Codementor Events

Getting started with Serverless in 5 minutes

Published Jul 13, 2018Last updated Jan 09, 2019
Getting started with Serverless in 5 minutes

Unless you've been hiding under a rock you probably heard about serverless by now. You've seen the articles, tweets, and the endless string of conferences promoting serverless so much so that you might start to think serverless is one those annoying buzzwords people use to make themselves sounds smart like quantum computing, dark data, blockchain technology or the greenhouse effect. OK, maybe not the last one.

So what's all the fuss?

Technically serverless does not mean you don't have a server but having a service provider manage the server for you. In return, you adhere to a set of rules like the maxim execution duration for the request in 300 seconds and the maximum size of a compressed Lambda function is 50mb.

Granted you'll have to get used to these new rules but if you stop thinking about them as limitations and see them for what they really are which is a best practice recommendation. What I mean is that you could have Lambdas that are bigger than 50mb, sure, but your code will be hard to maintain. You could have request executing longer than 5 minutes but that's something you really want to avoid on traditional servers too, right?

So where's the magic?

The technology behind serverless is nothing new. It was around way longer than you realize. Serverless computing is a cloud computing execution model in which the service provider takes care of the infrastructure. The only reason why you are stuck reading this article right now is because providers like AWS Lambda or Microsoft Azure (among others) have made it possible for everybody to use serverless technology by offering their service at a more than just affordable price.

Ok, fine but how exactly does it work?

You signup with a provider that offers FaaS (Functions as a Service) or PaaS(Platform as a Service) and you use their infrastructure to get your code running without managing anything under the hood (the server).

Yeah, but how?

Take AWS Lambda, the most popular FaaS provider at the moment. You create small functions that you upload into Lambdas. Those functions act like microservices. They execute a bit of code and return a value, or call a second Lambda.
So instead of having a traditional server running PHP or NodeJs that you'd use to create a microservice for let's say returning the sum of two numbers, you upload this logic to a Lambda and sending a request with the two numbers.

“Ok, but how come our old servers aren't ok anymore?”

It's not that traditional servers aren't ok or they are going to disappear, in fact, you might not even want to use Serverless for your app. It's not for a universal solution for all of the mankind problems.
That being said, Serverless has a few benefits that got it so popular. It's so popular in fact that the adoption rate in 2017 was at about 667%.
There are three main benefits of serverless that I personally think stand out: cost, production speed, scalability. Let me explain.

1. Cost

This is a big one. A recent study has shown that companies saved an average of 24% some have reported going from a $10 000 per month bill to $370 (that's right. No I did not forget a zero. It's $370). Here's a link to a serverless case study. Basically, with services like Lambda or Azure, you only pay what you use.

2. Development Speed

After switching to serverless you save around 4 developer work day each month. Since you don't need to worry about the backend you can focus on the app, website and especially on the business logic. Imagine what you can achieve if your team has an extra 4 days each month to work on something else.

3. Scalability

Without getting into too many details, I'm sure you'll agree that scalability is a problem on traditional servers. You've seen plenty of examples where websites and app crash when you get a spike in traffic. With Serverless, this problem goes away as your provider makes sure the app scales gracefully.

Let’s try it out!

Creating a website on serverless is easy, so easy in fact that you can launch a site in under 15 minutes. I won’t bore you with the details but I will however share a link for those that want to start small and move up slowly.

I was conflicted trying to figure out what is the one thing that every website has that would work great on serverless and after careful consideration I realized that contact forms are found on about 99% of the websites I’ve visited in the past years.

Before serverless you’d have to use php to send emails, at least that was the most popular ways of creating a form that sends a message. The code that sends out the form looks something like this:

<?php
$to      = 'nobody@example.com';
$subject = 'the subject';
$message = 'hello';
$headers = 'From: webmaster@example.com' . "\r\n" .
    'Reply-To: webmaster@example.com' . "\r\n" .
    'X-Mailer: PHP/' . phpversion();

mail($to, $subject, $message, $headers);
?>

It seems straightforward, right? On paper is does, but with the php mail() method you’ll inevitably run into problems so you’ll be stuck debugging and looking at logs for a uncomfortably long time. Even after you got it to work the first time, there’s a number of factors that are almost always out of your control that will cause your form to not work, things like ports closing or changing, the host might limit or disable the email system, etc.

