Storage API

This is Team 1's Assignment 6 for ECE 4574 at Virginia Tech

• Accessing the Website/Server ui
  • What is this repository for?

• Single Docker Container
  • Prerequisites to Run
  • How to generate a single Docker container and run locally
  • How to Stop the Docker Container
  • To configure host/ui address
  • Changing the port number
• Local Docker Swarm
  • Prerequisites to Run
  • Creating a Docker Swarm
    • Creating the Docker Machines
    • Initializing the Swarm
    • Building the container in the docker machines
    • Deploying the service
    • Testing the service
    • Stopping the Swarm
• Online Docker Swarm
  • Prerequisites to Run
  • Creating a Docker Swarm
  • To configure host/ui address
  • Changing the port number

Accessing the Website/Server ui

To access Existing EC2 Servers Go to this link Main Website

What is this repository for?

• This project is to provide a Database and API for storage and retrieval of information.
• Version 3.0

Single Docker Container

Prerequisites to Run

• Docker is installed
• Your device is connected to the internet (to clone the repository and download required libraries)

How to generate a single Docker container and run locally

1. go ahead and gitclone this repo git clone https://github.com/ldevr2t1/docker_storage.git
2. Run git checkout local
3. Navigate into the web directory (i.e. cd web)
4. run docker commands to get server running - may have to sudo
   • docker-compose build
   • docker-compose run
5. Access your machine-ip address (docker-machine ip) in your web browser
   • The UI should be viewable at **192.168.99.100/v2/ui**
6. If you cannot access the UI then change the 'host' address in the swagger.yaml file
   • To get Machine ip address: Run docker-machine ip
   • Update swagger.yaml: host: "<Machine Ip-Address>"
   • Repeat steps 4 and 5

How to Stop the Docker Container

Run the command docker stop [id], where [id] is the generated id number of your container. If you do not know what your container id is, use the command docker ps to view all running containers.

To configure host/ui address

1. To change the server's IP-address edit the 'host' parameter in main.

   • File: web/swagger_server/__main__.py
     • app.run(host='<your_address>', port=<port_number>)

2. Change the host parameter in the swagger.yaml file for the UI to work

   • web/swagger_server/server/swagger.yaml
     • host: "<your_address>:<port_number>"

Changing the port number

1. Change the <port_number> in the same files for configuring the host/ui address
2. Go into the base directory /docker_pathfind and edit the docker-compose.yml
   • ports: - "<port_number>:<port_number>"
3. Change the Dockerfile (i.e. change the EXPOSE #)
   • #Expose port # for testing - EXPOSE <port_number>

Local Docker Swarm

Prerequisites to Run

• Docker is installed
• VirtualBox is installed
• Your device is connected to the internet (to clone the repository and download required libraries)

Creating a Docker Swarm

In order to build a Docker Swarm locally on your machine, follow the steps below:

Creating the Docker Machines

1. Build the Manager with docker-machine create <name_of_manager>, where name_of_manager is the name of your machine
2. Build the Workers with docker-machine create <name_of_worker>, where name_of_worker is the name of your machine. You may repeat this as many times as you'd like, but remember the machine names!.

Initializing the Swarm

Run these commands in the following order:

1. SSH into your manager directory with docker-machine ssh <name_of_manager>
2. Run docker swarm init --advertise-addr <ip_of_manager> to initialize the manager
3. Copy the returned command, and detach from the container with ^d
4. Now, ssh into the other workers using the command in part 1, pasting in the command returned from part 3.

Building the container in the docker machines

1. Run docker-machine env <name_of_worker> to get access to a command in the commments. This command allows you to reroute your container from your own files, and is different depending on the OS. Run that command now.
2. go ahead and gitclone this repo git clone https://github.com/ldevr2t1/docker_storage.git
3. Navigate to the cloned directory
4. Run git checkout local
5. Run docker-compose -f docker-compose-swarm.yml build
6. Repeat step 1 and 5 with all workers.

Deploying the service

1. Run docker-machine env <name_of_manager> to get access to a command in the commments. This command allows you to reroute your container from your own files, and is different depending on the OS. Run that command now.
2. run docker stack deploy -c docker-compose-swarm.yml <name_of_service, where name_of_service is the name of your swarm.

Testing the service

1. Type docker-machine ls to see the addresses hosting your machines
2. Go to a browser and type :8000/v3/ui/, where address is the ip of one of the machines

Stopping the Swarm

docker-machine stop <name_of_machine>

Online Docker Swarm

In order to build a Docker Swarm remotely, follow the steps below:

Prerequisites to Run

• Docker is installed
• Your device is connected to the internet (to clone the repository and download required libraries)

Creating a Docker Swarm

Follow the steps located in Local Docker Swarm, but use docker-machine create driver amazonec2 --amazonec2-access-key <access_key> --amazonec2-secret-key <secret_key> <name_of_machine> instead.

To configure host/ui address

1. To change the server's IP-address edit the 'host' parameter in main.

   • File: web/swagger_server/__main__.py
     • app.run(host='<your_address>', port=<port_number>)

2. Change the host parameter in the swagger.yaml file for the UI to work

   • web/swagger_server/server/swagger.yaml
     • host: "<your_address>:<port_number>"

Changing the port number

1. Change the <port_number> in the same files for configuring the host/ui address
2. Go into the base directory /docker_pathfind and edit the docker-compose.yml
   • ports: - "<port_number>:<port_number>"
3. Change the Dockerfile (i.e. change the EXPOSE #)
   • #Expose port # for testing - EXPOSE <port_number>

Authors

• Joshua Chung
• David Gwizdala