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

[dlandon]

11 hours ago, unr41dus3r said:

Config 1: Read Only on server configured (does not work)

NFS Server, i create the Share with only Read rights for a single client (unraid2) "clientIP(sec=sys,r)"

NFS Client, i can't mount this NFS share with my unraid2 Client, because i recieve the error that the share is unmountable.

I mounted a share on my main server using UD on my test server with the UD NFS Rule:

*(r,sec=sys,insecure,anongid=100,anonuid=99,no_root_squash)

 

and it mounted just fine:

Sep 13 19:32:48 BackupServer unassigned.devices: Mount NFS command: /sbin/mount -t 'nfs' -o ro,soft,noac,noatime,nodiratime,retrans=4,timeo=300 'MEDIASERVER:/mnt/user/Public' '/mnt/remotes/MEDIASERVER_Public'

 

It mounted RO on the local server becuse I set Read Only on.

 

If you are having trouble with this, it's probably something with your Rules.

[unr41dus3r]

18 hours ago, dlandon said:

I mounted a share on my main server using UD on my test server with the UD NFS Rule:

*(r,sec=sys,insecure,anongid=100,anonuid=99,no_root_squash)

 

and it mounted just fine:

Sep 13 19:32:48 BackupServer unassigned.devices: Mount NFS command: /sbin/mount -t 'nfs' -o ro,soft,noac,noatime,nodiratime,retrans=4,timeo=300 'MEDIASERVER:/mnt/user/Public' '/mnt/remotes/MEDIASERVER_Public'

 

It mounted RO on the local server becuse I set Read Only on.

 

If you are having trouble with this, it's probably something with your Rules.

 

Ok thanks for testing this.

I will try it again with your ruleset

[manofoz]

Hello. I hope this is the right place to ask for some help. I have all three unassigned devices plugins installed and here are the versions:

 

I am trying to copy data from an 4TB external HDD which is connected via USB to the motherboard. It appears like this in unRAID:

 

The guides I watch seem to have a mount button for the devices they want to access in the disks folder. I'm not sure why that is not the case for this, it is formatted NTFS.

 

Thanks! 

 

Update - I was able to get this drive to show up in a Windows 11 VM. This is not ideal but works as a workaround. The drive moved to historical devices when that happened. I will need to look up what that means:

 

Update 2 - The drive has mounted itself now that the VM is shut down. This is odd but desirable.

Edited September 15 by manofoz

[dlandon]

8 hours ago, manofoz said:

The guides I watch seem to have a mount button for the devices they want to access in the disks folder. I'm not sure why that is not the case for this, it is formatted NTFS.

There is a bug in UD that shows the 'Udev' indicator when there is a partition with no file system.  It will be fixed in the next release of UD.

 

8 hours ago, manofoz said:

Update 2 - The drive has mounted itself now that the VM is shut down. This is odd but desirable.

You probably have the "auto Mount' switch set so the disk will auto mount.

 

A work around for this situation is to set the "Auto Mount" switch on, detach the disk, then re-attach the disk.  The disk will be mounted.  Unmount it by clicking on the second partition 'Unmount' button.

[manofoz]

13 minutes ago, dlandon said:

A work around for this situation is to set the "Auto Mount" switch on, detach the disk, then re-attach the disk.  The disk will be mounted.  Unmount it by clicking on the second partition 'Unmount' button.

 

Thanks for the workaround! When you say detach do you mean unplug the USB and plug it back in or is there a way to detach from the UI? My other disks shouldn't have empty partitions as I formatted them myself before using them for Media but I have one more WD external to migrate. I guess I could always detach it by connecting it to my Window's VM and then re-attach so auto mount works by shutting the VM down (like what happened with this first drive). 

[dlandon]

7 minutes ago, manofoz said:

 

Thanks for the workaround! When you say detach do you mean unplug the USB and plug it back in or is there a way to detach from the UI? My other disks shouldn't have empty partitions as I formatted them myself before using them for Media but I have one more WD external to migrate. I guess I could always detach it by connecting it to my Window's VM and then re-attach so auto mount works by shutting the VM down (like what happened with this first drive). 

Click on the device settings for the disk (three gears icon) when the device is not mounted and you'll see a 'Detach' button.  Click on the same icon for that device in the historical devices and you'll see an 'Attach' button.

[Bayoubuddha]

 

I just installed a 1 TB m.2 NVME and after I format it, this is what I'm seeing. Any help would be appreciated. 

 

 

[dlandon]

2 hours ago, Bayoubuddha said:

 

I just installed a 1 TB m.2 NVME and after I format it, this is what I'm seeing. Any help would be appreciated. 

 

 

Post diagnostics.

[Bayoubuddha]

Sorry. Should have done that from the start. 

jupiter-diagnostics-20230916-2233.zip

[dlandon]

1 hour ago, Bayoubuddha said:

Sorry. Should have done that from the start. 

jupiter-diagnostics-20230916-2233.zip 167.11 kB · 0 downloads

Does the UD page show a file system under 'FS'?  The udev info in your diagnostics doesn't show a file system.

