7 minutes ago, Espressomatic said:

I'll take a look.  Having multiple machines, it wasn't just shares potentially going offline, but also not being available until well after the system (mounting them) has started.

Delay the mounting of remote shares using the UD Setting.

1 hour ago, dlandon said:

Delay the mounting of remote shares using the UD Setting.

 

The problem isn't the delay, it's that mounting only happens once. I might have a system come up seconds or minutes behind another one, or it may be hours. And since the shares are crossed, not one-way, it can affect any of the systems - each of the four Unraid systems mounts one or more shares from each of the other three. Specifically, three systems each share out one volume and the other system (general-purpose NAS) shares out 5.

 

Not Unraid-specific, but a similar situation exists on on my Mac, which mounts shares from all of the servers. I have a program (Automounter) running at all times that mounts any defined shares when their server(s) come online, as MacOS also doesn't have built-in automounting except at startup.

Edited September 19, 20241 yr by Espressomatic

7 hours ago, dlandon said:

You shouldn't have to change permissions.  Explain more of what you are trying to do.

Sorry should have been clearer - I am trying to get my arr apps to write to the NAS drives - specifically Radarr and Sonarr.

media2 is now working since I changed the permissions - everything I download is moved to the correct folder.

media3 has some folders that it works and others that it doesn't.

 

Here are the permissions I have setup on media3 (Synology) - 192.168.1.251 is my OMV server, .48 is my Unraid

On 9/19/2024 at 10:47 AM, Espressomatic said:

 

The problem isn't the delay, it's that mounting only happens once. I might have a system come up seconds or minutes behind another one, or it may be hours. And since the shares are crossed, not one-way, it can affect any of the systems - each of the four Unraid systems mounts one or more shares from each of the other three. Specifically, three systems each share out one volume and the other system (general-purpose NAS) shares out 5.

 

Not Unraid-specific, but a similar situation exists on on my Mac, which mounts shares from all of the servers. I have a program (Automounter) running at all times that mounts any defined shares when their server(s) come online, as MacOS also doesn't have built-in automounting except at startup.

While not a total solution to your situation, I am implementing a retry on mounting remote shares on server startup.  There will be multiple tries (right now I'm thinking 3 times) with the UD "Remote Share Mount Wait Time" timeout setting between tries.  This will help some on server startup when the remote server is delayed coming online.

 

I prefer to count on the NFS and CIFS built-in robustness to keep the remote shares active and recover from network and remote server issues.

Anyone able to tell me why a mounted device would keep mounting into read only while read only is unchecked:

 

Sep 28 10:46:45 HomeLab unassigned.devices: Mount cmd: /sbin/mount -t 'ntfs' -o rw,relatime,nodev,nosuid,nls=utf8,umask=000 '/dev/sdc2' '/mnt/disks/anime'
Sep 28 10:46:47 HomeLab ntfs-3g[31567]: Mounted /dev/sdc2 (Read-Only, label "Plex", NTFS 3.1)

 

I ended up finding the error out side of the mounting logs.

Sep 28 11:02:00 HomeLab unassigned.devices: Mount warning: The disk contains an unclean file system (0, 0). Metadata kept in Windows cache, refused to mount. Falling back to read-only mount because the NTFS partition is in an unsafe state. Please resume and shutdown Windows fully (no hibernation or fast restarting.) Could not mount read-write, trying read-only

 

If anyone else ends up having this issue you can simply put it into a windows machine and safely eject it and it will clear up the issue.

 

Edited September 29, 20241 yr by Amino
found error

First time creating a remote SMB share using UD.  I created a read-only mount from my Unraid test-->main server and it mounts fine.  When browsing the remotes folder either from SSH or Unraid GUI, some folders are seen as "empty" even though they contain data.  UD sees the folder, but no files inside.  The permissions are identical to the files/folders that display correctly.  I don't have the cache dirs plugin, but do have Dynamix File Manager on the test server.  Any ideas?

 

 

21 hours ago, dboonthego said:

When browsing the remotes folder either from SSH or Unraid GUI, some folders are seen as "empty" even though they contain data.  UD sees the folder, but no files inside. 

