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


Recommended Posts

38 minutes ago, munit85 said:

Just wanted to hop in here and say I have 2 unraid servers connected together with unassigned devices.  I experience trouble when I reboot one server without unmounting the share via the second server.

 

Once i reboot server "unraid" while server "Jefflix" is still running I guess "jefflix" tries to connect to the "unraid" and it seems to crash unassigned devices. It will show unassigned devices as loading without ever kicking in.  Eventually it locks up the whole server and a hard reboot is the only solution. Right now my mover is "running" but not actually moving anything, presumably because of this slow lockup process.

 

This problem is super consistent. i just hard booted "jefflix" so of course now "unraid" unassigned devices is locked up.  Although the icon in the bottom tray show my two preclears are running. I can't see anything under the unassigned devices area.

 

shares are connected as "//JEFFLIX/Media" for example and //Unraid/media" the other direction.

 

Unraid 6.6.6, I believe the entirety of the 6.x I've had this problem.

 

I have them directly connected with mellanox 10gb cards and a DAC cable.

 


Apr 11 15:51:10 Jefflix nginx: 2019/04/11 15:51:10 [error] 10799#10799: *6055578 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.1.9, server: , request: "POST /webGui/include/Download.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "jefflix", referrer: "http://jefflix/Tools/Diagnostics"
Apr 11 15:57:38 Jefflix php-fpm[10740]: [WARNING] [pool www] server reached max_children setting (20), consider raising it

 

i tried to run diagnostics but that locked up as well.

Be sure you have the latest version of UD installed.  Your server should not lock up with remote shares going off line.  UD is probably very busy trying to deal with the preclear status updates and your remote shares going offline.  Remove the preclear plugin and try again.  Post your diagnostics.  You can get the diagnostics from the command line with 'diagnostics'.

Link to comment
59 minutes ago, dlandon said:

Be sure you have the latest version of UD installed.  Your server should not lock up with remote shares going off line.  UD is probably very busy trying to deal with the preclear status updates and your remote shares going offline.  Remove the preclear plugin and try again.  Post your diagnostics.  You can get the diagnostics from the command line with 'diagnostics'.

i have to grab them before rebooting right?

 

the preclear was the only thing still working. the other server not running pre-clear had identical problems.  What seems to be happening is my dockers that are expecting the files on the other server seem to lock up and I can't kill them no matter what and that is the real problem with locking up. docker kill commands are completely ineffective.  

Edited by munit85
Link to comment
15 minutes ago, munit85 said:

i have to grab them before rebooting right?

 

the preclear was the only thing still working. the other server not running pre-clear had identical problems.  What seems to be happening is my dockers that are expecting the files on the other server seem to lock up and I can't kill them no matter what and that is the real problem with locking up. docker kill commands are completely ineffective.  

Yes.

 

UD will eventually time out when the remote shares are not accessible.  It depends on how many remote shares you have mounted.  The issue with the Dockers hanging has nothing to do with UD.  I can imagine they struggle a lot when the remote mounts go away.  Show a screen shot of the UD webgui so I can see what you are mounting remotely.

 

Why so much remote mounting?  You are probably asking for problems like you have when your network goes off, or the remote server is rebooted.  Why not move some of the data to the local server?

Link to comment
20 hours ago, dlandon said:

Yes.

 

UD will eventually time out when the remote shares are not accessible.  It depends on how many remote shares you have mounted.  The issue with the Dockers hanging has nothing to do with UD.  I can imagine they struggle a lot when the remote mounts go away.  Show a screen shot of the UD webgui so I can see what you are mounting remotely.

 

Why so much remote mounting?  You are probably asking for problems like you have when your network goes off, or the remote server is rebooted.  Why not move some of the data to the local server?

"unraid" assignments

https://i.imgur.com/6DYx0Rr.jpg

 

"jefflix" assignments

https://i.imgur.com/aQ6Wkvc.jpg

 

web gui settings for both

https://i.imgur.com/1XNCrlG.jpg

Link to comment
40 minutes ago, dlandon said:

My guess is that your dockers are locking up when they lose access to the shares.

hmm yeah so not really a UD problem. I wonder what I could do to help with that. I just can't get them to shutdown gracefully at all. I figured they'd time out and move on when they can't access files anymore, but yeah..

 

anyway, thanks.

Link to comment

I apologize if this is noted somewhere in the 116 pages of notes, but I couldn't get my SMB shares to work and found that I had to change my complex SMB password. It originally contained a comma (as the special character), which was causing the issue. Removing the comma to another special character resolved the failure. Thanks for the great work! I've been using it for years. I recently decided to move from NFS to SMB shares from my Synology and this issue took some trail and error to resolve.

Link to comment
7 hours ago, mibsy said:

I apologize if this is noted somewhere in the 116 pages of notes, but I couldn't get my SMB shares to work and found that I had to change my complex SMB password. It originally contained a comma (as the special character), which was causing the issue. Removing the comma to another special character resolved the failure. Thanks for the great work! I've been using it for years. I recently decided to move from NFS to SMB shares from my Synology and this issue took some trail and error to resolve.