 

Click on the double arrows on the UD webpage and see if it removes the 'Udev' display.

 

How did you format the disk?

[Bayoubuddha]

It's formatted as XFS. I got it working by accident when I clicked on the toggle for Read Writes Display. The Mount button appeared and mounted when I clicked on it. 

[Johann]

On 9/9/2023 at 1:02 PM, dlandon said:

Yes.  The good thing is it's powered off too.

Oh awesome! I wish i found out about this easier, I accumulated a handful of power-on hours and only found out about this feature recently. I guess there are pros and cons to keeping a drive on vs off as backup.

[asopala]

Hey guys, I'm having an issue with remote SMB shares.  After having to move and getting new IP addresses for my server and my Windows PC, I can't seem to connect the SMB remote shares I had set up.  I mounted them originally using the PC name, and that hasn't changed, but they won't mount.  Trying to make new SMB shares doesn't seem to work either, as searching for servers only shows the Unraid server (itself).  Not sure what's going on here.  It's worth mentioning I can access my Unraid shares from my PC, but I can't seem to do the other way around.  Anyone have any insight?  I've attached the diagnostics.

 

Just to add, the drives in my Windows PC are set to sharing, and I had no problems with it before.

alexnas-diagnostics-20230919-0958.zip

Edited September 19 by asopala
Additional information

[dlandon]

Your PC is not responding to a ping and UD thinks it is offline:

Sep 19 09:57:08 AlexNAS unassigned.devices: Mounting Remote Share '//SOPALA-MAINPC/A'...
Sep 19 09:57:09 AlexNAS unassigned.devices: Remote Server 'SOPALA-MAINPC' is offline and remote share 'A' cannot be mounted.
Sep 19 09:57:09 AlexNAS unassigned.devices: Mounting Remote Share '//SOPALA-MAINPC/D'...
Sep 19 09:57:11 AlexNAS unassigned.devices: Remote Server 'SOPALA-MAINPC' is offline and remote share 'D' cannot be mounted.

 

Go to an Unraid command line and ping the PC by name and IP address and see if it responds.  Check your network sharing on the Winfows PC.

 

41 minutes ago, asopala said:

Just to add, the drives in my Windows PC are set to sharing, and I had no problems with it before.

Something changed in your network setup and that's why UD can't see the PC now.

[murkus]

I daresay there is again a bug with the smb config file handling (some may call it a feature). After rebooting the unraid server, UA found all SMB and all NFS share mounts that are in the same DNS domain to be invalid. The cause seems to be that the entries contain full domain names (FILER.SUB.NET), and if I create a new smb / nfs mount, it will just use the hostname (FILER), even if I specify the server by full domain name. I actually would prefer it if UA uses the actual name provided by the user and wouldn't insist on throwing away the rest of the FQDN.

 

Note that I wrote that this is only happening for the servers that are in the same DNS domain. What do I mean:

 

UNRAID.SUB.NET -> DNS domain is SUB.NET

FILER.SUB.NET -> is in same DNS domain, UA wants it to be just FILER

FILER2.OTHER.NET -> is NOT in the same DNS domain, UA accepts it as FILER2.OTHER.NET

 

has this to do with domain in /etc/resolv.conf somehow?

here for sake of the example:

domain sub.net

search sub.net other.net other2.net

 

Edited September 20 by murkus

[dlandon]

27 minutes ago, murkus said:

I actually would prefer it if UA uses the actual name provided by the user and wouldn't insist on throwing away the rest of the FQDN.

It will remove the "Local TLD" part of the FQDN because name resolution is faster without it.  You have "Local TLD" set to "SUB.NET".

 

31 minutes ago, murkus said:

UNRAID.SUB.NET -> DNS domain is SUB.NET

FILER.SUB.NET -> is in same DNS domain, UA wants it to be just FILER

This should work after removing the "SUB.NET".   "UNRAID" and "FILER" should resolve wihout the "SUB.NET".

[murkus]

Yes, it DOES work. But with the rationale applied by you, you'd use the "search" definition and remove all local TLDs that are mentioned there. But you may end up with ambiguous hostnames, if the same host names exist in different of these TLDs.

 

I personally don't think that it makes a substantial performance improvement. You can prove me wrong of course. I would still prefer the FQDN being used, if the user chooses to put that in.  At the least it should not refuse to use the FQDN if that is in the config file, as this is how it was in the past here for me and I had to edit the config file manually.... (faster than deleting and creatig the shares again).

 

the BUG part for me was that after a reboot UA refused to recognize the existing shares. It should at least have corrected those entries on its own and not just say nay. It was suboptimal UX.

