Skip to content

README.md

Latest commit

 

History

History
99 lines (71 loc) · 7.57 KB

README.md

File metadata and controls

99 lines (71 loc) · 7.57 KB
Raw

ChunkyCloud Render Node

Render node for ChunkyCloud, a distributed rendering service for Chunky.

This service runs headlessly, connects to a ChunkyCloud API server, polls for render tasks, downloads the required scene data and resource packs, renders the assigned tile with Chunky, reports progress, uploads the rendered result, and marks the task as finished.

Requirements

Getting started

  1. Get a render node API key (also called render node token) from the ChunkyCloud server you want to connect to

  2. Download the latest release

  3. Run the jar

    java -jar rendernode-v2.0.1.jar \
  --api-key your-render-node-api-key

Important

You may only run a single process per API key. If you want to host multiple render nodes, get a different API key for each of them.

Configuration

Option Default Description
--api https://api.chunkycloud.net ChunkyCloud API endpoint
--api-key Not set Render-node API key
--api-key-file Not set File containing the render-node API key. Useful for container secrets
--cpu-load 100 Maximum Chunky CPU load
-t, --thread-count Number of CPU cores Number of render threads used by Chunky
--job-path ./cc_jobs Directory for temporary per-task data
--texturepacks-path ./cc_texturepacks Directory for downloaded resource packs
--cache-directory ./cc_cache HTTP cache directory for downloaded scene resources
--max-cache-size 512 Maximum HTTP cache size, in MB
--advertised-memory Not set Memory limit advertised to the scheduler, in MB. The render node will only be assigned jobs whose declared memory requirements do not exceed this value. This is not a hard memory limit, as scheduling relies on the memory requirements specified when jobs are created. If omitted or set to 0, no memory-based filtering is applied.

The API key can also be provided through the API_KEY environment variable.

Runtime directories

When no custom paths are provided, the render node creates these directories in the current working directory:

Directory Purpose
cc_jobs Temporary task working directories
cc_texturepacks Downloaded resource packs
cc_cache HTTP cache for API and scene downloads
cc_chunky Chunky settings directory used by the headless renderer

For long-running nodes, mount these directories on persistent storage so resource packs and cached scene files survive container restarts.

Docker

docker run --rm \
  -e API_KEY=your-render-node-api-key \
  -v chunkycloud-render-node-data:/opt/cc-rendernode/data \
  ghcr.io/chunkycloud/render-node:latest

The container runs from /opt/cc-rendernode/data, so the default runtime directories are created inside the mounted data volume. Mounting a volume is recommended so that cached data (e.g. resource packs and scene files) are persisted across container restarts.

You can use a Docker secret to specify the API key, just launch with --api-key-file /run/secrets/chunkycloud-node-token instead of using an environment variable.

Development

Caution

Please do not run customized render nodes against leMaik's ChunkyCloud server. Your contributions are very welcome, but please test them against a local ChunkyCloud server instance.

The project builds a self-contained jar file with its dependencies.

./gradlew build

The runnable jar file is written to:

build/libs/rendernode.jar

The main entry point is de.lemaik.chunkycloud.renderer.Main.

License

This project is licensed under the GNU General Public License v3.0 or later. See LICENSE.txt.