carpenterd-worker

The carpenterd worker process for executing builds.

Install

npm install carpenterd-worker --save

Usage

const carpenterd = require('carpenterd-worker');

const app = new Map();
// Select the directory that contains a `config` directory so that `slay-config`
// works appropriately
app.rootDir = require('path').join(__dirname);

carpenterd(app)
  .start(carpenterd.worker);

Description

carpenterd-worker is responsible for receiving job messages from carpenterd, fetching the fully built npm tarball from a place like amazon s3, execute thewebpack/browserify/babel build using workers-factory.

Dependencies

nsq for job distribution to all the workers in given cluster.
amazon s3 or an s3 like store for storing built assets and for fetching tarballs stored by carpenterd.
A cassandra cluster that is storing the warehouse.ai system data.

Configuration

See the example config config.example.json in this repo.

notes

nsq config option is for a fork of nsq.js that has kubernetes support but otherwise has the same options.
assets.prefix in the config is the bucket name for where the public CDN assets are uploaded.
http is the http port the healhcheck listens on
npm-tars the npm tarball bucket to fetch from
datastar the configuration for cassandra that gets passed directly to datastar.

fritzmonkey / carpenterd-worker