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


Recommended Posts

On 5/6/2020 at 3:58 PM, dlandon said:

UD mounts the remote mount, but has no control over the operation of the NFS share.  Linux manages the remote share.  UD does not unmount the share if the server goes off-line.  With that said, please post diagnostics when this happens and I'll see if here is more that can be done to prevent hangs.

I was able to get the issue to occur again and got diagnostics with the NFS share stuck and diagnostics after I resolved it. Here is the process I went through.

 

1. Detected issue on Main Tab. says that the NFS mount has a size of 0.

2. Tried to force an unmount "umount -f /mnt/disks/192.168.1.10_Plex_Library"

   a. Get and error saying the device is busy "umount.nfs: /mnt/disks/192.168.1.10_Plex_Library: device is busy"

3. Tried to find what was keeping the mount busy.

   a. lsof | grep '/mnt/disks/192.168.1.10_Plex_Library'

   b. Get an error back "lsof: no pwd entry for UID 472"

4. Tried "fuser /mnt/disks/192.168.1.10_Plex_Library" and then "umount -f /mnt/disks/192.168.1.10_Plex_Library" which worked.

5. From the UD GUI I could remount the NFS share and all is back to normal.

 

 

Attachments. 1349 is while the share was stuck. 1357 is after I resolved it with the steps listed above.

deathstar-diagnostics-20200514-1357.zip deathstar-diagnostics-20200514-1349.zip

Link to comment

Hi!

 

I've mounted an NFS share from another NAS on my network using Unassigned Devices, and it seems to automatically then share this via SAMBA on my network. I'd prefer it not do this, as my NAS already shares the location itself.. 

 

Is there a way to disable this?

Link to comment
6 minutes ago, Cadr said:

I've mounted an NFS share from another NAS on my network using Unassigned Devices, and it seems to automatically then share this via SAMBA on my network. I'd prefer it not do this, as my NAS already shares the location itself.. 

 

Link to comment

Sorry everyone I somehow managed to merge this thread with a completely different thread by accident. I corrected the title (sort of) but the end result is any links to the thread are now broken, including the plugin support link.

 

EDIT

 

OK, I think I now know what happened at least. While I was trying to merge 2 threads (crossposted) from a new user, he apparently deleted one of the threads to be merged, with the result that a different thread (this one) got merged instead.

 

The user only had those 2 posts, and one other, which is in this thread 2 posts up.

Link to comment
  • trurl pinned this topic
7 hours ago, Cadr said:

Hi!

 

I've mounted an NFS share from another NAS on my network using Unassigned Devices, and it seems to automatically then share this via SAMBA on my network. I'd prefer it not do this, as my NAS already shares the location itself.. 

 

Is there a way to disable this?

No.  That's what UD does.  If you just want a local mount point, you should do it through a User Script.

Link to comment
4 hours ago, Stoodo said:

Is there a way to keep the unassigned drives mounted when the array is stopped?  It would be nice to keep the NFS share to ESXI up while making changes to the actual Array.  Thank you.

All UD devices are unmounted when the array is stopped.  Leaving devices mounted would prevent a clean array shutdown.

Link to comment
On 4/27/2020 at 5:42 AM, johnnie.black said:

One request if you don't mind, I would like to see the devices grouped more closely together, similar to the array devices, so they didn't take so much vertical real estate and more could fit on the page, e.g. 8 arrays devices vs 8 UDs:

 

image.thumb.png.8cfdd64d568e82ce827d899b86d7c28d.png

 

image.thumb.png.7fecf0e483562f20b2027687dc9e6c1d.png

 

Also not a very big fan of the new color, maybe because they are links now? But not a big deal.

 

 

Fixed in latest release.

  • Thanks 1
Link to comment
On 4/13/2020 at 11:42 AM, ogi said:

Figured this had been requested before, with 167 pages in the thread, and the search results for "temperature" was still at least 3 pages.... was hard to determine what the current status is.

 

Thanks for the suggested workaround!

 

Were you able to get this to work? I have 2 Samsung 970 NVME drives that consistently run in the 40-45C range as Unassigned Devices to hold my VM images, so I manually went in and set all of my spinning disks to have have 40C warning/ 45C critical and then changed my global disk setting (Settings->Disk Settings-> Default warning disk temp and Default critical disk temp) to 65C warning and 70C critical.

 

I still see warnings whenever I log into my server that one of the NVMEs was running in the 45-50C range, which is pretty common during VM disk images backups.  I've tried changing this up and down thinking maybe its not sticking, but it seems like the temps for my unassigned devices are stuck at the default temperatures from when unraid installed and aren't observing the values in this setting. Wasn't sure if I was doing something wrong here, or if there was an easy fix I was just missing?

Link to comment

In the latest release of UD, I've added a 'Share' switch to the remote mounts (SMB and NFS).  This allows you to mount a remote share and not have it shared again on Unraid.  I'm not sure why you'd want to do this other than having a local mount point for the remote share.  If it were me, I'd access the remote share through SMB or NFS directly and not through a local mount point.  CIFS mounts really struggle when there are networking issues.

  • Like 1
Link to comment
21 minutes ago, dlandon said:

If it were me, I'd access the remote share through SMB or NFS directly and not through a local mount point. 

Not all containers support SMB.  (ie: all my containers require mount points for external shares)

Link to comment

Hi, I have an issue which other seem to have reported too where unassigned devices section becomes unresponsive and smb shares to usb devices hang. Usually when copying a large amount of data to an smb share of a usb disk. Other shares on the main array are fine when this happens.

 

it means I currently can’t backup my PC as there is not enough space on the array to complete the backup so needs to go direct to the usb share.

