Test runner for distributed systems based on mininet and tape
npm install tapenet
var test = require('tapenet')
var h1 = test.createHost()
var h2 = test.createHost()
var s1 = test.createSwitch()
h1.link(s1)
h2.link(s1)
test('start a server and connect to it', function (t) {
t.run(h1, function () {
// this is run inside a mininet container on h1
var net = require('net')
h2.on('listening', function (ip) {
var socket = net.connect(10000, ip)
var bufs = []
socket.on('data', data => bufs.push(data))
socket.on('end', function () {
t.same(Buffer.concat(bufs).toString(), 'hello from h2')
t.end()
})
})
})
t.run(h2, function () {
// this is run inside a mininet container on h2
var net = require('net')
var server = net.createServer(function (socket) {
t.pass('got connection')
socket.write('hello from h2')
socket.end()
})
server.listen(10000, function () {
h2.emit('listening', global.ip)
})
})
})
Similar to tape, you can either run a test by simply running a test file with node or you can use the cli runner
npm install -g tapenet
tapenet *.net.js
var h = test.createHost()Same as mininet.createHost()
var s = test.createSwitch()Same as mininet.createSwitch()
test.mininetThe underlying mininet instance.
test(name, runner)Start a test. name is a description of your test and runner should be a function.
When the test is run, runner is ran with the argument t which is a tape test instance.
t.run(host, source)In addition to all the other tape test methods, t.run will run a source function inside a mininet host.
The host will have access to the test object through the t variable, and a set of shared event emitters h1, h2, ...
If you emit an event on any of those event emitters it will be emitted across all mininet hosts currently running.
tapenet comes bundles with some useful topologies.
// same as h1.link(s1), h2.link(s1)
var {h1, h2} = test.topologies.basic()
For more see topologies/
MIT