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


dlandon

Recommended Posts

On 4/7/2021 at 4:32 PM, Andiroo2 said:

I'm also hitting the "Cancelling wait for mid" issue on my Unassigned Disk remote SMB share connected to a Mac:  

 


Apr  7 14:35:03 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close unmatched open for MID:ad201
Apr  7 14:35:03 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close cancelled mid failed rc:-9
Apr  7 14:37:35 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close interrupted close
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1443535 cmd: 6
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close interrupted close
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close cancelled mid failed rc:-9
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1682517 cmd: 5
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1682518 cmd: 14
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close unmatched open for MID:19ac55

This goes on and on in my log.  Diagnostics attached...thanks!

tower-diagnostics-20210407-1631.zip 147.57 kB · 0 downloads

Are you using Jumbo frames?

Link to comment
On 4/7/2021 at 4:32 PM, Andiroo2 said:

I'm also hitting the "Cancelling wait for mid" issue on my Unassigned Disk remote SMB share connected to a Mac:  

 


Apr  7 14:35:03 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close unmatched open for MID:ad201
Apr  7 14:35:03 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close cancelled mid failed rc:-9
Apr  7 14:37:35 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close interrupted close
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1443535 cmd: 6
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close interrupted close
Apr  7 14:42:37 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close cancelled mid failed rc:-9
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1682517 cmd: 5
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918 Cancelling wait for mid 1682518 cmd: 14
Apr  7 14:45:08 Tower kernel: CIFS: VFS: \\MACMINI-581918\DATA Close unmatched open for MID:19ac55

This goes on and on in my log.  Diagnostics attached...thanks!

tower-diagnostics-20210407-1631.zip 147.57 kB · 1 download

Are you using Jumbo frames?

Link to comment

If you are using SMB or NFS remote mounts and have issues with used and free showing zero, or excessive strange log messages related to CIFS mounts, be sure you are not using Jumbo frames on your network.  It's difficult to set up Jumbo frames on a network because every device on the network must be set up properly.  This includes network switches.

 

It s recommended that you do not have Jumbo frames enabled anywhere on your network if you are using CIFS mounts (SMB or NFS).

Link to comment

I'm sorry if this is a duplicate post.  But I'm still having the issue. My NFS mounts frequently seem to unmount themselves. I'm not sure where to start to look for errors.  The LOG appears blank.  I'm not sure how to turn it on. One of the mounts seems to go out more frequently than the others; however, this mount sees the most activity.  I've notices that a common trigger is when I attempt to write to the drive. Though I'm not sure that this is anything more than observation bias.

 

Any ideas where to start?

 

I should note that these mounts were working fine previously, and seemed to start this activity after the December update (which I didn't initiate until March). Also, the mounts point to a completely separate Unraid installation (separate hardware).  I had wondered if it had something to do with 6.9.1 also.

Capture.PNG.13eaecc5035cb85f19ef59e80aa10d00.png

Link to comment
56 minutes ago, Andiroo2 said:

It’s a Mac mini. I’ve got a SMB connection from the Unraid server to the Mac via the UD plugin. 

I think this issue is related to some remote servers needing to have the version specified when mounting a CIFS mount.  I have updated UD and added a setting to possibly help with this issue.

 

Procedure:

  • Update UD plugin.
  • Unmount all SMB remote shares.
  • Go to settings and set 'Specify SMB Version when Mounting Remote Shares?' to 'Yes'.
  • Mount remote SMB shares.

Let me know if this takes care of the excessive log messages.

Link to comment
4 hours ago, mikesp18 said:

I'm sorry if this is a duplicate post.  But I'm still having the issue. My NFS mounts frequently seem to unmount themselves. I'm not sure where to start to look for errors.  The LOG appears blank.  I'm not sure how to turn it on. One of the mounts seems to go out more frequently than the others; however, this mount sees the most activity.  I've notices that a common trigger is when I attempt to write to the drive. Though I'm not sure that this is anything more than observation bias.

 

Any ideas where to start?

 

I should note that these mounts were working fine previously, and seemed to start this activity after the December update (which I didn't initiate until March). Also, the mounts point to a completely separate Unraid installation (separate hardware).  I had wondered if it had something to do with 6.9.1 also.

