Tdarr Documentation

Tdarr uses a conditional plugin stack system to automate media library management. Each plugin in the stack is evaluated, and actions are only performed if specific conditions are met. This allows you to create complex workflows, such as:

1. Transcode non-HEVC files to HEVC
2. Remove subtitles
3. Remove metadata
4. Add AAC stereo audio if missing
5. Remove closed captions

Plugins are written in JavaScript and are located in the Tdarr/Documents directory. You can use community plugins or create your own using the plugin creator interface.

Tdarr is a distributed transcoding system comprised of two main components:

1. Tdarr_Server: The central process that manages tasks and coordinates all connected nodes.
2. Tdarr_Node: The worker processes that run on the same or different devices to perform the actual transcoding/remuxing tasks.

Nodes can be deployed on Windows, Linux (including ARM/ARM64), macOS, and via Docker. For hardware-accelerated transcoding, specific plugins (like Nvidia) may be required depending on your environment (e.g., unRAID or Ubuntu with Nvidia runtime).

When creating custom plugins via the Tdarr interface, the logic is divided into Filters and Actions. Actions are only executed if the associated Filters return true.

Available Filters:

• Filter by codec
• Filter by medium
• Filter by age
• Filter by resolution
• Filter by size

Available Actions:

• Transcode - HandBrake basic options
• Transcode - HandBrake/FFmpeg custom arguments
• Transcode - Standardise audio stream codecs
• Transcode - Add audio stream
• Transcode - Keep one audio stream
• Remux container

Tdarr manages workers to process files in parallel. Workers are categorized into four types:

• Transcode CPU
• Transcode GPU
• Health Check CPU
• Health Check GPU

Worker limits can be managed manually or automatically by the scheduler. The system includes a worker stall detector to ensure efficiency.