Serverless solves these issues with ports and avoids any hosting issues by having everything work from the “almighty” cloud.

I’ll assume you already have node installed. Let’s open your terminal and install a simple boilerplate template: serverless create --template aws-nodejs

Next up we get all the packages we need installed. First we generate our package.json with npm init -y and then we install our dependencies with npm i -s body-parser cors express serverless-http aws sdk

Once that completes you’ll have to load your project in an editor and open your handler.js file and and invoke the SES - Simple Email Service from AWS. Invoking SES is simple enough and looks something like this: ses.sendEmail(params,function(err,data){}). Checkout their documentation to learn more: https://docs.aws.amazon.com/AWSJavaScriptSDK/latest/AWS/SES.html.

What we’ll need to do is sign up for AWS SES and you do that by going to AWS Console, click on SES(you can use the search bar to search for SES), go to the left sidebar and click on email address. Add your email and verify it. It’s a two minute process.

Now let’s gets back to coding the contact form. In your project folder create a new folder called “form”. In that folder we’ll create our form assets. Create a file called script.js and style.css. We’ll go over each of them individually.

Our script.js looks like this:

let name = document.querySelector("#name");
let email = document.querySelector("#email");
let message = document.querySelector("#message");
let error = document.querySelector(".error");
let btn = document.querySelector("button");
let success = document.querySelector(".success");

btn.addEventListener("click", submit);

function submit(e) {
  e.preventDefault();

  var xhr = new XMLHttpRequest();
  xhr.open("POST", "yourendpoint", true);

  xhr.setRequestHeader("Content-type", "application/json");
  xhr.onreadystatechange = function() {
    if (xhr.readyState == XMLHttpRequest.DONE && xhr.status == 200) {
    }
  };
  var data = {
    name: name.value,
    email: email.value,
    message: message.value
  };

  if (name.value && email.value && message.value) {
    success.style.display = "block";
    success.innerHTML = "Thanks for submitting";
    document.querySelector(".all").style.display = "none";
    xhr.send(JSON.stringify(data));

  } else {
      error.style.display = "block";
      error.innerHTML = "Please Fill  All Details";
  }
}

You’ll probably see familiar things in here, as you should. There’s no Lambda “black magic” in here. We are defining our parameters that we’ll use in the form, create the submit function that will call our endpoint. Yes, the endpoint(line 14) is where the magic is. Make a note of that because we’ll get back to it in a second.

Since we are here let’s create the .css too, just to get it out of the way.

*{
box-sizing: border-box;
padding: 0;
margin: 0
}

body{
    font-family: sans-serif;
}

.items{
    display: flex;
    flex-direction: column;
    justify-content: center;
    align-items: center;
    margin: 2rem;
    font-size: 1.1rem;

    padding: 1rem;
}

input{
    padding: 1rem;
    margin: .8rem;
    width: 25rem;
    height: 3rem;
    font-size: 1.1rem;
    border: 3px solid rgb(31, 121, 255);
    outline: none;
}

textarea{
    width: 25rem;
    height: 10rem;
    padding: 1rem;
    display: flex;
    justify-content: center;
     outline: none;
 border: 3px solid rgb(31, 121, 255);
    font-size: 1.1rem;
}

.area{
    display: flex;
}

.area-label{
    margin-left:-2rem;
}

.txt{
    margin-left:.6rem;
}

button{
padding: 1rem;
width: 10rem;
margin-top: 2rem;
text-align: center;
background-color: rgb(75, 224, 75);
font-size: 1.3rem;
color: rgb(2, 2, 15);
box-shadow: 0 .2rem .2rem black;
}

.success{
   background-color: rgb(75, 214, 224);
   font-size: 1.2rem;
   text-transform: capitalize;
    text-align: center;
    margin-top: 10rem;
    padding: 2rem;
    box-shadow: 0 .2rem .2rem rgb(0, 0, 0);
   display: none;
}

.error{
    display: none;
    animation: move .2s ease-in ;
}
@keyframes move{

    0%{
        opacity: 0;
        transform: translateY(-80%);
    }
    50%{
        opacity: .6;
        transform: translate(-20%);
    }

    70%{
        opacity: .7;
        transform: translate(20%);
    }

    100%{
        opacity: 1;
        transform: translate(0);
    }

}

I’m not going to insist on the .css file but if you think I need to explain anything please leave a comment.

