THIS BRANCH IS DEPRECATED

Spark

Spark is a command-line tool used to manage node server processes written by Tj Holowaychuk and Tim Caswell.
It's part of the Connect framework, however can be used standalone with any node server.

This is a highly modified fork of the original Spark library which adds a number of process management and logging features — see the feature list below. This README documents the extra functionality that has been added to this fork.

Please report problems with any of the fork-specific features to Alex Wolfe, rather than the Tj or Tim. Fixes to the forked repository will be merged.

Installation

Using the standard npm installation for Spark will get you the original library. This version is not hosted in the NPM package repository. You must clone and install it locally:

git clone git://github.com/alexkwolfe/spark.git
cd spark
npm install .

Features

• Port/Host to listen on
• SSL hosting with specified key/certificate
• Spawns a configurable number of worker processes
• Environment modes (development/test/production)
• [this fork] Server restart on file system changes
• [this fork] Respawn worker processes gracefully on server restart
• [this fork] Graceful or forcible server stop
• [this fork] Server logging to ./logs/{environment mode}.js (configurable)
• [this fork] Automatically loads configuration from ./config/{environment mode}.js (configurable)
• User/Group to drop to after binding (if run as root)
• Modify the node require paths
• Can change the working directory before running the app
• --comment option to label processes in system tools like ps or htop
• Pass arbitrary code or environment variables to the app

Making an app spark compatible

Any node server can be used with spark. All you need to do it create a file called app.js that exports the instance of http.Server or net.Server.

A hello-world example would look like this:

module.exports = require('http').createServer(function (req, res) {
    res.writeHead(200, {"Content-Type":"text-plain"});
    res.end("Hello World");
});

And then to run it you simply go to the folder containing the app.js and type:

$ spark

The output you'll see will be:

Spark server(34037) listening on http://*:3000 in development mode

Where 34037 is the process id. If you want 4 processes to balance the requests across, no problem.

$ spark -n 4

And you'll see:

Spark server(34049) listening on http://*:3000 in development mode
Spark server(34050) listening on http://*:3000 in development mode
Spark server(34052) listening on http://*:3000 in development mode
Spark server(34051) listening on http://*:3000 in development mode

Commands

Spark supports starting, stopping, restarting, and checking the process status of your app. See spark -h for more information.

The spark restart command is graceful. Configuration will be reloaded (some configuration changes such as host and port require a full stop/start). Each worker will be respawned after currently pending requests have completed.

The spark stop command is also graceful. The server process will wait to shut down until all currently pending requests have completed. To forcibly stop the server without waiting for requests to complete, use the spark interrupt command.

Auto-restart on file changes

Spark will restart when the server's source files are changed on the file system. By default, all .js files in the current working directory will be watched.

When running in an environment other than "development" (i.e. --env production), auto-restart is disabled unless you specify the --watchfile option. The --watchfile option to specifies single file to watch — just touch the watched file to gracefully restart your app.

Auto-restart signals HUP in all environments except development, which always signals INT on file system changes. That means that in development, any pending requests will be forcibly terminated on file system changes.

Logging

Spark exposes 'process.sparkEnv.log' to each server worker process. This logger writes to the file specified in the --logfile option, which defaults to ./logs/{env.name}.log.

Log levels

You can use the log.debug, log.info, and log.error functions to write to the configured file. Use the --loglevel option to control the verbosity of your application's log output. For example, setting --loglevel info will log 'info', and 'error' log messages, but not 'debug' messages. The log level uses sensible defaults based on --env.

Sending an Error to the logger

Logging an Error object will automatically log it as an 'error' and will include the stack trace.

Writing to output

When you use the --verbose option, the Spark process itself will write to STDERR, and will also direct your apps log statements to STDOUT (while adhering to the configured log level).

Customizing log statement format

To customize the date format that the logger uses, replace the log.formatDate(date) function. It takes a single Date parameter and returns a String.

To customize the format of the log statement, replace the log.format(date, level, msg) function with your own function that returns a String.

Logging options in the configuration file

All logging configuration goes under the log key in your config file:

module.exports = {
   host: 'localhost',
   port: 80,
   log : { 
	 level : 'warn', 
	 file  : './logs/applog.log'
   }
}

Log rotation

Spark does not automatically rotate your log files — logrotate is recommended. Make sure to use the copytruncate option, so that your app can continue writing to the same log file:

/var/www/vhosts/your_project/logs/*log {
   daily
   rotate 7
   missingok
   copytruncate
}

MIT License

Copyright (c) 2010 Sencha Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.