Docker

[hurricanehrndz 149]

Hi,

 

I tried to do an update and apparently messed something up. I probably should not have forced the remove on my container. I should really go through some docker tutorials, but I figured it's a fundamental someone could point out quickly. 

 

~$ sudo docker rm -f <containerID>

~$ sudo docker exec emby-server update

Error response from daemon: No such container: emby-server

~$ systemctl status emby-server.service

● emby-server.service - Emby Media Server

Loaded: loaded (/etc/systemd/system/emby-server.service; enabled; vendor preset: enabled)

Active: failed (Result: timeout) since Sun 2016-12-18 09:22:50 PST; 18s ago

Process: 3175 ExecStartPre=/usr/bin/docker pull emby/embyserver:latest (code=killed, signal=TERM)

 

Dec 18 09:22:32 avocado docker[3175]: 33046b3b32fa: Pulling fs layer

Dec 18 09:22:32 avocado docker[3175]: 8f5cd8a98a50: Pulling fs layer

Dec 18 09:22:32 avocado docker[3175]: 650d4863e9f8: Pulling fs layer

Dec 18 09:22:32 avocado docker[3175]: 650d4863e9f8: Waiting

Dec 18 09:22:32 avocado docker[3175]: 8f5cd8a98a50: Verifying Checksum

Dec 18 09:22:32 avocado docker[3175]: 8f5cd8a98a50: Download complete

Dec 18 09:22:50 avocado systemd[1]: emby-server.service: Start-pre operation timed out. Terminating.

Dec 18 09:22:50 avocado systemd[1]: Failed to start Emby Media Server.

Dec 18 09:22:50 avocado systemd[1]: emby-server.service: Unit entered failed state.

Dec 18 09:22:50 avocado systemd[1]: emby-server.service: Failed with result 'timeout'.

~$ docker ps -all

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES

~$ docker images -a

REPOSITORY TAG IMAGE ID CREATED SIZE

emby/embyserver latest a2e6e3cd4c0f 5 weeks ago 681.9 MB

emby/embyserver <none> d8925dfdae7d 8 weeks ago 681.9 MB

emby/embyserver <none> e8db73a21f89 9 weeks ago 590.3 MB

emby/embyserver <none> 270443911717 10 weeks ago 592 MB

hello-world latest c54a2cc56cbb 5 months ago 1.848 kB

`docker exec emby-server update` is an internal update that works only when the container is running. You executed `docker rm -f` which stop the conatiner and thus the update command fails. docker pull emby/embyserver is what you want. I will update the wrapper script and documentation to included to new functions an internal update mechanism and a docker pull. I will let you know once this is done.

[teiva 18]

`docker exec emby-server update` is an internal update that works only when the container is running. You executed `docker rm -f` which stop the conatiner and thus the update command fails. docker pull emby/embyserver is what you want. I will update the wrapper script and documentation to included to new functions an internal update mechanism and a docker pull. I will let you know once this is done.

@@hurricanehrndz Thank you!

[Pancho 9]

I looked through the scripts it is all good from our end. Perhaps unRAID changed how they apply updates. Perhaps they no longer filter via a tag. I'm not sure.

 

Looking through the list of builds on the hub, I can see that the build prior the latest stable, was the beta and it's version is 259 something. Please remember to be patient, that updates take time. It takes approximately 4 hours after the release has been tagged on github.

 

I'm only speculating here, but I think unRAID now updates to whatever the latest build was and ignores the tag. I check the release notes for 6.2.4 and don't see anything about the update mechanism being changed.

 

Hurricane,

 

Appreciate the follow up.   I just have never experienced anything like that in the past.  In either case, the docker did update and I am now running on the 3.1.0.0 version.

 

Thanks.

[wedgekc 12]

I have problem upgrading to 3.1.0.

 

1. did a docker exec emby-server update ...updated to 3.1.0 without issue and used for a couple hours.

2. restarted the container stuck on upgrading emby database (30%) for hours.

3. tried to copy db from before upgrade and now get connection refused

4. tried new container with no db got the some issue, cannot access web config.

 

 

 

first.txt

[Luke 37242]

I have problem upgrading to 3.1.0.

 

1. did a docker exec emby-server update ...updated to 3.1.0 without issue and used for a couple hours.

2. restarted the container stuck on upgrading emby database (30%) for hours.

