Part 1: AWS Managed Database Instances Setup

In this section, you will configure MySQL and Redis as AWS Managed Services. Following the configuration steps below:

RDS MySQL Instance

1. Create DB Subnet Group
2. Name: mysql
3. Description: mysql
4. VPC: cmpe281
5. Add All Subnets Related to this VPC

** Need to have at least Two Private

** Subnets accross two AZ' s

1. Create an RDS MySQL Instance
2. Select "Dev/Test - MySQL" Use Case
3. Select Version 5.5.61
4. Select "Only enable options eligible for RDS Free Usage Tier"
5. Select Instance Type "db.t2.micro"
6. Instance Name: mysql
7. DB User: admin
8. DB Pass: midterm281
9. VPC: cmpe281
10. DB Subnet Group: mysql
11. Public accessibility: No
12. AZ Info: No Preference
13. Security Group: Chose or Create one with MySQL Port Open
14. DB Name: cmpe281
15. Port: 3306
16. IAM DB authentication: Disable
17. Disable auto minor version upgrade
18. Remaining Options: Use Defaults

Install MySQL Client in your "Jump Box"

sudo yum update
Install mysql client
sudo yum install mysql
mysql -u admin -p -h <host>

ElasticCache Redis Instance

1. Cluster Engine: redis
2. Name: redis
3. Description: redis
4. Version: 4.0.10
5. Port: 6379
6. Parameter group: default.redis4.0
7. Node type: cache.t2.micro
8. Number of Replicas: 2
9. Subnet Group: Create New
10. Subnet Group Name: redis
11. SG Description: redis
12. VPC: cmpe281
13. Subnets: select the two private subnets
14. Security Group: Chose or Create one with Redis Port Open

Install Redis Client in your "Jump Box"

sudo yum install gcc
sudo yum install wget
wget http://download.redis.io/redis-stable.tar.gz
tar xvzf redis-stable.tar.gz
cd redis-stable
make
src/redis-cli -h <host> -p 6379

Part 2: Deploy Gumball Cluster with an Internal ELB

• Modify the Go Gumball Code to connect to your Redis and MySQL Servers from Part 1.
• Build the Code and deploy it to AWS (as two EC2 Instances running as Docker Hosts)
• Then put these two EC2 instances behind an Internal Elastic Load Balancer

NOTE: All Instances in this section are in the "Private Network"

After you have the this up, test the Gumball API with the following CURL commands: Note: Replace "host" with the DNS name of your ELB.

curl http://host:3000/ping (Links to an external site.)Links to an external site.
curl -X GET http://host:3000/gumball (Links to an external site.)Links to an external site. 
curl -X POST http://host:3000/order (Links to an external site.)Links to an external site. 
curl -X POST http://host:3000/order/<orderid> (Links to an external site.)Links to an external site.

Then connect to your MySQL and Redis instances and run the following commands:

MySQL:

select * from gumball

Redis:

keys *
get <key>

Part 3: Deploy Kong API Gateway with Auth Key to Gumball

• Deploy Kong API gateway on your publicly facing Docker Host (from Labs)
• Configure Kong to route to the internal Gumball ELB as an upstream API
• Add an API Key in Kong to the Gumball Load Balanced API

Kong Container Dockerized

Consumer Created

Key Value set

Testing

Part 4: Modify & Deploy Node.js on a Local Docker Instance

• Modify the code in Node.js to point at your AWS Kong API Gateway
• Deploy the Node.js App to you local docker host
• Run the Node.js App