It's best to not use special characters in passwords used in UD.

Link to comment

Hi All

 

I have 3 mounts to my other server and I have some of the same problems it doesn't time out?

Only way to get them working is stopping the array (Which could be because that stops all dockers?)

 

Is there any way by console to stop UAD? and restart the service like stopping a single docker and not stopping the array?

(Sorry if this has been asked before long thread)

 

Best regards

Casperse

 

Please wait... collecting diagnostic information (this may take several minutes).......

 

Edited by casperse
Link to comment
36 minutes ago, casperse said:

Hi All

 

I have 3 mounts to my other server and I have some of the same problems it doesn't time out?

Only way to get them working is stopping the array (Which could be because that stops all dockers?)

 

Is there any way by console to stop UAD? and restart the service like stopping a single docker and not stopping the array?

(Sorry if this has been asked before long thread)

 

Best regards

Casperse

 

Please wait... collecting diagnostic information (this may take several minutes).......

 

I assume you are saying that your dockers are getting hung up when a remote server goes offline.  UD is not a service and has no involvement in mounted devices and remote shares once mounted.  It is only a UI that manages mounting, unmounting, and monitoring unassigned devices and remote mounts.  The reason UD looks like it is hanging up your system is that it can't perform some of the work needed to monitor the devices.  It does eventually time out if given the chance, but it can't control how long it takes to get the status of the mounted devices.  It is not 'hanging up' your dockers.  Your dockers are hanging themselves.  Stop the offending dockers and UD should become responsive again.  There is very little I can do with UD to solve a CIFS remote mounted share when the server is offline.  That capability is built into the Linux kernel and all I can do is possibly adjust the mount options.  I've done about as much as I can to help with this situation.

 

Remote mounting a share for a docker is probably going to cause a lot of issues like this when the remote server goes offline.  I'm unclear why you are remote mounting the shares in the first place.  So you can have a local mount point for the docker?  You should probably find a better way to do this.  Move the data to the local server?

 

Posting your diagnostics would help me see where you are having problems.

Link to comment
58 minutes ago, dlandon said:

Remote mounting a share for a docker is probably going to cause a lot of issues like this when the remote server goes offline.  I'm unclear why you are remote mounting the shares in the first place.  So you can have a local mount point for the docker?  You should probably find a better way to do this.  Move the data to the local server?

For some things there are definitely other ways to do this, and perhaps some people are overlooking or are unaware of those other ways. For example, the plex application itself has the ability to access media over SMB, so there is no need to remote mount that media in Unraid. Possibly other applications have similar capability.

  • Like 1
Link to comment
  • 2 weeks later...

Hi all,

 

Complete unRAID n00b, so apologies if I'm missing the obvious.

I have my unRAID working fine with some discs, but I'd also like to connect and SMB Share my USB3 Drobo via my unRAID box. And let some Dockers access the files on it.

 

Unassigned Devices seemed to be the answer, but I've found that my Drobo (which is formatted NTFS) is mounted as owned by Root, and therefore the Dockers which run as nobody can't read the drive. (At least I presume that's the problem - I'm mapping them into the Docker container, but the container can't read the contents of the folder).

What am I missing here?

I don't really want to run the Dockers are root, too. But I'd like them to be able to read-write to the Drobo as normal. 

 

Should they access via the SMB share instead of the /mnt/disk ? 

 

Any tips gratefully received.

Link to comment
25 minutes ago, Squid said:

 

25 minutes ago, trurl said:

Are you sure you aren't having this problem instead?

 

https://forums.unraid.net/topic/57181-real-docker-faq/?tab=comments#comment-564314

 

Ah, I restarted the container a few times, but not the whole docker service. 

 

I'll give that a go tonight, thank you! 

Link to comment
5 minutes ago, AndiTails said:

Ah, I restarted the container a few times, but not the whole docker service. 

That is not the permanent fix, unless you somehow manage to insure that the dockers don't start until after the UD is mounted.

 

The last sentence at that link is what you really want to do:

On 5/26/2017 at 10:57 AM, Squid said:

you need to adjust your volume mappings mode (usually either RW or Read Only) to be RW:Slave or Read Only:Slave) by editing the template, and then clicking "Edit" on the volume being passed to the container.

  • Like 1
Link to comment
  • 2 weeks later...

