Tower kernel: XFS (sdh1): Mounting V5 Filesystem
Dec 6 02:01:38 Tower kernel: XFS (sdh1): Ending clean mount
Dec 6 02:01:38 Tower unassigned.devices: Successfully mounted '/dev/sdh1' on '/mnt/disks//mnt/disks/download_cache'.
Dec 6 02:01:38 Tower unassigned.devices: Issue spin down timer for device '/dev/sdh'.
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 Sense Key : 0x5 [current]
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 ASC=0x24 ASCQ=0x0
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 CDB: opcode=0x2a 2a 08 12 a1 75 b2 00 00 02 00
Dec 6 02:02:09 Tower kernel: print_req_error: critical target error, dev sdh, sector 312571314
Dec 6 02:02:09 Tower kernel: print_req_error: critical target error, dev sdh, sector 312571314
Dec 6 02:02:09 Tower kernel: XFS (sdh1): metadata I/O error in "xlog_iodone" at daddr 0x12a17572 len 64 error 121
Dec 6 02:02:09 Tower kernel: XFS (sdh1): xfs_do_force_shutdown(0x2) called from line 1271 of file fs/xfs/xfs_log.c. Return address = 000000007d943554
Dec 6 02:02:09 Tower kernel: XFS (sdh1): Log I/O Error Detected. Shutting down filesystem
Dec 6 02:02:09 Tower kernel: XFS (sdh1): Please umount the filesystem and rectify the problem(s)

I assume it happened because I tried to mount it to "/mnt/disks/download_cache" not knowing that "/mnt/disks/" is already appended in front of it automatically...

Link to comment

I am setting up a new server, but have an issue.  I'm attempting to mount an external NAS as an Unassigned Device for file transfers.  I have the unraid server and NAS each connected to the same router via ethernet cable.

 

I am able to add the NAS as an Unassigned Device by:  

 

  1. Searching for (and locating) the serer
  2. Entering the Username
  3. Entering the Password
  4. Leaving the domain blank
  5. Choosing the Share directory on the NAS

 

At this point, I see the green indicator next to smb.  However, when I click "Mount", the mount box goes grey, says "Mounting" and spins.  After a few seconds, it goes back to "Mount". 


Yesterday, I was able to mount the same NAS and see the files in Krusader.  I can't figure out why it would recognize the server and give me the green indicator, but not mount properly.  I am 100% sure I entered the correct username and password as well - when I search for shares, the appropriate directories appear.  I select the public folder.  

 

I am not sure what I could be doing wrong.  The values under "Source" are //MyServer/MyShare and the value under "Mount Point" is "MyServer_MyShare". 

 

I checked the logs, and get this:

Dec 5 20:34:39 MyServer kernel: CIFS VFS: cifs_mount failed w/return code = -13
Dec 5 20:34:39 MyServer unassigned.devices: Mount of '//MyNAS/Public' failed. Error message: 'mount error(13): Permission denied Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) and kernel log messages (dmesg) '.

Dec 5 20:35:12 MyServer unassigned.devices: Mount SMB share '//MyNAS/Public' using SMB default protocol.
Dec 5 20:35:12 MyServer unassigned.devices: Mount SMB command: /sbin/mount -t cifs -o rw,nounix,iocharset=utf8,file_mode=0777,dir_mode=0777,uid=99,gid=100,credentials='/tmp/unassigned.devices/credentials_Public' '//MyNAS/Public' '/mnt/disks/MyNAS_Public'

Dec 5 20:35:12 MyServer kernel: No dialect specified on mount. Default has changed to a more secure dialect, SMB2.1 or later (e.g. SMB3), from CIFS (SMB1). To use the less secure SMB1 dialect to access old servers which do not support SMB3 (or SMB2.1) specify vers=1.0 on mount.

This is strange because I have enabled SMB security in the Unassigned Devices settings.

 

Can anyone offer some guidance on what I could be doing wrong?  I don't understand why it can find the shared directories, accept my select, but fail to mount.

Edited by JoseCheckers
Link to comment
1 hour ago, iD4NG3R said:

To prevent unnecessary spin-up of my array, and not to wear down my SSD cache any further I decided to add a basic USB disk to my server and use UD to create a share on it, I somehow messed up horribly and have no idea how to fix this.


