I'm hoping someone can help me. I followed Spaceinvaderone's tutorial for setting this up. I am using a GTX1050ti on my unraid server, however I cannot get it to use the GPU to transcode, it keeps using the CPU.

 

My config in unraid is as follows:

 

 

Tdarr config:

 

 

 

 

I'm sure it's something stupid I've done, but I cannot for the life of me figure out it. I've read multiple pages here and in the previous thread that linked here, but cannot figure out where I'm going wrong here. Any help would be greatly appreciated.

If I missed any needed info, please let me know.

How to Add a New Transcoded Audio Stream with Custom Bitrate in Tdarr

 

If you want to keep other audio streams like DTS or TrueHD and add a new transcoded audio stream with a custom bitrate, follow these steps:

 

Install Plugin:

Install the Tdarr_Plugin_00td_action_add_audio_stream_codec (Add Audio Stream Codec) plugin.

 

Edit the Local downloaded Plugin:

Open the plugin in the Tdarr interface.

Do not change the name, ID, or description of the plugin, as it will cause the plugin to fail to load.

 

Replace the Code:

Delete the existing code in the plugin.

Replace it with the code from this link: https://codefile.io/f/FecLfQx7dA.

 

 

This will allow you to keep your original audio streams and add a new audio stream with a custom bitrate.

 

Happy transcoding!

Installed Tdarr today to transcode. Super slow with my older processor of course, but wanted to try it out.  Went out, bought an Nvidia GT 1030. Not ideal, but it fits in the case with the other cards. 

Went through Space Invaders one tutorial and after applying the settings, I get the following error:
 

--runtime=nvidia 'gher. io/haveagitgat/tdarr'
docker: Error response from daemon: unknown or invalid runtime name: nvidia.

 

Without the --runtime=nvidia it installs fine, but no nvidia transcoding.  I've also installed the nvidia drivers for unraid using plugin.

 

Not sure what the error is trying to say. It looks like it's trying to pull from a repo and failing, but could be completely wrong.

Any ideas?

1 hour ago, csimpson said:

looks like it's trying to pull from a repo and failing

The command works fine on mine but it appears you’re using a different repo from mine. You could try using haveagitgat/tdarr in the repository field.

1 hour ago, csimpson said:

Any ideas

Also, did you install the nvidia plugin? Its required for this to work IIRC.

Yes, the nvidia plugin is installed and seemingly working. I've also got the GPU plugin showing me the stats on my dashboard.

I tried using the repo you mentioned and it gave me the exact same error. 

 

When I remove, --runtime=nvidia it installs fine.

 

But then, I can't seem to use the GPU. I also read that the 1030 might not be able to support NVENC but not sure why that would affect the installation process of the docker?

Also, it gives me the same error when I try to do the same thing with Dizquetv.

Edited June 1, 20242 yr by csimpson

Hi, I'm completely new to Tdarr. I came from using Unmanic. I have set up Tdarr server and a node, and everything is detected properly. I imported a flow, and health checks run and the staging fills, but no transcoding is happening. What am I missing? I even upgraded to pro thinking it was some feature I missed, but still the same issue... I was hoping to resolve this before I go on a trip so I can leave it running for a few weeks.

Hi, Anybody tried to transcode audio to be in line with Cinematic Reference Level? Most media out there is with audio level to high. Happy for any feedback.

Anyone else on the latest versions of "ghcr.io/haveagitgat/tdarr_node" & "ghcr.io/haveagitgat/tdarr" getting a mismatched version error? "Server version is 2.25.01 but Node version is 2.24.05.

Please make sure they are the same version. Auto-updating non-Docker Nodes is disabled, you can enable this on the Options tab." In desperation, I have tried enabling the mentioned option but seems to have no impact (because the node is a docker instance). I would have thought the fix is for GitGat to update the node, but I would have expected that lots of people would be reporting the same issue - I presume I must have something set wrong somewhere.

 

Any help is much appreciated!

How can I add another tdarr node container for the second GPU?

 

Cheers,

Is it possible to make the reorder stream plugin run on the GPU? It seems to only want to run on a CPU transcode worker.

TDarr Fails to start.  

 

Have been using Tdarr for a few years now. Recently, when I tried to start it, it presented the error "Image can not be deleted, in use by other container(s)."  When I try to uninstall the image, I get the error "server error."

 

When I listed the docker containers via the console with "docker ps -a" the list presented had Tdarr as "dead."

