How do I SSH into a portainer Stack (Jellyfin)

ErikUden · 31. Dezember 2020

Hello!

To make it short: I've recently set up Jellyfin as a portainer stack on my OMV and all works nice! One of the best features Jellyfin offers is hardware accelerated encoding, which I could use since I believe that I have an NVIDIA graphics card and an Intel CPU (Intel Celeron N4100). The issue is that I also would have to set up that hardware accelerated encoding and provide versions of ffmpeg and other stuff into the portainer. I know how to SSH into my OpenMediaVault, but how do I SSH from there into a docker virtual environment?

(If you know anything about the transcoding and how to set it up via Jellyfin, I'd be glad to hear that too!)

TL; DR: How do I SSH into a portainer stack, or possibly even use the FTP to connect to that (if possible)?

gderf · 31. Dezember 2020

Did you read the image documentation Application Setup section?

macom · 31. Dezember 2020

Zitat von ErikUden

How do I SSH into a portainer stack, or possibly even use the FTP to connect to that (if possible)?

I don't use yellyfin, but in general to execute a shell in the container from your ssh connection to OMV:

docker exec -it <container_name> bash

https://docs.docker.com/engine/reference/commandline/exec/

johnvick · 31. Dezember 2020

In you use the linuxserver docker for Jellyfin Intel HW transcoding should work without any extra measures. Lots on this in the Reddit Jellyfin forum.

ErikUden · 31. Dezember 2020

Thank you! This helped a lot.

Zitat von johnvick

In you use the linuxserver docker for Jellyfin Intel HW transcoding should work without any extra measures. Lots on this in the Reddit Jellyfin forum.

Zitat von macom

I don't use yellyfin, but in general to execute a shell in the container from your ssh connection to OMV:

docker exec -it <container_name> bash

https://docs.docker.com/engine/reference/commandline/exec/

Have a wonderful new year's eve!

pcon · 1. Januar 2021

Acceleration on Docker. In order to use hardware acceleration in Docker, the devices must be passed to the container. See the Jellyfin documentation Hardware Acceleration the Parameters and Hardware Acceleration from Docker linuxserver/jellyfin

ErikUden · 1. Januar 2021

Zitat von johnvick

In you use the linuxserver docker for Jellyfin Intel HW transcoding should work without any extra measures. Lots on this in the Reddit Jellyfin forum.

What is Intel HW transcoding? All my options on the standard Playback transcoding menu are:

-None

-AMD AMF

-INTEL QUICK SYNC

-MediaCodec Android

-OpenMAX OMX

-Nvidia NVENC

-Video Acceleration API

-Exynos V4L2 MFC

-Video Toolbox

macom · 1. Januar 2021

Serious? You are still not reading the links that people are providing to you?

Zitat von gderf

Did you read the image documentation Application Setup section?

Zitat von pcon

See the Jellyfin documentation Hardware Acceleration the Parameters and Hardware Acceleration from Docker linuxserver/jellyfin

from https://hub.docker.com/r/linuxserver/jellyfin

Zitat

Hardware Acceleration
Intel
Hardware acceleration users for Intel Quicksync will need to mount their /dev/dri video device inside of the container by passing the following command when running or creating the container:
--device=/dev/dri:/dev/dri
We will automatically ensure the abc user inside of the container has the proper permissions to access this device.

ErikUden · 1. Januar 2021

I've tried all the options provided to me, yet whenever I turn the hardware accelerated transcoding option with Intel Quick Sync on, no video will play. These are all video devices that root can detect, but they are either not mounted to the docker container, or need more configuration

Code

00:00.0 Host bridge: Intel Corporation Gemini Lake Host Bridge (rev 03)
        Flags: bus master, fast devsel, latency 0