 

Something is definitely strange.  The folders contain a single file.  I can create a dummy file ex. "dummy.txt" inside each folder and both files appear.  If I subsequently delete the dummy file, then it goes back to empty. 

1 hour ago, dboonthego said:

Something is definitely strange.  The folders contain a single file.  I can create a dummy file ex. "dummy.txt" inside each folder and both files appear.  If I subsequently delete the dummy file, then it goes back to empty. 

Please provide diagnostics for further help.

4 hours ago, dlandon said:

Please provide diagnostics for further help.

Upgrading Unraid OS from 6.12.9 to 6.12.13 resolved it.  I found this thread indicating a bug fix in 6.12.10.  Thanks for the reply.

 

I'm having an issue with SMB shares. Every time I boot, my SMB share is not mounted and the button to mount is grayed out, as shown in this picture:

 

If I delete the share and add it back, it works without any issues. But as soon as I reboot, I have to do it all over again. Is this a known issue? Or is an issue on my side?

1 hour ago, s1l3nce said:

I'm having an issue with SMB shares. Every time I boot, my SMB share is not mounted and the button to mount is grayed out, as shown in this picture:

 

If I delete the share and add it back, it works without any issues. But as soon as I reboot, I have to do it all over again. Is this a known issue? Or is an issue on my side?

What is happening is the remote server is not seen by UD to be ready to accept SMB connections.  UD checks that the port for SMB is open on that server.  When you remove and add back the SMB share, the remote server is checked by UD and found to be on line and the 'Mount' button is enabled.  Normally UD checks remote servers every 15 seconds for being online.

 

You can click on the double arrows icon in the the upper right of the UD page to force an update instead of removing and adding back the remote share.

 

The next release of UD will attempt to mount remote shares three times with the UD "Remote Share Wait Time" between tries.  The remote online status will be checked before each try to mount remote shares to be sure it is current.

4 hours ago, dlandon said:

You can click on the double arrows icon in the the upper right of the UD page to force an update instead of removing and adding back the remote share.

I've tried that but it did not do anything. Only removing it and adding it again works.

Is there a log somewhere that might be showing some problems that I can share? I do not recall having this issue a while back (I'm still not using Unraid on a daily basis; I'm just experimenting with it to get comfortable when I move from my old Synology NAS).

--------
Edit:

I've just rebooted the UNRAID server and I found this in the system log:

Oct 1 21:23:09 UNRAID unassigned.devices: Mounting Remote Share '//NameOfMyNAS/data'...

Oct 1 21:23:10 UNRAID unassigned.devices: Remote Server 'NameOfMyNAS.LOCAL' is offline and remote share 'data' cannot be mounted.

So apparently it's not able to find the share but it does find it when I re-add it. I don't get it.

Btw, I have downloaded the ud_diagnostics zip and the only relevant thing that I could find are those lines from the syslog.

Edited October 1, 20241 yr by s1l3nce
I found important entries in the log file

41 minutes ago, s1l3nce said:

I've tried that but it did not do anything. Only removing it and adding it again works.

Is there a log somewhere that might be showing some problems that I can share? I do not recall having this issue a while back (I'm still not using Unraid on a daily basis; I'm just experimenting with it to get comfortable when I move from my old Synology NAS).

--------
Edit:

I've just rebooted the UNRAID server and I found this in the system log:

Oct 1 21:23:09 UNRAID unassigned.devices: Mounting Remote Share '//NameOfMyNAS/data'...

Oct 1 21:23:10 UNRAID unassigned.devices: Remote Server 'NameOfMyNAS.LOCAL' is offline and remote share 'data' cannot be mounted.

So apparently it's not able to find the share but it does find it when I re-add it. I don't get it.

Btw, I have downloaded the ud_diagnostics zip and the only relevant thing that I could find are those lines from the syslog.

Post the ud_diagnostics.zip file.

Just now, dlandon said:

Post the ud_diagnostics.zip file.

Here it is.

ud_diagnostics-20241001-215048.zip

BTW, I don't know if this is relevant or not but I'm using Unraid inside Proxmox.

57 minutes ago, s1l3nce said:

BTW, I don't know if this is relevant or not but I'm using Unraid inside Proxmox.

