aws-serverless-applications-study

☁️ A deep study about serverless applications on aws

When we talk about Serverless, it basically means:

You have a piece of code

You want it to get executed by some event (HTTP, Topic Message, Cron, etc), only when this event happens

You waits for an useful output

Example:

You have a brand that stores pictures, but in order to not get it expensive, you usually process the pictures to decrease its size

The picture is uploaded to AWS S3 (Bucket)
The AWS S3 triggers an event "PictureCreated"
The Lambda is invoked by the event "PictureCreated"
The Lambda processes the picture in order to decrease its size
The Lambda shuts down

In this study we'll be using the Serverless Framework which you can use for almost all the cloud platforms, with help of too many programming languages, such as Javascript, C#, Python and so on

Projects inside this repository

image-analysis: A lambda the makes a simple image analysis based on the image url sent to it via HTTP Request.

# patterns
- Factory Programming Pattern

# aws
- AWS Rekognition
- AWS Translate
- AWS API Gateway

# libs
- Axios.js

dynamo-event-source: A lambda triggered by the event of putting data on DynamoDB.

# patterns
- Decorator Programming Pattern
- Factory Programming Pattern
- Serverless Mocked Data Test Pattern

# aws
- AWS DynamoDB

# libs
- Joi.js

scheduler: A cron lambda that gets invoked to read random website page data.

# patterns
- Multiple Environments Pattern
- Serverless Modulated Configuration Pattern

# aws
- AWS DynamoDB

# libs
- Axios.js
- Cheerio.js

meme-maker: A lambda that receives a image url and some text in order to turn it into a meme.

# patterns
- Decorator Programming Pattern
- Serverless Mocked Data Test Pattern

# aws
- AWS Lambda Layers
- AWS API Gateway

# libs
- Axios.js
- Joi.js
- GraphicsMagick

stream-processing: A lambda that processes data by chunk using SQS and S3 services.

# patterns
- Factory Programming Pattern
- Serverless Mocked Data Test Pattern

# aws
- AWS S3
- AWS SQS
- AWS API Gateway

# libs
- LocalStack

graphql-interface: A lambda which runs a GraphQL Interface able to make modifications on a DynamoDB.

# patterns
- Factory Programming Pattern
- Repository Programming Pattern
- Serverless DynamoDB Migrations
- Serverless DynamoDB Seeders

# aws
- AWS DynamoDB
- AWS API Gateway

# libs
- LocalStack
- GraphQL

auth: A lambda which makes authentication and uses other lambda as a middleware to validate authentication.

# aws
- AWS Api Gateway

# libs
- JSON Web Token

website: A serverless app that makes build of a static website, creates a custom url and cdn distribution.

# patterns
- Serverless Components

# aws
- AWS S3
- AWS CloudFront

# libs
- React.js

api-management: A serverless app to manage api selling/distribution with features like rate limit, throttle, etc.

# aws
- AWS API Gateway

# libs
- moment.js

private-database: A lambda that connects to a serverless database using a vpc in order to create some data.

# aws
- AWS RDS Aurora
- AWS VPC
- AWS Secrets Manager
- AWS API Gateway

# libs
- sequelize.js
- moment.js

stackoverflow-surveys-processing: A serverless project that get kaggle reports from StackOverflow and consolidate it into a excel file.

# aws
- AWS Systems Manager
- AWS SES
- AWS S3
- AWS VPC
- AWS EC2
- AWS ECS
- AWS ECR
- AWS CloudWatch
- AWS Lambda

# libs
- mailcomposer.js
- env-var.js

How to init a project

After installing the framework, just type sls on terminal and follow the steps till you get your project started.

Good practices

After starting a new project, try to always get it deployed to the cloud asap in order to avoid some awful and unexpected environmental problems
Trigger the lambda on the cloud and mock the received data, in order to make your tests using it.
Create a config file where you can put all the scripts you execute to the Cloud CLI.

Common errors

My docker is running, I use the flag --docker with sls command and still receive the error: Please start the Docker daemon to use the invoke local Docker integration.

Solution: Execute the command as root user (add sudo before the command).

Useful commands

# Install the framework globally for javascript users
npm install -g serverless

# Init project
sls

# Deploy the project to the cloud
sls deploy

# Invoke the function on cloud
sls invoke -f $FUNCTION_NAME
# Ex: sls invoke hello

# Invoke the function locally
sls invoke local -f $FUNCTION_NAME
# Ex: sls invoke local -f hello

# Configure serverless dashboard
sls dashboard

# Listen for invocation logs (like a CloudWatch)
sls logs -f $FUNCTION_NAME -t
# Ex: sls logs -f hello -t

# Make requests to invoke function based on the request.json file
sls invoke local -f $FUNCTION_NAME --path request.json
# Ex: sls invoke local -f image-analysis --path request.json

# Setup aws credentials for serverless framework
serverless config credentials --provider aws --key $YOUR_ACCESS_KEY --secret $YOUR_SECRET_KEY