Capture.PNG.13eaecc5035cb85f19ef59e80aa10d00.png

Try this:

 

Link to comment

Hello,

I am having an issue in the official Plex docker on Unraid 6.9.2.  I have used unassigned devices to mount an NVME SSD which the purpose of is to be an exclusive SSD for Plex metadata.  When I route the config to the mounted SSD, I get some files that won't play.  The error shown is "Conversion Failed.  A required codec could not be found or failed to install."  I have narrowed it down to files that have audio encoded in EAC3 5.1.  If the mapped route is /mnt/disks/ everything will start up fine, but just have the EAC3 issue.  If I map to /mnt/remotes/ then I get an unhealthy start up that won't allow the Plex GUI to be accessible.  If I map the metadata to the cache via /mnt/user/appdata/, the problem goes away, but obviously the metadata is on the cache instead of the NVME.  I am assuming that there is some sort of permission problem with the unassigned devices plug in.  I have tried to put the NVME in pass through with no luck either.  I have tried using the legacy mount points as well, but get an unhealthy start with an inaccessible gui. 

 

trgxpt5.pngsEunddC.png

 

Edited by RelativeAsh
Link to comment
55 minutes ago, RelativeAsh said:

Hello,

I am having an issue in the official Plex docker on Unraid 6.9.2.  I have used unassigned devices to mount an NVME SSD which the purpose of is to be an exclusive SSD for Plex metadata.  When I route the config to the mounted SSD, I get some files that won't play.  The error shown is "Conversion Failed.  A required codec could not be found or failed to install."  I have narrowed it down to files that have audio encoded in EAC3 5.1.  If the mapped route is /mnt/disks/ everything will start up fine, but just have the EAC3 issue.  If I map to /mnt/remotes/ then I get an unhealthy start up that won't allow the Plex GUI to be accessible.  If I map the metadata to the cache via /mnt/user/appdata/, the problem goes away, but obviously the metadata is on the cache instead of the NVME.  I am assuming that there is some sort of permission problem with the unassigned devices plug in.  I have tried to put the NVME in pass through with no luck either.  I have tried using the legacy mount points as well, but get an unhealthy start with an inaccessible gui. 

 

trgxpt5.pngsEunddC.png

 

There are several things you are getting wrong.  ALL disks mount at /mnt/disks.  Remote (SMB and NFS) mounts will mount at /mnt/remotes, or /mnt/disks if the legacy option is enabled.  Physical disks always mount at .mnt.disks.

 

UD only mounts disks.  It does not have any control over the permissions or r/w access.

 

Move your nvme disk to the Unraid pool and use it from there.  The support is better for the things you are trying to do.

 

Based on the error you are seeing, it appears to be a Plex issue transcoding certain file types.

Link to comment

Thank you.

 

Since I am on 6.9.2, I did just add the lone drive as an additional cache pool as you suggested.  Changing that and mapping the meta data to that drive allows Plex to transcode EAC3 correctly.

 

I think there is more to this issue though, due to the fact that I can turn the transcode error on and off simply by where I map the meta data to.

Link to comment
6 minutes ago, trurl said:

When mapping to an Unassigned Device, whether local disk or remote share, you must set the Access Mode to slave for the mapping.

Agreed.  Fix Common Problems made sure that was done.  I still had the issue present when set as a slave.

Link to comment

I use UD to mount shares from my Synology, which my unRAID setup will be replacing once my new hardware arrives.  I started having an issue in UD where the Mount button was grayed out and not clickable.  Not understanding what was going on I removed the share from UD and attempted to add it back and saw the following.  I can search for servers and the Synology is listed, I select it and enter the same username/password I've been using, domain is left blank.  Upon searching for shares, it's empty. If I manually type the share name in I'm left with a share where the mount button is grayed out and not clickable, as before.  I've got 2 Synology, one primary that backs up to the second.  Both are accessible from a Windows machine using the same user/pass as I am attempting to use in UD.  The 2nd Synology, the backup, has the same shares setup on it and reacts the same way in UD, I cannot get a list of the shares on it.  However, and I know this is going to point me back to my Synology, I can put in a Windows box under UD and it list the shares without issue.  I don't understand why it would have stopped working.  Any help would be appreciated.