Now that we have our form assets created let’s load them in a html. Go to the root of your project and create a new file named index.html.
Since the code is straight forward just paste the following and you are all set.

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name= "viewport" content="width=device-width, initial-scale=1.0">
    <meta http-equiv="X-UA-Compatible" content="ie=edge">
    <title>Contact us</title>
    <link rel="stylesheet" href="form/style.css">

</head>
<body>

<div class="all">
  <form action="" class="items">
    <h1>Contact Us</h1>

      <div class="error-items">
<p style="color:red"  class="error"></p>
</div>
<div>
    <label for="name">Name</label>
    <input type="text"  id="name"  required  placeholder="Enter Your Name" class="name" />
</div>

<div>
    <label for="email">Email</label>
<input type="email"  required placeholder="Your Email" id="email"  class="normal" />
</div>

<div class="area">
<label for="message" class="area-label">Message</label>
<textarea type="text" placeholder="Message" required id="message" class="txt"></textarea>
</div>
<button>Send</button>
  </form>
</div>

<p class="success"></p>
    <script src="./form/script.js"></script>
</body>
</html>

I know you got bored following the article up until now, nothing cool has happened. Well, hold on to your seat because we are going to write our serverless code up next.

We’ll have to first address the handler.js file. Open it from the root of your project and paste in the following code:

const serverless = require("serverless-http");
const AWS = require("aws-sdk");
const express = require("express");
const cors = require("cors");
const bodyParser = require("body-parser");

const app = express();

if (!AWS.config.region) {
  AWS.config.update({
    region: "us-east-1"
  });
}

const ses = new AWS.SES();

app.use(cors());
app.use(bodyParser.urlencoded({ extended: false }));
app.use(bodyParser.json());

app.post("/", (req, res) => {
  const name = req.body.name;
  const email = req.body.email;
  const message = req.body.message;

  const emailParams = {
    Source: "yourname@mail.com", // Your Verified Email
    Destination: {
      ToAddresses: ["yourname@mail.com"] // Your verfied Email
    },
    ReplyToAddresses: [req.body.email],
    Message: {
      Body: {
        Text: {
          Charset: "UTF-8",
          Data: `${message}  from  ${req.body.email}`
        }
      },
      Subject: {
        Charset: "UTF-8",
        Data: "You Received a Message from www.domainname.com"
      }
    }
  };

  ses.sendEmail(emailParams, (err, data) => {
    if (err) {
      res.status(402).send(`${err} ${err.stack}`);
    }
    if (data) {
      res.send(data);
    }
  });
});

module.exports.form = serverless(app);

What we just did is to create the method that calls SES and sends the message. Pretty straightforward stuff.

Next up we are going to have a look at the serverless yml. This is where we are just going to select what region we’ll be using, defining the actual function that the Lambda will execute and the action, which is ses:SendEmail.

service: contact-form

provider:
  name: aws
  runtime: nodejs8.10
  region: us-east-1
  iamRoleStatements:
    - Effect: "Allow"
      Action:
        - "ses:SendEmail"
      Resource: "*"

functions:
  app:
    handler: handler.form
    events:
        - http: ANY /
        - http: 'ANY {proxy+}'


Next stop: delopyment. Sounds complicated but it’s actually a simple task. All you do is open your terminal and run: sls deploy.

You’ll see the log on the screen and when it’s done you’ll have to copy the endpoint and paste it in your script.js file I mentioned earlier. Open script.js and on line 14 replace the word “yourendpoint” with the endpoint supplied. That’s it! Open your html and test out your contact form.

Nicely done! High five all around! It's been quite a journey but it's worth it. You are the proud owner of a brand new website.

23yriddx3cajprkhuk95.gif

Congratulations!

Before I let you go, I’ve got to tell you about monitoring your app which is very important when working with serverless. You'll want to keep an eye out for health problems, invocations time and any errors you might have. Not keeping tabs on your serverless app might cost a lot and it’s not something startups can afford. Or even if they could afford the cost what they simply can’t do is have their service interrupted because a Lambda is taking too long to execute creating a bad experience for their customers.

You can check on your app either via the AWS Console but this is not something that can be easily understood or through a third party like Dashbird.io, a service that monitors your AWS Lambda providing insights on what happens behind the scenes.

Discover and read more posts from John Demian
get started
post commentsBe the first to share your opinion
Show more replies