00:00.1 Signal processing controller: Intel Corporation Celeron/Pentium Silver Processor Dynamic Platform and Thermal Framework Processor Participant (rev 03)
        Subsystem: Intel Corporation Celeron/Pentium Silver Processor Dynamic Platform and Thermal Framework Processor Participant
        Flags: bus master, fast devsel, latency 0, IRQ 24
        Memory at 80000000 (64-bit, non-prefetchable) [size=32K]
        Capabilities: [d0] Power Management version 3
        Capabilities: [e0] Vendor Specific Information: Len=0c <?>
        Kernel driver in use: proc_thermal
        Kernel modules: processor_thermal_device

00:02.0 VGA compatible controller: Intel Corporation UHD Graphics 605 (rev 03) (prog-if 00 [VGA controller])
        Subsystem: Device 1e39:a002
        Flags: bus master, fast devsel, latency 0, IRQ 135
        Memory at a0000000 (64-bit, non-prefetchable) [size=16M]
        Memory at 90000000 (64-bit, prefetchable) [size=256M]
        I/O ports at f000 [size=64]
        [virtual] Expansion ROM at 000c0000 [disabled] [size=128K]
        Capabilities: [40] Vendor Specific Information: Len=0c <?>

Alles anzeigen

Intel Corporation Celeron/Pentium Silver Processor = Intel(R) Celeron(R) N4100 CPU @ 1.10GHz

Intel Corporation UHD Graphics 605 = Integrated Graphics Card

I added

Code

    devices:
      - /dev/dri:/dev/dri

to my Jellyfin stack and re-deployed it.

I also installed a mod called Jellyfin-opencl-intel with this under the environment variables part of the stack:

Code

- DOCKER_MODS=linuxserver/mods:jellyfin-opencl-intel

Now, obviously this isn't everything I need to do, as can be read here

But I'm really not that sure what all I need to do...

Within Jellyfin there seems to be already a folder titled "ffmpeg"

How do I find its version via the CLI so that I can confirm that it is older than specified?

The tutorial linked above says that:

Zitat

1. QSV is based on VAAPI device on Linux, so please confirm whether you have completed the VAAPI configuration first

Does this mean that I'll have to add VAAPI devices to my portainer stack as described here?

Also, the "Acceleration on Docker" section says that the render group may need to be added to the docker permissions. Do I do this with "group_add:" and then the ID that I find at /etc/group/render?

The documentation also talks about configuring VAAPI acceleration for Debian, will I need to do that too?

The reason why I was so confused about the links, is because they both talk about wildly different things that I need to do in order to configure it correctly.

Linuxserver.io, whom I have the Jellyfin image from, says that I'll only have to add the

Code

--device /dev/dri

Code in order to get it all running, but they also mention VAAPI, yet

jellyfin.org says that I'll have to configure:

1. Acceleration on Docker

(for which I'll need to find the render group ID and add it to the stack with group_add:

2. Maybe OpenCL Accelerated Tone Mapping?? I don't know

3. VAAPI acceleration on Debian/Ubuntu

(by doing a lot)

4. Configuring Intel QuickSync on Debian/Ubuntu

(which requires VAAPI)

I'm just very confused what I'll actually need to do. But as far as I understand it, I'll need to do whatever Jellyfin.org says on the CLI on OMV and then just mount all of that to the Docker image (with the few steps the Docker configuration explained) so that docker can use it.

Is that correct, or did I misunderstand something?

johnvick · 1. Januar 2021

To recap on how I did this on a similar CPU with Ubuntu and on OMV.

Install linuxserver/jellyfin docker (It does the HW transcoding config for you)

set up your media folders

Set up Jellyfin to use VAAPI - disable QuicSync (hard to get going and if VAAPI works then little need)

Optional step - install vainfo on base system to confirm encoding/decoding capabilities of your Intel chip.

Find a 10 bit HEVC file and play it. Open an HTOP window to see what's happening - this will tell if HW transcoding is working.

Optional step - install intel_gpu_top and run it - this will tell if HW transcoding is working.