3. tried to copy db from before upgrade and now get connection refused

4. tried new container with no db got the some issue, cannot access web config.

 

Hi @@wedgekc, this log is from the previous server release, are you sure you updated?  What I would suggest doing is making sure you're actually on 3.1, then delete the library.db file, start up the server and run a library scan. Let us know how it goes. Thanks !

[wedgekc 12]

Thanks for the reply.

 

I noticed that too , although the time stamp is after I upgraded.  I simply did a docker logs emby-server while waiting because I could not open the dashboard (white screen: wait while your emby dabatse is being updted %30) and that is what I got.  I know I was on 3.1 because I noticed the interface changes after the initial update and asked to update plugins, etc

 

I will try deleting library.db and let you know, hopefully I can get to the dashboard and run a scan.

 

Thanks

[wedgekc 12]

I'm still having issues here but slowly sorting them out.

 

Decided to start from scratch. Not sure why but after the following steps the ports for emby were not exposed, hence no dashboard.

1. moved .emby-server from my home directory

2. deleted container and image

3. sudo docker run -it --rm -v /usr/local/bin:/target -e "APP_USER=chris" \emby/embyserver instl

4. emby-server

 

after doing that I cannot access the dashboard until I expose the ports for emby. Once doing that I can access the dashboard. Let me know if I haven't done something incorrectly.

 

starting with the wrapper emby-server seems to cause issues with the ports.  If I use docker start emby-server it seems to work.

 

Thanks

docker_inspect.txt

Edited December 21, 2016 by wedgekc

[hurricanehrndz 149]

I'm still having issues here but slowly sorting them out.

 

Decided to start from scratch. Not sure why but after the following steps the ports for emby were not exposed, hence no dashboard.

1. moved .emby-server from my home directory

2. deleted container and image

3. sudo docker run -it --rm -v /usr/local/bin:/target -e "APP_USER=chris" \emby/embyserver instl

4. emby-server

 

after doing that I cannot access the dashboard until I expose the ports for emby. Once doing that I can access the dashboard. Let me know if I haven't done something incorrectly.

 

starting with the wrapper emby-server seems to cause issues with the ports.  If I use docker start emby-server it seems to work.

 

Thanks

Thank you for reporting the issue. With all the changes and features going into the wrapper script it seems like the network mode was accidentally removed from the wrapper. We will get it sorted by the end of the week.

[hurricanehrndz 149]

Build has been triggered to fix wrapper script.

[wedgekc 12]

 

Build has been triggered to fix wrapper script.

 

Thanks! New build solves the problem.

[Luke 37242]

Thanks for confirming.

[chvb 1]

hey,

 

UNRAID now supports the kernel driver for the iGPU.

So now we can Hardware transcode inside a docker.

 

With Plex everything is working out of the box.

 

I have to set this extra parameter:

--device /dev/dri:/dev/dri

 

I also tried this with the Emby Docker and enabled Quicksync and VAAPI.

But it doesn't work. He don't start's to play any media.

In the log he say that he can find the VA device /dev/dri/renderD128

 

Any Ideas?

[hurricanehrndz 149]

I would check permissions and ensure before anything else

 

Sent from my ONEPLUS A3000 using Tapatalk

[chvb 1]

I would check permissions and ensure before anything else

 

Sent from my ONEPLUS A3000 using Tapatalk

 

thanks for your answer @@hurricanehrndz.

Can you tell me, how i can check the permissions?

 

Many thanks.

[simonk83 7]

Ok, I'm going to have to ask for some help here.

 

I run a Synology NAS with Docker and have several set up so far (all working great), being Sonarr, Radarr, NzbGet, Unifi, Jackett and Deluge.

 

I have Emby running as well, and all was fine until I noticed that my metadata was being saved in a folder in the Emby config and I wanted it in with all my media.  Long story short, it seems Emby is having trouble writing to my shares.  I use the same UID (1053) and GID (100) as all my other Dockers (Sonarr writes metadata to the same folders just fine for example), but in the Emby log I see a bunch of Unauthorized Access errors:

