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

[Revan335]

1 hour ago, dlandon said:

Post diagnostics.

On the Way!
Maybe you need the mirrored syslog from flash drive?

I have rebootet the Unraid Server before you posted.

I the USB Disk are not showed after the Reboot, i pluged this to a Linux Client and thats worked. I can access and managed Data on there. I have checked the FS with GParted no Errors.

Than i pluged the USB Disk to the Unraid Server and he was showed to UD.

 

[K1LLA_KING_KING]

I have have just tested an old 2TB disk through SMART and pre-clear. Both failed. I'm not really sure how to interpret the results. But does this mean I should not use this disk at all? And secondly is there anything I can do to fix a disk?

alien-smart-20230820-1106.zip preclear_disk_WD-WMAZA4904886_18807.txt

Edited August 20, 2023 by K1LLA_KING_KING

[[email protected]]

I have a strange problem that i cannot get fix.

 

Since the recent Unassigned devices updates, i can no longer attach to any windows shares in my estate.

 

Initially i had the issue with attaching to $shares but the updates have fixed that.

 

I have simplified to connections for testing, One is to a share in a Windows Server 2019 server, and the other to a share on a windows 2016 server. Neither will work when i try to connect with UD. They both fail with logon failure, and are also reported in Windows logs a logon failures.

 

If i do a manual connect from the unraid server copying the UD line (from syslog) and substitute the credentials file reference for username and password. The connection works (clearly i have to manually create the mount point first.

 

ie:   sudo /sbin/mount -t 'cifs' -o rw,noserverino,nounix,iocharset=utf8,file_mode=0777,dir_mode=0777,uid=99,gid=100,username='xxxxxxx',password='xxxxxxxx' '//192.168.5.10/unraidbu' '/mnt/remotes/192.168.5.10_unraidbu'

 

This mounts and i can access the files on this mount point.

 

What is even odder... but may point to the issue.... i created for testing an OMV vm, and created an smb share with exactly the same share name and password, and this i can connect to that fine with UD.

 

So problem only to real windows machines. I have tried setting the smb version setting on and off.... no difference.

 

This used to work, but seemed to stop at the time of the recent UD updates..

 

Lost as what to try next.

avon-unraid-diagnostics-20230820-0902.zip

[dlandon]

3 hours ago, [email protected] said:

This used to work, but seemed to stop at the time of the recent UD updates..

The log shows this:

Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: Attempting to mount \\192.168.5.10\unraidbu
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: Status code returned 0xc000006d STATUS_LOGON_FAILURE
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: VFS: \\192.168.5.10 Send error in SessSetup = -13
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: VFS: cifs_mount failed w/return code = -13

Which can be a credentials and/or permissions issue.

 

UD creates a credentials file in the tmpfs and refers to it in the command line.  This way credentials are not shown in the log.  There may a tmpfs file issue, but I doubt it.  In your diagnostics I see some things:

• It does not show any output from the 'df' command that would normally show mount usage.
• Your log is very short because it was rotated, but I don't see any previous logs.
• Your flash has some check files from a file system scan:

  -rw-------  1 root root     32768 Dec 31  1979 FSCK0000.REC
  -rw-------  1 root root     32768 Dec 31  1979 FSCK0001.REC
  -rw-------  1 root root     32768 Dec 31  1979 FSCK0002.REC

  I'm not sure what any of this means.

Here's what I suggest:

• It looks like you are using AD.  Make sure you have the domain correct in UD.  It's my understanding that the domain has to be specified with the credentials.
• Verify your AD policies on the server.
• Change the WIdows credentils to plain text with no special characters.  UD may not be handling special characters the same as before.
• On a command line type 'df' and check the tmpfs usage.
• Power down and check your flash drive.
• Reboot.

[dlandon]

11 hours ago, Revan335 said:

Have another Problems to?

I cant't say from which 6.12. Version this exists.

 

After a disk is mounted by UD, it has no interaction with the file system or disk except to monitor it's mounted status (size, used and free space).  Anything that happens with the file system is under the control of Linux.

 

I suggest you do a bfrfs scrub to check the file system, and try a different USB port.

[dlandon]

9 hours ago, K1LLA_KING_KING said:

But does this mean I should not use this disk at all?