ErikUden · 2. Januar 2021

Thank you so much! I am an idiot, I thought "HW Transcoding" was its own thing, sorry. Since QuickSync requires VAAPI to work anyway, and Jellyfin also has VAAPI as an option , there really is no need to set QuickSync up with all the complications it has.

I will do the above and post my issues if I encounter any, if I encounter none, I will tell everyone that it works (which it most likely will)

ErikUden · 2. Januar 2021

I've run into a bit of an issue since I do not know what VAAPI drivers to install

https://01.org/linuxmedia/vaapi

I have the Intel Celeron N4100 CPU which has the integrated graphics card "Intel UHD Graphics 600"

I installed vainfo yet whenever I run it on my main server CLI I get this:

The Jellyfin Docker environment seems to have everything set up the right way:

Jellyifin.org says I'll need to follow the relevant documentation which needs drivers to be installed. I don't know what drivers I need since my CPU is a "Gemini Lake" from 2017, meaning Goldmont Plus. It's part of the Atom roadmap. Skylake, which is mentioned in the documentation, is from 2015. Everything after that requires also "linux-firmware," yet I do not know what I require in the first place. VAINFO probably does this because I have no drivers installed, but I do not know what drivers I need. Can you help me with this?

ErikUden · 2. Januar 2021

Edit: I am an idiot. I simply installed the "Intel i965 driver" by typing

Code

sudo apt install i965-va-driver

and then vainfo should return something like:

Code

vainfo
error: XDG_RUNTIME_DIR not set in the environment.
error: can't connect to X server!
libva info: VA-API version 1.4.0
libva info: va_getDriverName() returns 0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so
libva info: Found init function __vaDriverInit_1_4
libva info: va_openDriver() returns 0
vainfo: VA-API version: 1.4 (libva 2.4.0)
vainfo: Driver version: Intel i965 driver for Intel(R) Gemini Lake - 2.3.0
vainfo: Supported profile and entrypoints
VAProfileMPEG2Simple            : VAEntrypointVLD
VAProfileMPEG2Main              : VAEntrypointVLD
VAProfileH264ConstrainedBaseline: VAEntrypointVLD
VAProfileH264ConstrainedBaseline: VAEntrypointEncSlice
VAProfileH264ConstrainedBaseline: VAEntrypointEncSliceLP
VAProfileH264Main               : VAEntrypointVLD
VAProfileH264Main               : VAEntrypointEncSlice
VAProfileH264Main               : VAEntrypointEncSliceLP
VAProfileH264High               : VAEntrypointVLD
VAProfileH264High               : VAEntrypointEncSlice
VAProfileH264High               : VAEntrypointEncSliceLP
VAProfileH264MultiviewHigh      : VAEntrypointVLD
VAProfileH264StereoHigh         : VAEntrypointVLD
VAProfileVC1Simple              : VAEntrypointVLD
VAProfileVC1Main                : VAEntrypointVLD
VAProfileVC1Advanced            : VAEntrypointVLD
VAProfileNone                   : VAEntrypointVideoProc
VAProfileJPEGBaseline           : VAEntrypointVLD
VAProfileJPEGBaseline           : VAEntrypointEncPicture
VAProfileVP8Version0_3          : VAEntrypointVLD
VAProfileHEVCMain               : VAEntrypointVLD
VAProfileHEVCMain10             : VAEntrypointVLD
VAProfileVP9Profile0            : VAEntrypointVLD
VAProfileVP9Profile2            : VAEntrypointVLD

Alles anzeigen

Jellyfin now tells me that:

Which is quite cool.

The issue is that I get insane frame bugs. Something within the decoding must go extremely wrong, here's an example:

This is horrendous, any ideas on how to fix?

How do I SSH into a portainer Stack (Jellyfin)

Acceleration on Docker. In order to use hardware acceleration in Docker, the devices must be passed to the container. See the Jellyfin documentation Hardware Acceleration the Parameters and Hardware Acceleration from Docker linuxserver/jellyfin

Jetzt mitmachen!

Tags