Huntarr [Tdarr Edition]: Scale/Kill Tdarr Node Based on PLEX Transcoding

A Docker container that manages Tdarr nodes based on Plex transcoding activity. This tool helps optimize your GPU usage by dynamically adjusting Tdarr workers or stopping Tdarr completely when Plex needs transcoding resources.

Configuration Options

Variable	Description	Example
TDARR_NODE_LOG_PATH	Path to your Tdarr Node log file	/tdarr/logs/Tdarr_Node_Log.txt
TDARR_ALTER_WORKERS	Whether to adjust workers or kill container	true or false
TDARR_DEFAULT_LIMIT	Maximum number of GPU workers	3
TAUTULLI_URL	URL to your Tautulli API	http://10.0.0.10:8181/api/v2
TAUTULLI_API_KEY	Your Tautulli API key	abcdef1234567890
TDARR_API_URL	URL to your Tdarr server	http://10.0.0.10:8265
CONTAINER_NAME	Name of your Tdarr Node container	tdarr_node
OFFSET_THRESHOLD	Transcodes to start reducing GPU workers	1
TRANSCODE_THRESHOLD	Transcodes to kill the Tdarr node	1
RESTART_DELAY	Delay before restarting workers/node	30

Method 1: Worker Scaling Mode

This mode dynamically adjusts the number of Tdarr GPU workers based on Plex transcoding activity, allowing both Plex and Tdarr to share your GPU resources. When Plex needs more resources for transcoding, Huntarr automatically reduces Tdarr workers to ensure smooth playback. When transcoding stops, Tdarr workers are restored.

Docker Run:

docker run -d \
  --name huntarr-tdarr \
  --restart unless-stopped \
  -v /var/run/docker.sock:/var/run/docker.sock:ro \
  -e TAUTULLI_URL=http://10.0.0.10:8181/api/v2 \
  -e TAUTULLI_API_KEY=your_api_key \
  -e TDARR_API_URL=http://10.0.0.10:8265 \
  -e TDARR_NODE_LOG_PATH=/tdarr/logs/Tdarr_Node_Log.txt \
  -e TDARR_ALTER_WORKERS=true \
  -e TDARR_DEFAULT_LIMIT=3 \
  -e CONTAINER_NAME=tdarr_node \
  -e OFFSET_THRESHOLD=1 \
  huntarr/4tdarr:latest

Docker Compose:

version: '3'

services:
  huntarr-tdarr:
    image: huntarr/4tdarr:latest
    container_name: huntarr-tdarr
    restart: unless-stopped
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock:ro
    environment:
      - TAUTULLI_URL=http://10.0.0.10:8181/api/v2
      - TAUTULLI_API_KEY=your_api_key
      - TDARR_API_URL=http://10.0.0.10:8265
      - TDARR_NODE_LOG_PATH=/tdarr/logs/Tdarr_Node_Log.txt
      - TDARR_ALTER_WORKERS=true
      - TDARR_DEFAULT_LIMIT=3
      - CONTAINER_NAME=tdarr_node
      - OFFSET_THRESHOLD=1

With this configuration:

• Maximum of 3 GPU workers when no transcodes are active
• With OFFSET_THRESHOLD=1:
  • 0 transcodes → 3 workers
  • 1 transcode → 3 workers
  • 2 transcodes → 2 workers
  • 3 transcodes → 1 worker
  • 4+ transcodes → 0 workers

Method 2: Node Killer Mode

This mode completely stops the Tdarr container when Plex transcoding is detected. This is ideal for systems with limited GPU resources where Plex and Tdarr cannot effectively run simultaneously. When transcoding is detected, the Tdarr node is immediately stopped, and it will only restart when transcoding activity has ceased for the specified delay period.

Docker Run:

docker run -d \
  --name huntarr-tdarr \
  --restart unless-stopped \
  -v /var/run/docker.sock:/var/run/docker.sock:ro \
  -e TAUTULLI_URL=http://10.0.0.10:8181/api/v2 \
  -e TAUTULLI_API_KEY=your_api_key \
  -e TDARR_API_URL=http://10.0.0.10:8265 \
  -e TDARR_NODE_LOG_PATH=/tdarr/logs/Tdarr_Node_Log.txt \
  -e TDARR_ALTER_WORKERS=false \
  -e TDARR_DEFAULT_LIMIT=2 \
  -e CONTAINER_NAME=tdarr_node \
  -e TRANSCODE_THRESHOLD=1 \
  -e RESTART_DELAY=60 \
  huntarr/4tdarr:latest

Docker Compose:

version: '3'

services:
  huntarr-tdarr:
    image: huntarr/4tdarr:latest
    container_name: huntarr-tdarr
    restart: unless-stopped
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock:ro
    environment:
      - TAUTULLI_URL=http://10.0.0.10:8181/api/v2
      - TAUTULLI_API_KEY=your_api_key
      - TDARR_API_URL=http://10.0.0.10:8265
      - TDARR_NODE_LOG_PATH=/tdarr/logs/Tdarr_Node_Log.txt
      - TDARR_ALTER_WORKERS=false
      - TDARR_DEFAULT_LIMIT=2
      - CONTAINER_NAME=tdarr_node
      - TRANSCODE_THRESHOLD=1
      - RESTART_DELAY=60

With this configuration:

• When a single transcode is detected, the entire Tdarr node is stopped immediately
• The node will restart only when all transcodes are finished and after the restart delay
• RESTART_DELAY=60 means it will wait 60 seconds before restarting the node after transcodes stop

Logs

View the logs to monitor operations:

docker logs -f huntarr-tdarr

Troubleshooting

• If the application can't find the node ID, check that the TDARR_NODE_LOG_PATH is correct and that the container has access to the log file
• If Docker operations fail, ensure that your container has access to the Docker socket
• Check Tautulli connection by verifying your API key and URL