 

I installed a second instance of the container from the app store, using the same settings as before (tdarr-1).  It started and performed its tasks.

 

Is there a way to force remove the first container, or determine why it will not start?  Diagnostic data attached.

 

Thank-you for your time.

diagnostics-20250215-1402.zip

I get a long list of files that are unable to be processed. Can you help me find out where the issue lays?

 

Thanks so much.

Tdarr.txt

Has anyone been able to use Tdarr for AV1 with an intel battlemage card like the Arc B580? I am struggling to get it to work. I'm on the beta with support for battlemage but keep getting all sorts of errors. I don't know if it's just the version of ffmpeg not supporting battlemage yet.

So I am using a RAM risk as my transcode cache and gave it 24G which is fine for about 95% of what I'm doing. Where I seem to have an issue is that if it has to reorder the streams first, it makes a copy of the original file in transcode cache then proceeds to transcode it so between the reordered original file and then the transcoding, some files are too big to fit in the ramdisk this way.

 

Is there any way to tell Tdarr to replace the existing file on disk with the reordered file and then proceed to transcode it so it doesn't use up so much memory? Or could I even get more fancy that if a file is larger than X, use a different directory for transcode cache?

On 4/3/2025 at 10:02 PM, TallMan206 said:

I get a long list of files that are unable to be processed. Can you help me find out where the issue lays?

 

Thanks so much.

Tdarr.txt

I am also experiencing similar issues to you. Recently been getting lots of " CUDA_ERROR_NOT_INITIALIZED: initialization error" but it only happens with the GPU(1660 super) installed in my unraid server. If I use my RTX 3080 as a remote transcode node on my gaming PC works perfectly fine zero issues. So I'm leaning to something happening in unraid that is dropping the gpu from docker containers. something in tdarr that is dropping nvidia gpu's from view or the lastest nvidia driver v575.51.02 could have bugs in it that are causing issues.

I also get the same issue using a regular tdarr ffmpeg stack not a custom plugin. only started happening the last week or so.

Current Unraid Version: 7.1.2
Latest tdarr version.

O6sW895sO-log.txt

On 5/23/2025 at 2:06 AM, widowshyper said:

I am also experiencing similar issues to you. Recently been getting lots of " CUDA_ERROR_NOT_INITIALIZED: initialization error" but it only happens with the GPU(1660 super) installed in my unraid server. If I use my RTX 3080 as a remote transcode node on my gaming PC works perfectly fine zero issues. So I'm leaning to something happening in unraid that is dropping the gpu from docker containers. something in tdarr that is dropping nvidia gpu's from view or the lastest nvidia driver v575.51.02 could have bugs in it that are causing issues.

I also get the same issue using a regular tdarr ffmpeg stack not a custom plugin. only started happening the last week or so.

Current Unraid Version: 7.1.2
Latest tdarr version.

O6sW895sO-log.txt

Hi,

I found out the bframes option was on because I had a RTX3060 before in my server, but replaced it for a P2000. The P2000 does not support frames.

After setting it to false, I get transcoded files again!

I am using the guyide to get AV1 encoding working in Unraid here:
GitHub - plexguide/Unraid_Intel-ARC_Deployment: Unraid - Comprehensive Guide to Enabling AV1 Encoding, Plex Deployment & GPU Management with Tdarr on Unraid 7.0+

But I have found that even if I use the newest Docker container and Unraid version 7.14 I cant encode with the Intel Arc A380

The intel plugin sees the card but I cant transcode any files

After spending 2 days with AI I get:

According to the guide any Unraid above v.7 should support the A380.
Does anyone have this working?
Is there a FW update to the card I need?

Really grasping for straws now....

Hope someone have the answer AI isnt cutting it :-) LOL

Consistently getting this error message when trying to build my Tdarr and Tdarr_node dockers after swapping out my 1660 for an A380. I've removed all references to Nvidia specific stuff from my config, not sure what to do about the shim task error. Deleting containers and starting fresh is giving me the same error. Anyone encountered this before?

"docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running prestart hook #1: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: initialization error: nvml error: driver not loaded: unknown."

(I had ChatGPT summarize my long and fruitless conversation on solving this issue, hence the AI-ish feel of the text below)

---

🛑 Issue: Transcoded file not found — still writes to /tmp despite configured cache path

Hi everyone,

I'm running into a persistent issue with my setup and could really use some help.