Link to comment
14 minutes ago, StirCwazy said:

I use UD to mount shares from my Synology, which my unRAID setup will be replacing once my new hardware arrives.  I started having an issue in UD where the Mount button was grayed out and not clickable.  Not understanding what was going on I removed the share from UD and attempted to add it back and saw the following.  I can search for servers and the Synology is listed, I select it and enter the same username/password I've been using, domain is left blank.  Upon searching for shares, it's empty. If I manually type the share name in I'm left with a share where the mount button is grayed out and not clickable, as before.  I've got 2 Synology, one primary that backs up to the second.  Both are accessible from a Windows machine using the same user/pass as I am attempting to use in UD.  The 2nd Synology, the backup, has the same shares setup on it and reacts the same way in UD, I cannot get a list of the shares on it.  However, and I know this is going to point me back to my Synology, I can put in a Windows box under UD and it list the shares without issue.  I don't understand why it would have stopped working.  Any help would be appreciated.

Post your diagnostics.

Link to comment
10 hours ago, StirCwazy said:

Thanks for the quick reply.  As requested, my diags.

sedecore-diagnostics-20210411-0854.zip 88.03 kB · 4 downloads

 

I'm not sure where to start to help you.  You seem to have a lot if issues with disks:

Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: DHCP, IP range 192.168.122.2 -- 192.168.122.254, lease time 1h
Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: DHCP, sockets bound exclusively to interface virbr0
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: reading /etc/resolv.conf
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 71.10.216.1#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 71.10.216.2#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 192.168.0.1#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: read /etc/hosts - 2 addresses
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: read /var/lib/libvirt/dnsmasq/default.addnhosts - 0 addresses
Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: read /var/lib/libvirt/dnsmasq/default.hostsfile
Apr 11 07:34:18 SEDECORE kernel: virbr0: port 1(virbr0-nic) entered disabled state
Apr 11 07:34:18 SEDECORE kernel: L1TF CPU bug present and SMT on, data leak possible. See CVE-2018-3646 and https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/l1tf.html for details.
Apr 11 07:39:22 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 1 TIMES] ###
Apr 11 07:44:00 SEDECORE root: Fix Common Problems Version 2021.04.10
Apr 11 07:55:06 SEDECORE kernel: isci 0000:05:00.0: sci_remote_device_frame_handler: in wrong state: STP_DEV_IDLE
Apr 11 07:55:06 SEDECORE kernel: sas: Enter sas_scsi_recover_host busy: 1 failed: 1
Apr 11 07:55:06 SEDECORE kernel: sas: ata9: end_device-7:2: cmd error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata7: end_device-7:0: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata8: end_device-7:1: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata9: end_device-7:2: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata10: end_device-7:3: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 1 tries: 1
Apr 11 08:00:32 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 9 TIMES] ###
Apr 11 08:00:47 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 1 TIMES] ###
Apr 11 08:04:02 SEDECORE unassigned.devices: Removing configuration '//DS1/Downloads'.
Apr 11 08:04:12 SEDECORE kernel: net_ratelimit: 10 callbacks suppressed
Apr 11 08:04:12 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 2 TIMES] ###
Apr 11 08:04:25 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 2 TIMES] ###
Apr 11 08:14:28 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 9 TIMES] ###
Apr 11 08:14:33 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
Apr 11 08:54:34 SEDECORE kernel: isci 0000:05:00.0: sci_remote_device_frame_handler: in wrong state: STP_DEV_IDLE
Apr 11 08:54:34 SEDECORE kernel: sas: Enter sas_scsi_recover_host busy: 1 failed: 1
Apr 11 08:54:34 SEDECORE kernel: sas: ata9: end_device-7:2: cmd error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata7: end_device-7:0: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata8: end_device-7:1: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata9: end_device-7:2: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata10: end_device-7:3: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 1 tries: 1

 

I am not a network expert, but something doesn't seem right with your network.

 

UD won't turn on the 'Mount' button if the remote server won't respond to a ping.  Go to a command line and manually ping the server in question.  If you don't get a response, you'll have to figure out why not.