2017-02-08 07:05:13.5253 Error ProviderManager: Error in metadata saver
    *** Error Report ***
    Version: 3.2.0.0
    Command line: /usr/lib/emby-server/bin/MediaBrowser.Server.Mono.exe -programdata /config -ffmpeg /bin/ffmpeg -ffprobe /bin/ffprobe -restartpath /usr/lib/emby-server/restart.sh
    Operating system: Unix 3.10.77.0
    64-Bit OS: True
    64-Bit Process: True
    Mono: 4.6.2 (Stable 4.6.2.7/08fd525 Mon Jan 30 13:12:40 UTC 2017)
    Processor count: 4
    Program data path: /config
    Application directory: /usr/lib/emby-server/bin
    System.UnauthorizedAccessException: Access to the path "/carlytv" is denied.
     at System.IO.Directory.CreateDirectoriesInternal (System.String path) [0x00070] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.Directory.CreateDirectory (System.String path) [0x000a4] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.DirectoryInfo.Create () [0x00000] in <8f2c484307284b51944a1a13a14c0266>:0
     at (wrapper remoting-invoke-with-check) System.IO.DirectoryInfo:Create ()
     at System.IO.Directory.CreateDirectoriesInternal (System.String path) [0x0003f] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.Directory.CreateDirectory (System.String path) [0x000a4] in <8f2c484307284b51944a1a13a14c0266>:0
     at Emby.Common.Implementations.IO.ManagedFileSystem.CreateDirectory (System.String path) [0x00000] in <42c5174dc9d74c7a87560f62bb636308>:0
     at MediaBrowser.XbmcMetadata.Savers.BaseNfoSaver.SaveToFile (System.IO.Stream stream, System.String path) [0x0000c] in <6b741bebd2d349378c6fc59d689940be>:0
     at MediaBrowser.XbmcMetadata.Savers.BaseNfoSaver.Save (MediaBrowser.Controller.Entities.IHasMetadata item, System.Threading.CancellationToken cancellationToken) [0x00026] in <6b741bebd2d349378c6fc59d689940be>:0
     at MediaBrowser.Providers.Manager.ProviderManager+<SaveMetadata>c__async3.MoveNext () [0x001ff] in <7660f7892d2c49cc8fa7e4be412ff52e>:0
    System.UnauthorizedAccessException
     at System.IO.Directory.CreateDirectoriesInternal (System.String path) [0x00070] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.Directory.CreateDirectory (System.String path) [0x000a4] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.DirectoryInfo.Create () [0x00000] in <8f2c484307284b51944a1a13a14c0266>:0
     at (wrapper remoting-invoke-with-check) System.IO.DirectoryInfo:Create ()
     at System.IO.Directory.CreateDirectoriesInternal (System.String path) [0x0003f] in <8f2c484307284b51944a1a13a14c0266>:0
     at System.IO.Directory.CreateDirectory (System.String path) [0x000a4] in <8f2c484307284b51944a1a13a14c0266>:0
     at Emby.Common.Implementations.IO.ManagedFileSystem.CreateDirectory (System.String path) [0x00000] in <42c5174dc9d74c7a87560f62bb636308>:0
     at MediaBrowser.XbmcMetadata.Savers.BaseNfoSaver.SaveToFile (System.IO.Stream stream, System.String path) [0x0000c] in <6b741bebd2d349378c6fc59d689940be>:0
     at MediaBrowser.XbmcMetadata.Savers.BaseNfoSaver.Save (MediaBrowser.Controller.Entities.IHasMetadata item, System.Threading.CancellationToken cancellationToken) [0x00026] in <6b741bebd2d349378c6fc59d689940be>:0
     at MediaBrowser.Providers.Manager.ProviderManager+<SaveMetadata>c__async3.MoveNext () [0x001ff] in <7660f7892d2c49cc8fa7e4be412ff52e>:0
    

and a whole bunch of a similar error trying to access .nfo's in my other paths.

 

Logging into the Synology with my user shows:

Simon@Diskstation:~$ id
uid=1053(Simon) gid=100(users) groups=100(users),101(administrators),1023(http),65536(sc-download),65537(sc-media)

My media folders (Movies, TV and CarlyTV) are in a share called NAS:

 

 

These are the user groups with read/write to that share:

 

User Simon is part of the users group:

 

 

User Simon has the following rights:

 

 

 

Docker Environment Variables:

 

 

Docker Volumes:

 

 

 

Emby shares:

 

http://imgur.com/a/RqH0b

 

 

 

I did a bunch of reading and noticed there was an option to set APP_USER variable, so I removed the GID and UID and added that (APP_USER = Simon).   This actually did work and I started to see Emby populate my folders as expected.   This morning however, it's back to not working with the auth errors so I'm stumped.