Hello @dlandon I recently starting seeing issues with unassigned devices on one of my servers after about 2 days uptime (on unraid 6.6.7)


Any ideas the Main tab is just trying to load the mounts I have (to two servers running unraid 6.7.0)
the logs are showing this

May 12 09:44:06 Thor nginx: 2019/05/12 09:44:06 [error] 6904#6904: *598311 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 09:49:06 Thor login[496]: ROOT LOGIN on '/dev/pts/1'
May 12 09:52:48 Thor php-fpm[6877]: [WARNING] [pool www] server reached max_children setting (20), consider raising it
May 12 09:53:42 Thor nginx: 2019/05/12 09:53:42 [error] 6904#6904: *600563 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 10:09:51 Thor nginx: 2019/05/12 10:09:51 [error] 6904#6904: *605372 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 10:21:33 Thor login[27364]: ROOT LOGIN on '/dev/pts/2'
May 12 10:24:33 Thor nginx: 2019/05/12 10:24:33 [error] 6904#6904: *607714 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"

Link to comment
8 hours ago, Can0nfan said:

Hello @dlandon I recently starting seeing issues with unassigned devices on one of my servers after about 2 days uptime (on unraid 6.6.7)


Any ideas the Main tab is just trying to load the mounts I have (to two servers running unraid 6.7.0)
the logs are showing this

May 12 09:44:06 Thor nginx: 2019/05/12 09:44:06 [error] 6904#6904: *598311 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 09:49:06 Thor login[496]: ROOT LOGIN on '/dev/pts/1'
May 12 09:52:48 Thor php-fpm[6877]: [WARNING] [pool www] server reached max_children setting (20), consider raising it
May 12 09:53:42 Thor nginx: 2019/05/12 09:53:42 [error] 6904#6904: *600563 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 10:09:51 Thor nginx: 2019/05/12 10:09:51 [error] 6904#6904: *605372 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"
May 12 10:21:33 Thor login[27364]: ROOT LOGIN on '/dev/pts/2'
May 12 10:24:33 Thor nginx: 2019/05/12 10:24:33 [error] 6904#6904: *607714 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.0.0.10, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net", referrer: "https://94cedd8af31b7f0c6ca862dff854ebf6dcc7e6a0.unraid.net/Main"

Post diagnostics.

Link to comment
4 hours ago, dlandon said:

Post diagnostics.

I would love to but they never come via CLI or through the GUI best i can give you is the sys logs posts posted above that i have been leaving running.....I see my nginx is crashing too after a couple days of uptime, i can recover that without a reboot by killing nginx (stop command doesnt work then sarting it and restart php-fpm

 

Edited by Can0nfan
Link to comment
6 hours ago, dlandon said:

Post diagnostics.

I  have tried  something, I noticed nginx crashing since the bios update on my motherboard..I was using legacy boot, i switched over to UEFI boot and will have to wait a couple days to see if issues crop back up..in mean time i also enabled troubleshooting mode via fix common problems so hoping to catch this with that if it occurs again. as I cannot change settings, grab diagnostics at all once nginx and unassigned devices crash docker and my docker apps stay  running smoothly and i can stop and start them without issues....its odd for sure

Link to comment
On 4/21/2019 at 11:25 PM, dlandon said:

I assume you are saying that your dockers are getting hung up when a remote server goes offline.  UD is not a service and has no involvement in mounted devices and remote shares once mounted.  It is only a UI that manages mounting, unmounting, and monitoring unassigned devices and remote mounts.  The reason UD looks like it is hanging up your system is that it can't perform some of the work needed to monitor the devices.  It does eventually time out if given the chance, but it can't control how long it takes to get the status of the mounted devices.  It is not 'hanging up' your dockers.  Your dockers are hanging themselves.  Stop the offending dockers and UD should become responsive again.  There is very little I can do with UD to solve a CIFS remote mounted share when the server is offline.  That capability is built into the Linux kernel and all I can do is possibly adjust the mount options.  I've done about as much as I can to help with this situation.

 

Remote mounting a share for a docker is probably going to cause a lot of issues like this when the remote server goes offline.  I'm unclear why you are remote mounting the shares in the first place.  So you can have a local mount point for the docker?  You should probably find a better way to do this.  Move the data to the local server?

 

Posting your diagnostics would help me see where you are having problems.

Thanks for explaining this to me, your right I have to look at all the dockers!

 

I am pretty sure that if I could create internal SMB mounts inside the docker application this would help on this issue when rebooting the server that is mounted by your excellent tool inside the Docker applications.

 

Than main suspects would be:

Plex

Emby

resilio sync

 

@trurl You mention SMB mounts from plex, but using smb: does not seem to cut it?

Yes ideally I would place everything on one server but I need two :-) and sharing is what a NAS should do best

Link to comment
2 hours ago, casperse said:

You mention SMB mounts from plex, but using smb: does not seem to cut it?

Could you be more specific? I haven't done it that way for years but I used to have another NAS (hpmediavault) with plex server running on my PC and that is the way I accessed my media then. Plex forum may be the best place to get any of those questions answered though.

Link to comment
1 hour ago, trurl said:

Could you be more specific? I haven't done it that way for years but I used to have another NAS (hpmediavault) with plex server running on my PC and that is the way I accessed my media then. Plex forum may be the best place to get any of those questions answered though.

I have Plex docker running on Unraid and have one mounted folder on my Synology NAS through Unassigned Devices

I don't think you can access SMB shares through the Plex UI using smb:

But maybe directly using path in docker config settings for plex?

 

Edited by casperse
Link to comment
  • trurl pinned this topic

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.