The disk has too meny Current Pending Sectors:

Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: ATTRIBUTE              	INITIAL	NOW  	STATUS
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Reallocated_Sector_Ct  	0      	0    	-
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Power_On_Hours         	28076  	28089	Up 13
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Temperature_Celsius    	28     	31   	Up 3
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Reallocated_Event_Count	0      	0    	-
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Current_Pending_Sector 	7      	65518	Up 65511
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: Offline_Uncorrectable  	14     	14   	-
Aug 20 08:30:07 preclear_disk_WD-WMAZA4904886_18807: S.M.A.R.T.: UDMA_CRC_Error_Count   	21     	21   	-
\

I'd say it has failed.

 

9 hours ago, K1LLA_KING_KING said:

is there anything I can do to fix a disk?

Replace it.

[Revan335]

2 hours ago, dlandon said:

I suggest you do a bfrfs scrub to check the file system, and try a different USB port.

How can i scrub this UD?

[dlandon]

1 hour ago, Revan335 said:

How can i scrub this UD?

Click the check mark icon by the mount point when the disk is mounted.

[dlandon]

For anyone with name resolution issues with UD installed, I've found a situation where UD does not keep the name resolution up to date if a local server IP address changes.  For example, let's say you have a remote server 'Test' accessible on your LAN and it has an IP address of 192.168.1.100 and the IP is not static.  When the DHCP server assigns a new IP address to 'Test' of 192.168.1.150, UD does not update the 'Test' IP address to 192.168.1.150 and doing a ping on 'Test' will try to ping 192.168.1.100 and not 192.168.1.150.

 

This will be fixed in the next release.  The short term solution is to change your server in UD to 'Test.local' or "Test.local TLD' if your 'local TLD' is not set to 'local'.  This should resolve the name correctly.

[Revan335]

4 hours ago, dlandon said:

Click the check mark icon by the mount point when the disk is mounted.

Any Notification/Progress about the Status of the Scrub?

[dlandon]

On 8/18/2023 at 3:21 PM, StatMatt said:

I've been trying to search this forum/thread, but haven't found this specific issue. I have a few remote shares set up on a Win10 PC. Everything seemed to be working fine and I was able read/write to those shares until recently. I usually have that PC powered off while I'm away, and when I returned recently and booted up that computer, the remote shares won't reconnect.  The remote PC is on and I am able to log into it (local or RDP) and the shares should all be accessible - afaik nothing changed.  How can I best diagnose the issue?  I've tried adding a new share from within unRAID, and when I "search for servers" that PC name (E5-2670) comes up, but then when step thru the wizard and try to load shares the list comes up blank.  Hoping I can reconnect these shares and appreciate any tips.  Hopefully it's okay to post this question right in the thread here but I'd be happy to make a new thread if that's preferable. Thanks!

 

bonzo-diagnostics-20230818-1509.zip 197.18 kB · 0 downloads

I have found an issue in UD that probably caused this situation.  Basically, the problem is your remote server IP address.  If it is not static, when the DHCP server changes the IP address, UD was not detecting that and updating the name resolution.  Anything referencing the remote server (like UD and ping) would resolve the old IP address.

 

It will be fixed in the next release of UD.

[[email protected]]

On 8/20/2023 at 1:05 PM, dlandon said:

The log shows this:

Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: Attempting to mount \\192.168.5.10\unraidbu
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: Status code returned 0xc000006d STATUS_LOGON_FAILURE
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: VFS: \\192.168.5.10 Send error in SessSetup = -13
Aug 20 08:24:56 Avon-UNRAID kernel: CIFS: VFS: cifs_mount failed w/return code = -13

Which can be a credentials and/or permissions issue.

 

UD creates a credentials file in the tmpfs and refers to it in the command line.  This way credentials are not shown in the log.  There may a tmpfs file issue, but I doubt it.  In your diagnostics I see some things:

• It does not show any output from the 'df' command that would normally show mount usage.
• Your log is very short because it was rotated, but I don't see any previous logs.
• Your flash has some check files from a file system scan:

  -rw-------  1 root root     32768 Dec 31  1979 FSCK0000.REC
  -rw-------  1 root root     32768 Dec 31  1979 FSCK0001.REC
  -rw-------  1 root root     32768 Dec 31  1979 FSCK0002.REC

  I'm not sure what any of this means.

Here's what I suggest:

• It looks like you are using AD.  Make sure you have the domain correct in UD.  It's my understanding that the domain has to be specified with the credentials.
• Verify your AD policies on the server.
• Change the WIdows credentils to plain text with no special characters.  UD may not be handling special characters the same as before.
• On a command line type 'df' and check the tmpfs usage.
• Power down and check your flash drive.
• Reboot.

