Kong

Kong is a scalable and customizable API Management Layer built on top of nginx.

• Requirements
• Installation
• Usage
• Development

Requirements

• Lua 5.1
• Luarocks 2.2.0
• OpenResty 1.7.7.2
• Cassandra 2.1

Installation

From source

[sudo] make install

Usage

Use Kong through the bin/kong executable.

The first time ever you're running Kong, you need to make sure to setup Cassandra by executing:

bin/kong migrate

To start Kong:

bin/kong start

See all the available options, with bin/kong -h.

Development

Running Kong for development requires you to run:

make dev

This will create your environment configuration files (dev and tests). Setup your database access for each of these enviroments (be careful about keyspaces, since Kong already uses kong and unit tests already use kong_tests).

• Run the tests:

make test-all

• Run it:

bin/kong -c config.dev/kong.yaml -n config.dev/nginx.conf

Makefile

When developing, use the Makefile for doing the following operations:

Name	Description
install	Install the Kong luarock globally
dev	Setup your development enviroment (creates dev and tests configurations)
clean	Cleans the development environment
reset	Reset your database schema according to the development Kong config inside the dev folder
seed	Seed your database according to the development Kong config inside the dev folder
drop	Drop your database according to the development Kong config inside the dev folder
test	Runs the unit tests
test-proxy	Runs the proxy integration tests
test-web	Runs the web integration tests
test-all	Runs all unit + integration tests at once

Scripts

Those script provide handy features while developing Kong:

Name	Commands	Description	Arguments
migrate
	create --conf=[conf]	Create a migration file for all available databases in the given conf	--name=[name] Name of the migration
	migrate --conf=[conf]	Migrate the database set in the given conf
	rollback --conf=[conf]	Rollback to the latest executed migration (TODO)
	reset --conf=[conf]	Rollback all migrations
seed
	seed --conf=[conf]	Seed the database configured in the given conf	-s (Optional) No output
			-r (Optional) Also populate random data (1000 by default)
	drop --conf=[conf]	Drop the database configured in the given conf	-s (Optional) No output