Edited February 7, 2017 by simonk83

[hurricanehrndz 149]

Try using APP_UID and APP_GID as well

 

Sent from my ONEPLUS A3000 using Tapatalk

[simonk83 7]

Try using APP_UID and APP_GID as well

 

Sent from my ONEPLUS A3000 using Tapatalk

 

No joy I'm afraid.  It looked like it was working, but then:

System.UnauthorizedAccessException: Access to the path "/movies/moviename/moviename.nfo" is denied.

I'm wondering though, could it be that these .nfo's existed previously and Emby just doesn't have the rights to amend them for whatever reason (no idea why that would be).   Maybe I'll try deleting all of the .nfo's (they seem the main culprit) and have Emby recreate them from scratch.

[Sergio 4]

What filesystem is that located? If it was aufs, for example, it could be that permissions on the underlaying fss could be wrong, even when the pool Mount had right permissions.

Or being de-sincronized between disc and catalog.

 

Enviado desde mi A0001 mediante Tapatalk

Edited February 8, 2017 by Sergio

[simonk83 7]

What filesystem is that located? If it was aufs, for example, it could be that permissions on the underlaying fss could be wrong, even when the pool Mount had right permissions.

Or being de-sincronized between disc and catalog.

 

Enviado desde mi A0001 mediante Tapatalk

 

It's a Synology NAS, looks like it's ext4.

[hurricanehrndz 149]

Nfo could have been created by a kodi client

 

Sent from my ONEPLUS A3000 using Tapatalk

[Sergio 4]

No idea. Just make sure permissions are right and there are no acl.

 

Enviado desde mi A0001 mediante Tapatalk

[simonk83 7]

Nfo could have been created by a kodi client

 

Sent from my ONEPLUS A3000 using Tapatalk

 

Very likely.  I was using Kodi and mysql for years.  Just installed Emby recently to replace the mysql part. 

 

I've removed all the nfos and Emby is busy creating its own now.  Seems OK so far but I'll keep a keen eye on it

Edited February 8, 2017 by simonk83

[Luke 37242]

Thanks for reporting back !

[NNate 10]

Just got Emby up and running in Docker on Unraid - coming from the windows world

 

I didn't want to share my whole /mnt folder with the docker, so I made custom mappings.

 

Most things worked, but had some weird behavior when I tried to create a "/metadata" -> "/mnt/user/metadata" mapping for my metadata folder and use path substitution. 

 

For example, I pointed the custom metadata location at "/metadata" and then set up path substitution for "\\Tower\metadata". Looking at the nfos created, the actors images were pointing at "\\Tower\metadata\\Tower\metadata\..." - note the double substitution

 

Now, when I changed my docker to map my metadata folder to "/EmbyMetadata", the path substitution came out as you'd expect - "\\Tower\metadata\..."

 

I don't know if this is a bug unique to the docker or the server in general. Just throwing it out there.

 

Edit: This is with the 3.2.0 release. I noticed an update to 3.2.1 this morning - perhaps that fixed the bug I'm seeing, but guessing not?

Edited February 10, 2017 by NNate

[Luke 37242]

Just got Emby up and running in Docker on Unraid - coming from the windows world

 

I didn't want to share my whole /mnt folder with the docker, so I made custom mappings.

 

Most things worked, but had some weird behavior when I tried to create a "/metadata" -> "/mnt/user/metadata" mapping for my metadata folder and use path substitution. 

 

For example, I pointed the custom metadata location at "/metadata" and then set up path substitution for "\\Tower\metadata". Looking at the nfos created, the actors images were pointing at "\\Tower\metadata\\Tower\metadata\..." - note the double substitution

 

Now, when I changed my docker to map my metadata folder to "/EmbyMetadata", the path substitution came out as you'd expect - "\\Tower\metadata\..."

 

I don't know if this is a bug unique to the docker or the server in general. Just throwing it out there.

 

Edit: This is with the 3.2.0 release. I noticed an update to 3.2.1 this morning - perhaps that fixed the bug I'm seeing, but guessing not?

 

This is by design. When you configure the metadata folder to /X, we use /X/metadata. What's very common is that users will configure all of the custom paths to the same location, for example, metadata, transcoding, cache, etc, all pointing to /X. Then they would come back and report problems when different parts of the system were stepping on each other's toes due to all of them using the same folder. So this left us with no choice but to add the extra sub-folder. We've been doing it for a while.