Link to comment
1 hour ago, dlandon said:

 

I'm not sure where to start to help you.  You seem to have a lot if issues with disks:


Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: DHCP, IP range 192.168.122.2 -- 192.168.122.254, lease time 1h
Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: DHCP, sockets bound exclusively to interface virbr0
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: reading /etc/resolv.conf
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 71.10.216.1#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 71.10.216.2#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: using nameserver 192.168.0.1#53
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: read /etc/hosts - 2 addresses
Apr 11 07:34:18 SEDECORE dnsmasq[8990]: read /var/lib/libvirt/dnsmasq/default.addnhosts - 0 addresses
Apr 11 07:34:18 SEDECORE dnsmasq-dhcp[8990]: read /var/lib/libvirt/dnsmasq/default.hostsfile
Apr 11 07:34:18 SEDECORE kernel: virbr0: port 1(virbr0-nic) entered disabled state
Apr 11 07:34:18 SEDECORE kernel: L1TF CPU bug present and SMT on, data leak possible. See CVE-2018-3646 and https://www.kernel.org/doc/html/latest/admin-guide/hw-vuln/l1tf.html for details.
Apr 11 07:39:22 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 1 TIMES] ###
Apr 11 07:44:00 SEDECORE root: Fix Common Problems Version 2021.04.10
Apr 11 07:55:06 SEDECORE kernel: isci 0000:05:00.0: sci_remote_device_frame_handler: in wrong state: STP_DEV_IDLE
Apr 11 07:55:06 SEDECORE kernel: sas: Enter sas_scsi_recover_host busy: 1 failed: 1
Apr 11 07:55:06 SEDECORE kernel: sas: ata9: end_device-7:2: cmd error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata7: end_device-7:0: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata8: end_device-7:1: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata9: end_device-7:2: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: ata10: end_device-7:3: dev error handler
Apr 11 07:55:06 SEDECORE kernel: sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 1 tries: 1
Apr 11 08:00:32 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 9 TIMES] ###
Apr 11 08:00:47 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 1 TIMES] ###
Apr 11 08:04:02 SEDECORE unassigned.devices: Removing configuration '//DS1/Downloads'.
Apr 11 08:04:12 SEDECORE kernel: net_ratelimit: 10 callbacks suppressed
Apr 11 08:04:12 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 2 TIMES] ###
Apr 11 08:04:25 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
### [PREVIOUS LINE REPEATED 2 TIMES] ###
Apr 11 08:14:28 SEDECORE kernel: neighbour: arp_cache: neighbor table overflow!
### [PREVIOUS LINE REPEATED 9 TIMES] ###
Apr 11 08:14:33 SEDECORE dhcpcd[2224]: route socket overflowed - learning interface state
Apr 11 08:54:34 SEDECORE kernel: isci 0000:05:00.0: sci_remote_device_frame_handler: in wrong state: STP_DEV_IDLE
Apr 11 08:54:34 SEDECORE kernel: sas: Enter sas_scsi_recover_host busy: 1 failed: 1
Apr 11 08:54:34 SEDECORE kernel: sas: ata9: end_device-7:2: cmd error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata7: end_device-7:0: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata8: end_device-7:1: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata9: end_device-7:2: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: ata10: end_device-7:3: dev error handler
Apr 11 08:54:34 SEDECORE kernel: sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 1 tries: 1

 

I am not a network expert, but something doesn't seem right with your network.

 

UD won't turn on the 'Mount' button if the remote server won't respond to a ping.  Go to a command line and manually ping the server in question.  If you don't get a response, you'll have to figure out why not.

 

Hmmm, oddly enough I can't ping it by name, but does respond when using IP.  UD works without issue when I manually enter the IP vs telling it to search for servers.  Much appreciated on that, for some reason I never thought to look there.  As for the network issues, I just replaced my router/mesh system so I have a feeling that has something to do with it.

 

As for your comment about lots of issues with disks, care to elaborate?


Thanks for your assistance!

Link to comment
On 4/10/2021 at 10:45 AM, dlandon said:

Try this:

 

 

Quote

It appears there was a file on disk1 and then possibly it got moved maybe to cache or unassigned device?

You can probably get rid of the messages: go to: Settings/Global Share Settings, change Tunable (support hard links) to No.  But hard links in user shares will not be supported, but you may not need that anyway.

Please report back with findings.

From that thread, I tried turning off Tunable (support hard links), setting it to no instead.  This didn't work.

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.