I've had this reported several times and I'm beginning to develop a theory of how this happens.  When a remote share is added, there is a procedure to determine if the remote server is online that appears to be different than the process to determine if a remote share is on line after it has been added.  It's probably related to name resolution.

 

It may be a bit of a challenge for me to fix because I've not been able to reproduce the issue.  I'll see if I can figure it out.

 

BTW, you don't need to remove the remote share, just add it and it will overwrite the existing one and should get the same result as removing and adding.

2 hours ago, s1l3nce said:

Here it is.

ud_diagnostics-20241001-215048.zip 26.12 kB · 1 download

Give me the output of this command when the mount button is grayed out:

bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'; echo $?

 

and this:

bash -c '(echo >/dev/tcp/ZENO/445) &>/dev/null'; echo $?

 

and this:

bash -c '(echo >/dev/tcp/{IP Address of ZENO}/445) &>/dev/null'; echo $?

 

and this:

arp -a ZENO

 

and this:

arp -a ZENO.LOCAL

 

40 minutes ago, dlandon said:

bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'; echo $?

This one returns a 0 after 4-5 seconds but I have observed that a few seconds later, the share gets mounted 👌

 

40 minutes ago, dlandon said:

bash -c '(echo >/dev/tcp/ZENO/445) &>/dev/null'; echo $?

 

This one returns a 1 instantly

 

40 minutes ago, dlandon said:

bash -c '(echo >/dev/tcp/192.168.82.20/445) &>/dev/null'; echo $?

 

This one returns a 0 instantly.
 

40 minutes ago, dlandon said:

arp -a ZENO

 

ZENO: Unknown server error

 

40 minutes ago, dlandon said:

arp -a ZENO.LOCAL

 

ZENO.local (192.168.82.20) at {ZENO's Mac address} [ether] on br0


The fact that the first command mounts the share is really interesting.

Edited October 1, 20241 yr by s1l3nce
formatting

1 hour ago, s1l3nce said:

The fact that the first command mounts the share is really interesting.

I see it.  I have that command time out after 1 second.  It's too soon.

Correction: the command 

bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'

is not mounting the share but simply making it available; it goes green and the mount button is no longer grayed out.

I've tried adding that command to a custom script that runs at Startup of array but it doesn't work, unless I run it manually, which makes me think that maybe if I could delay the script for a few seconds, that might fix the first issue, which is making the shares available. The second issue would be to make the shares automount.

Edited October 2, 20241 yr by s1l3nce

5 minutes ago, s1l3nce said:

Correction: the command 

bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'

is not mounting the share but simply making it available; it goes green and the mount button is no longer grayed out.

I've tried adding that command to a custom script that runs at Startup of array but it doesn't work, unless I run it manually, which makes me think that maybe if I could delay the script for a few seconds, that might fix the first issue, which is making the shares available. The second issue would be to make the shares automount.

I'll have a fix in the next release of UD.  If you set the automount switch on, the remote share should auto mount once the online detection is fixed.

It wasn't automounting but I added this delay to the script and now it's working perfectly.
 

#!/bin/bash
sleep 5  # Delay execution for 5 seconds
bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'

 

For what it's worth, I have about 20 NFS shares configured, but not automounting.  They are there in case I want to look at the files on the remote server, but every single mount button is greyed out.  That led me here, and I see I'm not alone.

14 hours ago, s1l3nce said:

It wasn't automounting but I added this delay to the script and now it's working perfectly.
 

#!/bin/bash
sleep 5  # Delay execution for 5 seconds
bash -c '(echo >/dev/tcp/ZENO.LOCAL/445) &>/dev/null'

 

 

1 hour ago, House Of Cards said:

For what it's worth, I have about 20 NFS shares configured, but not automounting.  They are there in case I want to look at the files on the remote server, but every single mount button is greyed out.  That led me here, and I see I'm not alone.

Update UD and see if your issues are solved.

3 hours ago, dlandon said:

Update UD and see if your issues are solved.

Excellent @dlandon! It works

But after updating, when I rebooted the system, I got this command in console right before it posts the Unraid Version:
 

modprobe: FATAL: Module ntfs3 not found in directory /lib/modules/6.1.74-Unraid

I'm on version 6.12.8, in case that helps.

Edited October 2, 20241 yr by s1l3nce