statsd-grapher

📈 A statsd backend that serves an http page with the graphical data. This is an alternative to using graphite, and can simply be hosted on a free Heroku instance.

View the annotated source

Quick use

1. Install and set up statsd-grapher

$ npm install statsd statsd-grapher --save
$ cp node_modules/statsd-grapher/exampleConfig.js .
$ ./node_modules/.bin/statsd exampleConfig.js

1. Send some data to your statsd instance

2. Go to your statsd http server to view the graph. The default would be localhost:3000/stats

Configuration

Options

tokens ([String]) - An array of tokens used to authenticate access to the graph. Makes it so that the data can only be accessed by an authenticated user (recommended)
expressPort (Integer) - The port on which the backend http server should run (default: 3000)
sessionStore (Object) - The session store object that will be passed into the store option for express-session (default: MemoryStore)
sessionSecret (String) - The session secret (default: uuid.v4())
sessionExpires (Integer) - The session expiration time in milliseconds (default: 3600000)

Urls

http(s)://hostname:port/stats - Views all stats. This page may be slow to load if you have a lot of metrics. http(s)://hostname:port/stats/:namespace/:metric - Views stats from a particular namespace and metric. If metric is set to all, then all metrics will be returned for that namespace. Ex: /stats/some.name.space/timers
http(s)://hostname:port/metrics/:type/:name - Returns a JSON representation of the metric. Ex: /metrics/counter_rates/some.complete.name.space

Note: If you set tokens in your statsd configuration file, then you need to authenticate your session by appending the token as query parameter like this: /stats?token=your_token_value. The default session expiration time is 1 hour.

Graphing

The library currently uses plotly to graph the stats. It can often lag when handling large data sets, though I consider this to be a reasonable trade-off due to its robust set of features.

Example Graph

Contributors

• Submit an issue citing the feature request first before submitting a PR
• PRs offering different graphing libraries are encouraged 🌳
• Follow the coding style. Use ramda/sanctuary to write small generic functions with corresponding mocha tests. The functions should be composed in a point-free style whenever possible/practical for the DSL

License

Licensed under the MIT license