 

When it hangs the fix is to unplug the usb and remount.  When it stops the wave symbol on the whole unassigned devices section just shows loading forever and the smb shares do not respond but via ssh the usbs are still mounted and accessible.

 

please help.

 

Link to comment
8 hours ago, Rik Bignell said:

Hi, I have an issue which other seem to have reported too where unassigned devices section becomes unresponsive and smb shares to usb devices hang. Usually when copying a large amount of data to an smb share of a usb disk. Other shares on the main array are fine when this happens.

 

it means I currently can’t backup my PC as there is not enough space on the array to complete the backup so needs to go direct to the usb share.

 

When it hangs the fix is to unplug the usb and remount.  When it stops the wave symbol on the whole unassigned devices section just shows loading forever and the smb shares do not respond but via ssh the usbs are still mounted and accessible.

 

please help.

 

Post diagnostics.

Link to comment
10 hours ago, dlandon said:

Post diagnostics.

These are the errors im getting in syslog.  If you let me know which other parts of the diagnostic zip you want to see i'll post them? Posting the whole zip would expose a large amount of secure info.

 

May 16 19:32:35 Tower unassigned.devices: Error: shell_exec(/usr/bin/dd bs=446 count=1 if=/dev/sdk 2>/dev/null | /bin/sum | /bin/awk '{print }') took longer than 5s! May 16 19:33:16 Tower kernel: mdcmd (39): set md_num_stripes 1280 May 16 19:33:16 Tower kernel: mdcmd (40): set md_queue_limit 80 May 16 19:33:16 Tower kernel: mdcmd (41): set md_sync_limit 5 May 16 19:33:16 Tower kernel: mdcmd (42): set md_write_method May 16 19:33:16 Tower kernel: mdcmd (43): set spinup_group 0 0 May 16 19:33:16 Tower kernel: mdcmd (44): set spinup_group 1 0 May 16 19:33:16 Tower kernel: mdcmd (45): set spinup_group 2 0 May 16 19:33:37 Tower unassigned.devices: Error: shell_exec(/usr/bin/dd bs=1 count=1 skip=449 if=/dev/sde 2>/dev/null | /bin/sum | /bin/awk '{print }') took longer than 2s! May 16 19:33:40 Tower unassigned.devices: Error: shell_exec(/usr/bin/dd bs=1 count=1 skip=450 if=/dev/sde 2>/dev/null | /bin/sum | /bin/awk '{print }') took longer than 2s! May 16 19:33:42 Tower unassigned.devices: Error: shell_exec(/usr/bin/dd bs=1 count=1 skip=451 if=/dev/sde 2>/dev/null | /bin/sum | /bin/awk '{print }') took longer than 2s! May 16 19:33:50 Tower unassigned.devices: Error: shell_exec(/usr/bin/dd bs=1 count=1 skip=460 if=/dev/sde 2>/dev/null | /bin/sum | /bin/awk '{print }') took longer than 2s! May 19 21:28:26 Tower ntfs-3g[12391]: Version 2017.3.23 integrated FUSE 27 May 20 19:30:04 Tower ntfs-3g[4890]: Version 2017.3.23 integrated FUSE 27 May 21 23:39:23 Tower nginx: 2020/05/21 23:39:23 [error] 6881#6881: *1760428 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.23, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "192.168.0.18", referrer: "http://192.168.0.18/Main" May 21 23:45:24 Tower kernel: traps: lsof[13685] general protection ip:148d174bbb8e sp:80f7be55bc258d0f error:0 in libc-2.30.so[148d1749c000+16b000]

Link to comment

Just tried the backup again and its frozen again.

 

Nothing in the syslog other than a timeout when i try and load the unassigned devices page

 

root@Tower:~# tail -f /var/log/syslog 


May 23 09:46:32 Tower nginx: 2020/05/23 09:46:32 [error] 6881#6881: *2034262 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.23, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "192.168.0.18", referrer: "http://192.168.0.18/Main"

 

And via ssh the /mnt/disks ls hangs

 

root@Tower:~# ls -la /mnt/disks/

---HANGS---

 

 

As soon as i unplug the usb drive it all returns to normal.

 

May 23 10:02:46 Tower kernel: usb 3-2.1: USB disconnect, device number 5

 

Link to comment
55 minutes ago, Rik Bignell said:

And via ssh the /mnt/disks ls hangs

 

root@Tower:~# ls -la /mnt/disks/

---HANGS---

This is why UD is slow responding and times out.  It is not an issue with UD.

 

It's very hard to help you with a log snippet without line breaks that makes it impossible to read.  The complete diagnostics provides a lot more information to help with your issue, such as the disk SMART report.  This is the best I can do to help based on the little information you have provided.

 

You probably have a disk or USB port issue.  Try the following:

  • Run file system check.
  • Check SMART report for disk issues.
  • Change USB ports.
Link to comment
2 minutes ago, dlandon said:

This is why UD is slow responding and times out.  It is not an issue with UD.

 

It's very hard to help you with a log snippet without line breaks that makes it impossible to read.  The complete diagnostics provides a lot more information to help with your issue, such as the disk SMART report.  This is the best I can do to help based on the little information you have provided.

 

You probably have a disk or USB port issue.  Try the following:

  • Run file system check.
  • Check SMART report for disk issues.
  • Change USB ports.

“Ifyou let me know which other parts of the diagnostic zip you want to see i'll post them? Posting the whole zip would expose a large amount of secure info.”

Link to comment
49 minutes ago, Rik Bignell said:

“Ifyou let me know which other parts of the diagnostic zip you want to see i'll post them? Posting the whole zip would expose a large amount of secure info.”

You can anonymize the diagnostics when you make it.

He asks for the whole zip for a reason.

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.