Skip to content

Latest commit

 

History

History
290 lines (264 loc) · 13 KB

README.md

File metadata and controls

290 lines (264 loc) · 13 KB

API VDI

Description

This project is an API to manage the VDI infrastructure running on OpenStack of the INSA CVL. This API can manage the following elements :

  • Users
  • Groups
  • VMs
  • Templates

You can find the frontend of this API here.
The documentation below is for the installation of the API, scheduler and the frontend.

Installation without Docker

Requirements

  1. Linux : Ubuntu 20.04, Debian 11,12
  2. Database : Mariadb or MySQL
  3. Python 3.11.2
  4. Pip or Pipx : you can even create a virtual environment to run those scripts
  5. Create a user vdi with the home directory /home/vdi
  6. Create a folder /home/vdi/VDI-API for the API and and clone this repository in it
    6.1. mkdir -p /home/vdi/VDI-API
    6.2. cd /home/vdi/VDI-API
    6.3. git clone https://github.com/AlextTheGeek/api-vdi.git
    6.4. cd api-vdi
  7. Install the requirements of the api and scheduler: pip install -r requirements.txt
  8. Install the openstack client : apt install python3-openstacksdk or pip install python3-openstacksdk. Your server can need some other packages like python3-openstackclient that you can install with apt install python3-openstackclient or pip install python3-openstackclient.
  9. Create a folder /home/vdi/VDI-APP for the API and and clone this repository in it
    8.1. mkdir -p /home/vdi/VDI-APP
    8.2. cd /home/vdi/VDI-APP
    8.3. git clone https://github.com/loimax/vdi.git
    8.4. cd vdi
  10. Install the requirements of the vdi (frontend) : pip install -r requirements.txt
  11. Create a folder /var/log/VDI/API for the logs of the API : mkdir -p /var/log/VDI/APP
    10.1. Set the rights of the user who runs the API (here vdi) on this folder : chown -R vdi:vdi /var/log/VDI/APP
    10.2 Create all necessary files for the logs : touch /var/log/VDI/API/access.log /var/log/VDI/API/error.log
  12. Create a folder /var/log/VDI/API for the logs of the API : mkdir -p /var/log/VDI/API
    11.1. Set the rights of the user who runs the API (here vdi) on this folder : chown -R vdi:vdi /var/log/VDI/API 11.2 Create all necessary files for the logs : touch /var/log/VDI/API/access.log /var/log/VDI/API/error.log /var/log/VDI/APP/api-flask.log
  13. Create a folder /var/log/VDI/Scheduler for the logs of the scheduler : mkdir -p /var/log/VDI/SCHEDULER 12.1. Set the rights of the user who runs the scheduler (here vdi) on this folder : chown -R vdi:vdi /var/log/VDI/SCHEDULER

Database

We recommend using a MariaDB or MySQL database that you can install on your server.
After installation, you can create a database and a user for the API and the scheduler.

CREATE DATABASE vdi;
CREATE USER 'vdi'@'%' IDENTIFIED BY 'azerty';
GRANT ALL PRIVILEGES ON vdi.* TO 'vdi'@'%';
FLUSH PRIVILEGES;

API

  1. Modify the file main.py to set the environment variables :
  • URL_VNC : URL of the VNC server (ex: https://vnc.insa-cvl.com)
  • URL_VDI : URL of the VDI server (ex: https://vdi.insa-cvl.com)
  • URL_API : URL of the API server (ex: https://api.insa-cvl.com)
  • URL_API_CAS : URL of the API without https:// (ex: api.insa-cvl.com)
  • app.config['SECRET_KEY'] : Secret key of the API
  • app.config['TOKEN_SECRET_KEY'] : Secret key of the token
  • app.config['SESSION_COOKIE_DOMAIN'] : Domain of the session cookie (ex: insa-cvl.com)
  • app.config['SQLALCHEMY_DATABASE_URI'] : URI of the database (ex : mysql+mysqlconnector://vdi:azerty@vdi-db/vdi)
  1. Modify the file back_openstack.py to set the environment variables :
  • PRIVATE_NETWORK_ID : ID of the private network
  • USERNAME : Username of the OpenStack user (ex: vdi)
  • PASSWORD : Password of the OpenStack user (ex: vdi)
  • AUTH_URL : URL of the OpenStack authentication (ex: http://172.10.3.60:5000/v3)
  • PROJECT_NAME : Name of the OpenStack project (ex: vdi)
  • USER_DOMAIN_ID : ID of the OpenStack user domain (ex: default)
  • PROJECT_DOMAIN_ID : ID of the OpenStack project domain (ex: default)
  1. Create a file /etc/systemd/system/vdi-api.service with the following content :
[Unit]
Description=Gunicorn instance to serve API VDI
After=network.target

[Service]
User=vdi
Group=vdi
WorkingDirectory=/path/to/the/folder # Path to the folder containing the wsgi.py file
Environment="PATH=/home/vdi/.local/bin" # Necessary to find gunicorn if you installed it with pipx
ExecStart=/home/vdi/.local/bin/gunicorn --access-logfile /var/log/VDI/API/access.log --error-logfile /var/log/VDI/API/error.log --workers 3 --bind 0.0.0.0:5001 main:app # You need to check 
Restart=always