Dec 6 02:01:32 Tower unassigned.devices: Adding disk '/dev/sdh1'...
Dec 6 02:01:32 Tower unassigned.devices: Mount drive command: /sbin/mount -t xfs -o rw,noatime,nodiratime '/dev/sdh1' '/mnt/disks//mnt/disks/download_cache'
Dec 6 02:01:32 Tower unassigned.devices: Mount of '/dev/sdh1' failed. Error message: mount: /mnt/disks/mnt/disks/download_cache: /dev/sdh1 already mounted or mount point busy.
Dec 6 02:01:32 Tower unassigned.devices: Issue spin down timer for device '/dev/sdh'.
Dec 6 02:01:35 Tower unassigned.devices: Format disk '/dev/sdh' with 'xfs' filesystem result: meta-data=/dev/sdh1 isize=512 agcount=4, agsize=19535700 blks = sectsz=512 attr=2, projid32bit=1 = crc=1 finobt=1, sparse=1, rmapbt=0 = reflink=1 data = bsize=4096 blocks=78142798, imaxpct=25 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0, ftype=1 log =internal log bsize=4096 blocks=38155, version=2 = sectsz=512 sunit=0 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0
Dec 6 02:01:38 Tower unassigned.devices: Reloading disk '/dev/sdh' partition table.
Dec 6 02:01:38 Tower kernel: sdh: sdh1
Dec 6 02:01:38 Tower unassigned.devices: Reload partition table result: /dev/sdh: re-reading partition table
Dec 6 02:01:38 Tower unassigned.devices: Adding disk '/dev/sdh1'...
Dec 6 02:01:38 Tower unassigned.devices: Mount drive command: /sbin/mount -t xfs -o rw,noatime,nodiratime '/dev/sdh1' '/mnt/disks//mnt/disks/download_cache'
Dec 6 02:01:38 Tower kernel: XFS (sdh1): Mounting V5 Filesystem
Dec 6 02:01:38 Tower kernel: XFS (sdh1): Ending clean mount
Dec 6 02:01:38 Tower unassigned.devices: Successfully mounted '/dev/sdh1' on '/mnt/disks//mnt/disks/download_cache'.
Dec 6 02:01:38 Tower unassigned.devices: Issue spin down timer for device '/dev/sdh'.
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 Sense Key : 0x5 [current]
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 ASC=0x24 ASCQ=0x0
Dec 6 02:02:09 Tower kernel: sd 7:0:0:0: [sdh] tag#0 CDB: opcode=0x2a 2a 08 12 a1 75 b2 00 00 02 00
Dec 6 02:02:09 Tower kernel: print_req_error: critical target error, dev sdh, sector 312571314
Dec 6 02:02:09 Tower kernel: print_req_error: critical target error, dev sdh, sector 312571314
Dec 6 02:02:09 Tower kernel: XFS (sdh1): metadata I/O error in "xlog_iodone" at daddr 0x12a17572 len 64 error 121
Dec 6 02:02:09 Tower kernel: XFS (sdh1): xfs_do_force_shutdown(0x2) called from line 1271 of file fs/xfs/xfs_log.c. Return address = 000000007d943554
Dec 6 02:02:09 Tower kernel: XFS (sdh1): Log I/O Error Detected. Shutting down filesystem
Dec 6 02:02:09 Tower kernel: XFS (sdh1): Please umount the filesystem and rectify the problem(s)

I assume it happened because I tried to mount it to "/mnt/disks/download_cache" not knowing that "/mnt/disks/" is already appended in front of it automatically...

Turn off auto mount for the disk, reboot the server, then change the mount point without the //mnt/disks/.

 

I'll look into blocking this in the future.

Link to comment
7 minutes ago, JoseCheckers said:

I am setting up a new server, but have an issue.  I'm attempting to mount an external NAS as an Unassigned Device for file transfers.  I have the unraid server and NAS each connected to the same router via ethernet cable.

 

I am able to add the NAS as an Unassigned Device by:  

 

  1. Searching for (and locating) the serer
  2. Entering the Username
  3. Entering the Password
  4. Leaving the domain blank
  5. Choosing the Share directory on the NAS

 

At this point, I see the green indicator next to smb.  However, when I click "Mount", the mount box goes grey, says "Mounting" and spins.  After a few seconds, it goes back to "Mount". 


Yesterday, I was able to mount the same NAS and see the files in Krusader.  I can't figure out why it would recognize the server and give me the green indicator, but not mount properly.  I am 100% sure I entered the correct username and password as well - when I search for shares, the appropriate directories appear.  I select the public folder.  

 

I am not sure what I could be doing wrong.  The values under "Source" are //MyServer/MyShare and the value under "Mount Point" is "MyServer_MyShare". 

 

I checked the logs, and get this:


Dec 5 20:34:39 MyServer kernel: CIFS VFS: cifs_mount failed w/return code = -13
Dec 5 20:34:39 MyServer unassigned.devices: Mount of '//MyNAS/Public' failed. Error message: 'mount error(13): Permission denied Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) and kernel log messages (dmesg) '.

Dec 5 20:35:12 MyServer unassigned.devices: Mount SMB share '//MyNAS/Public' using SMB default protocol.
Dec 5 20:35:12 MyServer unassigned.devices: Mount SMB command: /sbin/mount -t cifs -o rw,nounix,iocharset=utf8,file_mode=0777,dir_mode=0777,uid=99,gid=100,credentials='/tmp/unassigned.devices/credentials_Public' '//MyNAS/Public' '/mnt/disks/MyNAS_Public'

Dec 5 20:35:12 MyServer kernel: No dialect specified on mount. Default has changed to a more secure dialect, SMB2.1 or later (e.g. SMB3), from CIFS (SMB1). To use the less secure SMB1 dialect to access old servers which do not support SMB3 (or SMB2.1) specify vers=1.0 on mount.

This is strange because I have enabled SMB security in the Unassigned Devices settings.

 

Can anyone offer some guidance on what I could be doing wrong?  I don't understand why it can find the shared directories, accept my select, but fail to mount.

You are getting a permission denied.  That's normally from a credentials problem.  Watch for special characters in the password.  Keep it simple with letters and numbers.

Link to comment
12 hours ago, dlandon said:

I've added a configuration setting to turn off the symlinks if not needed/wanted.

Thank you. The /mnt/disks folder is still there and it says Device or resource busy if I try to delete it manually. What's the best way to delete it completely?

Link to comment
7 minutes ago, pervel said:

Thank you. The /mnt/disks folder is still there and it says Device or resource busy if I try to delete it manually. What's the best way to delete it completely?

That folder stays there.  Don't delete it.  It is used for disk mounts.

 

That setting in UD is only to not add remote share mount symlinks to the /mnt/disks/ folder.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.