Even after configuring a proper cache location and setting up path translators, my node still writes temp files to c:/tmp on the Windows node and then fails to move/copy the transcoded file back to the library because it can’t find it.

---

⚙️ Setup

• Server version 2.46.01 running in Unraid docker

• Node version 2.46.01 running on Windows 11

• Library cache directory set to /mnt/media/tdarr-cache

• Path translators configured as:

"pathTranslators": [
  {
    "server": "/mnt/media/Movies/",
    "node": "T:/Movies"
  },
  {
    "server": "/mnt/media/TV/",
    "node": "T:/TV"
  },
  {
    "server": "/mnt/media/tdarr-cache",
    "node": "T:/tdarr-cache"
  }
]

---

⚡ Error

Here’s the relevant part of the log:

2025-09-11T12:47:54.357Z: "/mnt/media/Movies/Back to the Future Part III (1990)/Back to the Future Part III (1990) Remux-1080p.mkv"
2025-09-11T12:47:54.362Z: "/tmp/tdarr-workDir2-l4CR8eXuxhu/Back to the Future Part III (1990) Remux-1080p-TdarrCacheFile-wClKc7deJ.mkv"
...
2025-09-11T12:47:55.405Z: File move error: {"errno":-2,"code":"ENOENT","syscall":"rename","path":"/tmp/tdarr-workDir2-l4CR8eXuxhu/Back to the Future Part III (1990) Remux-1080p-TdarrCacheFile-wClKc7deJ.mkv","dest":"/mnt/media/Movies/Back to the Future Part III (1990)/Back to the Future Part III (1990) Remux-1080p-TdarrCacheFile-j5vEo-aB9d.mkv"}
2025-09-11T12:47:55.419Z: File copy error: Error: ENOENT: no such file or directory, lstat '/tmp/tdarr-workDir2-l4CR8eXuxhu/Back to the Future Part III (1990) Remux-1080p-TdarrCacheFile-wClKc7deJ.mkv'

---

📝 What I’ve Tried

• Set the Tdarr transcode cache folder to /mnt/media/tdarr-cache

• Added correct path translators between /mnt/media/tdarr-cache on the server and T:/tdarr-cache on the node

• Restarted both the server and the node

• Verified that T:/tdarr-cache exists and is writable on the node

---

🤔 Problem

Despite this setup, the Windows node writes the working files to /tmp/tdarr-workDir..., which is c:\tmp\, and when the move/copy step happens, it fails because the expected transcoded file is missing from the cache — resulting in a 0-byte file in the destination.

It seems like the node never actually uses the configured cache path and the transcoding job produces no output file.

---

💬 Any Ideas?

• Why is it ignoring the configured cache directory and still writing to /tmp?

• Is there something else I need to set for a mapped node to use the shared cache folder properly?

Any advice would be hugely appreciated — thanks!

On 4/8/2025 at 9:28 AM, JewelCrabCrab said:

Has anyone been able to use Tdarr for AV1 with an intel battlemage card like the Arc B580? I am struggling to get it to work. I'm on the beta with support for battlemage but keep getting all sorts of errors. I don't know if it's just the version of ffmpeg not supporting battlemage yet.

Have you been able to use your B580 yet?

On 8/4/2025 at 4:33 PM, shaivera said:

Consistently getting this error message when trying to build my Tdarr and Tdarr_node dockers after swapping out my 1660 for an A380. I've removed all references to Nvidia specific stuff from my config, not sure what to do about the shim task error. Deleting containers and starting fresh is giving me the same error. Anyone encountered this before?

"docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running prestart hook #1: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: initialization error: nvml error: driver not loaded: unknown."

Were you able to find a solution? I began encountering the same issue this morning.

UPDATE: I was able to resolve by changing my NVIDIA driver from "latest: v590.44.01" back to "Production Branch: v580.105.08" and rebooting the server.

Edited December 5, 2025Dec 5 by RoyalJackV

I'm trying to add an eac3 audio to all my movies with TrueHD audio. I tried to use Add Audio Stream Codec, but it doesn't do anything. It processes the file, but says nothing needs to happen.

When I google around it mentions a Tdarr_Plugin_075a_FFMPEG_Audio, but I don't see that plugin in the container.

This has got to be a common request, any help on how to get it working?

I have a TV that doesn't support truehd so for that TV I want eac3 available.

thanks

Edited December 16, 2025Dec 16 by lovingHDTV