API Highlight: Introduction

API Highlight: Introduction

By Henry Wagner, Chief Marketing Officer

One of the key features of the Megaport service is our flexible network model. The core of the system is the Megaport API and over the coming months we'll be doing a series of posts around the API with some usage information and examples.

Architecture

The software stack was designed to be modular, allowing us to easily add and remove layers while maintaining a consistent customer experience. One of the key points here for the technical team was to add a layer of abstraction between the network equipment and the management software. Initially this means we’re not locked in to a specific vendor but longer term it means that even moving to a new technology like OpenFlow can be achieved by adding a new module at the Agents layer while the rest of the stack remains the same. As can be seen in the stack, the Megaportal and our phone apps are all dependent on the RESTful API to talk to our backend systems. The intention here is that we use the same interface and mechanisms internally that we expose to customers.

Using the API

There are some useful browser extensions for working with RESTful APIs. Let’s start with something simple and authenticate. If you’d like to follow on with the examples you’ll need to make sure you replace the example credentials.

# Python can be used to format the JSON output if installed 
# | python -mjson.tools 
curl -X POST  
   -d 'username=example&password=secret'  
   https://api.megaport.com/login

Once we authenticate, we’re presented with an object that contains information about the user and the session token we’ll need later.

{
    "channelPartner": false,
    "companyId": 203,
    "companyName": "MEGAPORT",
    "expiryDate": null,
    "language": "English",
    "locale": "Australia",
    "loginDate": 1400643699384,
    "permissions": {
        "203": [
            "services",
            "approve_order",
            "tools",
            "notifications",
            "company_admin",
            "financials",
            "place_order",
            "virtual_employee",
            "tickets",
            "graphs"
        ]
    },
    "personId": 205,
    "session": "dafb8cd2-762a-4633-943b-7d5886c34124",
    "sourceIp": "10.3.1.119",
    "timeZone": "1000",
    "userAgent": "curl/7.30.0",
    "username": "example"
}

In this example you can see who the user is and what permissions they hold. With our [Service Agents] feature a single user account can have permissions under multiple companies. The permissions property will contain a list for each company the user account has access to. The session value contains the token that you’ll need for each subsequent request. Then we can list some information about the company that the user belongs to:

# /secure/party/company/:companyId 
curl -X GET https://api.megaport.com/secure/party/company/203?token=dafb8cd2-762a-4633-943b-7d5886c34124 

Conclusion

In the next post we’ll discuss the service construct and how to work with the Megaport API to report on services and get the related technical resources. If there are any specific things you’d like to see please let us know in the comments and we’ll do our best to cover it in a future writeup.

Related Posts

Multicloud Security: Challenges and Solutions

Multicloud Security: Challenges and Solutions

The more clouds you use, the more security risks your business can face. Here’s what to look out for, and how to stay safe.

Read More
The Future of Cloud Gaming Infrastructure

The Future of Cloud Gaming Infrastructure

As the COVID-19 pandemic continues, more people are turning to cloud gaming—powered by cloud infrastructure—for entertainment. What does its future hold?

Read More
Why Cloud-to-Cloud Connectivity is Key in Canada

Why Cloud-to-Cloud Connectivity is Key in Canada

Eliminate unnecessary latency and modernise your network with direct cloud-to-cloud connectivity. Let’s say your company is located in Calgary and uses Oracle and AWS clouds to run its Oracle databases, with data being sent between the two providers. Common examples of this are companies running Oracle databases within an AWS virtual machines. All of the major cloud providers in Canada – AWS, Microsoft Azure, Google Cloud – have their physical infrastructure located in Toronto, Montreal, or Quebec City. In Canada, there are no cloud regions west of Toronto and currently there are no plans for any of the big hyperscalers to build out new regions in central and western Canada. Your traffic would go from the Oracle region in Toronto or Montreal back to your on-premises network in Calgary and hairpin back out to the AWS infrastructure in Montreal. This routing scenario is illustrated in the diagram below.

Read More