 

Seems there was a network issue that was breaking the connection to the 2019 server... some asymetric routing causing chaos..... odd it worked before though, and if done manually still did... 

 

The second connection to the 2016 server went away when i upgraded that server to 2019 (some what overdue!!).

 

So all up and running now. 

 

Thanks for your support.

[tazire]

Weird behaviour on an nvme drive. So I did have proxmox installed on this drive but I now want to clear the drive and use it as a cache drive on my test server. The drive appears twice on UD both at 8.59GB (512GB drive). When I try to clear or format the drive it just fails. Is there anything I can do to force clear/format the drive to get it back working as it should?

I have a feeling this might just be whatever filesystem proxmox has on the drive causing the issue. Probably nothing to do with UD as such but any help would be great.  

tower-diagnostics-20230822-2339.zip

[dlandon]

2 minutes ago, tazire said:

I have a feeling this might just be whatever filesystem proxmox has on the drive causing the issue. Probably nothing to do with UD as such but any help would be great.  

Yes, the file format on the disk is a file system without a partition.  That's why it is marked as 'Pool'.  Unraid and UD want all disks to have a partition.

 

It looks like you clicked on the red 'X' to clear the disk.  Try these two steps to see if it clears up:

• Click the two arrows icon in the upper right on the UD page.
• Reboot.

[tazire]

4 minutes ago, dlandon said:

Yes, the file format on the disk is a file system without a partition.  That's why it is marked as 'Pool'.  Unraid and UD want all disks to have a partition.

 

It looks like you clicked on the red 'X' to clear the disk.  Try these two steps to see if it clears up:

• Click the two arrows icon in the upper right on the UD page.
• Reboot.

Thanks for the response. Just as you responded I had rebooted the server and the drive is back to normal operation and I can add it as a cache pool now. cheers. 

[apefray]

Hi.

 

I have just recovered my unraid server from a major power cut (I do have battery backup, but this only lasts an hour, and the power was off for nearly 3 hours). Everything is up and running again, unraid is running fine, and the array has started, except the 4 unassigned disks I have. All 4 are now showing under Historical Devices instead of under unassigned devices, ready for me to mount.

 

 

 

How can I get them back under Unassigend Devices? Please ignore the disk ending H45A as this is a failed disk from the array that I need to remove.

 

Cheers

 

 

Edited August 23, 2023 by apefray
Added Image

[dlandon]

38 minutes ago, apefray said:

How can I get them back under Unassigend Devices?

How are they connected.  USB, ATA?  Looks like a controller failure.

[apefray]

😳

 

Ahhhh, you would be correct in assuming a controller failure of some sorts. I have an external 4 bay drive enclosure, which I thought was set to auto start when power returns, just like the unraid server. The server did indeed start, but the external enclosure didn't, hence me being slightly embarrassed.

 

So yes, you are correct in the sense that it was a controller failure, but only because the external bay hadn't powered on.....

 

Lesson learned, check before posting........

[BeefIsForDinner]

Evening. Ever since the ~16th, I haven't been able to mount my SMB shares on my Windows machines. I have sharing enabled and smb security configured, but I'm not getting anywhere. I tried on two separate unRAID servers and I couldn't access a share from either one.

 

Config and logs attached.

hyperion-diagnostics-20230823-2253.zip

[murkus]

remote NFS shares have "Invalid configuration - Remove and Re-Add", but I cannot remove the share from UA (the "x" button does not remove the share). What now, how to fix?

[dlandon]

7 hours ago, BeefIsForDinner said:

Evening. Ever since the ~16th, I haven't been able to mount my SMB shares on my Windows machines. I have sharing enabled and smb security configured, but I'm not getting anywhere. I tried on two separate unRAID servers and I couldn't access a share from either one.

It'll be fixed in the next release.

[dlandon]

1 hour ago, murkus said:

remote NFS shares have "Invalid configuration - Remove and Re-Add", but I cannot remove the share from UA (the "x" button does not remove the share). What now, how to fix?

Show a screen shot of the NFS shares.

[BeefIsForDinner]

1 hour ago, dlandon said:

It'll be fixed in the next release.

Good deal, appreciate the insight!

[murkus]

[dlandon]

2 hours ago, murkus said:

What I was looking for were the device designations.  I'm pretty sure there are invalid characters in the device designations.

 

Irrespective of that, PM me the /flash/config/plugins/unassigned.devices/samba_mount.cfg file and I can walk you through how to remove those remote shares manually.  Don't post that file on the forum.