Edited September 20 by murkus

[dlandon]

21 hours ago, murkus said:

Yes, it DOES work. But with the rationale applied by you, you'd use the "search" definition and remove all local TLDs that are mentioned there. But you may end up with ambiguous hostnames, if the same host names exist in different of these TLDs.

Not really.  "server" and "server.local" are one in the same.  "server.some_tld" is not the same as "server" or "server..local".

 

UD does not remove all TLDs.  UD only removes the local tld defined in the "Local TLD" on that server.  For example, let's say "Local TLD" is set to "MY.SERVER":

• "server" remains "server"
• "server.MY.SERVER" is changed to "server".
• "server.OTHER.TLD" remains "server.OTHER.TLD".

 

21 hours ago, murkus said:

I personally don't think that it makes a substantial performance improvement. You can prove me wrong of course. I would still prefer the FQDN being used, if the user chooses to put that in.  At the least it should not refuse to use the FQDN if that is in the config file, as this is how it was in the past here for me and I had to edit the config file manually.... (faster than deleting and creatig the shares again).

The speed improvement is behind the scenes when the background task runs to refresh the server online status by pinging each server.

 

Removing the "Local TLD" when adding remote shares was done for several reasons:

• A "server" and "server.local" are the same and "server" is better for name resolution and there are no duplicates.
• When a "server" and "server.local" are both specified in Remote Shares, UD thinks the servers are different and the logic to protect from adding the same remote share as "server" and "server.local" doesn't work.
• UD has to keep up with IP address changes when the reomote server is using DHCP.  The logic to check for this is cleaner when the Local TLD is not speficied.
• If you decide to change the "Local TLD", you don't have to make any changes in the Remote Shares.  If you specified the "Local TLD" in the remote shares, you'd have to remove and re-add them.

21 hours ago, murkus said:

the BUG part for me was that after a reboot UA refused to recognize the existing shares. It should at least have corrected those entries on its own and not just say nay. It was suboptimal UX.

UD does not make configuration file changes.  Removing and then re-adding a remote share is not that difficult.

[asopala]

On 9/19/2023 at 10:51 AM, dlandon said:

Your PC is not responding to a ping and UD thinks it is offline:

 

Go to an Unraid command line and ping the PC by name and IP address and see if it responds.  Check your network sharing on the Winfows PC.

 

Something changed in your network setup and that's why UD can't see the PC now.

 

That did it.  Ethernet thought it was a public network and not a private one.  Thanks!

[shpitz461]

Hi, I'm having a weird issue with the drive listing:

 

 

Some drives show the 0UDEV thingy and cannot be mounted. They are all formatted ntfs (formatted in a Win11 VM). In UD they appear to show xfs as filesystem, which they were prior to doing the Windows ntfs format.

Some drives are listed twice: Venom is shown as Dev 10 as well as nvme3n1.

Any idea what is going on? Any logs I can pull?

A reboot clears up the duplicates, but the 0UDEV thingy still remains.

Thanks!

[dlandon]

14 minutes ago, shpitz461 said:

Any idea what is going on? Any logs I can pull?

Post diagnostics.  

[shpitz461]

32 minutes ago, dlandon said:

Post diagnostics.  

Dang you're quick!

Attached.

selanas-diagnostics-20230921-2012.zip

[dlandon]

10 hours ago, shpitz461 said:

Dang you're quick!

Attached.

selanas-diagnostics-20230921-2012.zip 368.08 kB · 1 download

How do you have so many nvme devices in one box?

 

Udev is reporting the disks marked "Udev" as having "XFS" file systems, but you said they are formatted NTFS.  Try clicking on the double arrows on the upper right side of the UD webpage and see if Udev will report the file system type correctly on those disks.

[shpitz461]

Ha, it was that easy? Didn't even notice the icon...

It did clear up the duplicates, shows the correct filesystem, and now I can mount them

I'm using two quad NVMe PCIe controllers, one Gen3, one Gen4, and also using the two NVMe ports on the motherboard, so 10 in total.

I tried using an octa NVMe controller (Highpoint Rocket 1108) but since it uses a PLX (the other two cards do not, the need motherboard bifurcation support), I couldn't get any of the drives to passthrough to a VM since they were all in the same iommu group. ACS did not help.

Edited Friday at 11:39 AM by shpitz461

[dlandon]

7 minutes ago, shpitz461 said:

It did clear up the duplicates, shows the correct filesystem, and now I can mount them

I've seen this a few times on my servers, it was a bug in the detach function in UD that has been fixed in the latest release of UD.  That's where the "Udev" marker on the mount button came from.  I added that to help indicate the problem and prevent any UD operations until it was sorted out.

 

10 minutes ago, shpitz461 said:

I couldn't get any of the drives to passthrough to a VM since they were all in the same iommu group. ACS did not help.

Are you passing through the UD disks to VMs?  If so, don't mount them in UD and in the VM at the same time.  You'll get file system corruption.