[Install]
WantedBy=multi-user.target
  1. Reload the systemd daemon : systemctl daemon-reload
  2. Enable the service : systemctl enable vdi-api.service
  3. Start the service : systemctl start vdi-apî.service
  4. Check the status of the service : systemctl status vdi-api.service

Scheduler

  1. Modify the file scheduler.py to set the environment variables :
  • DB_HOST : Host of the database (ex: vdi-db)
  • DB_NAME : Name of the database (ex: vdi)
  • DB_USER : User of the database (ex: vdi)
  • DB_PASSWORD : Password of the database (ex: azerty)
  1. Modify the file back_openstack.py to set the environment variables :
  • PRIVATE_NETWORK_ID : ID of the private network
  • USERNAME : Username of the OpenStack user (ex: vdi)
  • PASSWORD : Password of the OpenStack user (ex: vdi)
  • AUTH_URL : URL of the OpenStack authentication (ex: http://172.10.3.60:5000/v3)
  • PROJECT_NAME : Name of the OpenStack project (ex: vdi)
  • USER_DOMAIN_ID : ID of the OpenStack user domain (ex: default)
  • PROJECT_DOMAIN_ID : ID of the OpenStack project domain (ex: default)
  1. Create a file /etc/systemd/system/vdi-scheduler.service with the following content : 
    [Unit]
Description=Python VDI Scheduler Service
After=network.target

[Service]
User=vdi
Group=vdi
ExecStart=/usr/bin/python3 /path/to/the/folder/scheduler.py
Restart=always

[Install]
WantedBy=multi-user.target
  2. Reload the systemd daemon : systemctl daemon-reload
  3. Enable the service : systemctl enable vdi-scheduler.service
  4. Start the service : systemctl start vdi-scheduler.service
  5. Check the status of the service : systemctl status vdi-scheduler.service

APP

Github Repository

  1. Change the app.config['URL_API'] variable to the API domain (ex: https://api.insa-cvl.com) in the app.py file.
  2. Create a file /etc/systemd/system/vdi-app.service with the following content : 
    [Unit]
Description=Gunicorn instance to serve APP VDI
After=network.target

[Service]
User=vdi
Group=vdi
WorkingDirectory=/home/vdi/VDI-APP
Environment="PATH=/home/vdi/.local/bin"
ExecStart=/home/vdi/.local/bin/gunicorn --access-logfile /var/log/VDI/APP/access.log --error-logfile /var/log/VDI/APP/error.log --workers 3 --bind 0.0.0.0:5000 app:app

[Install]
WantedBy=multi-user.target
  3. Reload the systemd daemon : systemctl daemon-reload
  4. Enable the service : systemctl enable vdi-app.service
  5. Start the service : systemctl start vdi-app.service
  6. Check the status of the service : systemctl status vdi-app.service

Installation with Docker

You can use the docker-compose.all.yml file to run the API, the scheduler and the frontend. If you want to run the API, the scheduler and the frontend separately, you can use the docker-compose.yml file for each of them in the Docker folder and on the Github repository of the frontend.

Database

We recommend to use a MariaDB or MySQL database. You can find an example of a docker-compose.yml file to run a MariaDB database below.

  vdi-db:
    image: mariadb:10.11.4
    container_name: vdi-db
    restart: always
    environment:
      - MYSQL_ROOT_PASSWORD=azerty
      - MYSQL_DATABASE=vdi
      - MYSQL_USER=vdi
      - MYSQL_PASSWORD=azerty
    volumes:
      - ./vdi-db:/var/lib/mysql

API

  1. build the image : docker build -t vdi-api . or docker-compose build with the docker-compose.yml file
  2. Change the environment variables in the docker-compose.yml file
  • URL_VNC : URL of the VNC server (ex: https://vnc.insa-cvl.com)
  • URL_VDI : URL of the VDI server (ex: https://vdi.insa-cvl.com)
  • URL_API : URL of the API server (ex: https://api.insa-cvl.com)
  • URL_API_CAS : URL of the API without https:// (ex: api.insa-cvl.com)
  • DB_HOST : Host of the database (ex: vdi-db)
  • DB_NAME : Name of the database (ex: vdi)
  • DB_USER : User of the database (ex: vdi)
  • DB_PASSWORD : Password of the database (ex: azerty)
  • OPENSTACK_PRIVATE_NETWORK_ID : ID of the private network
  • OPENSTACK_USERNAME : Username of the OpenStack user (ex: vdi)
  • OPENSTACK_PASSWORD : Password of the OpenStack user (ex: vdi)
  • OPENSTACK_AUTH_URL : URL of the OpenStack authentication (ex: http://172.10.3.60:5000/v3)
  • OPENSTACK_PROJECT_NAME : Name of the OpenStack project (ex: vdi)
  • OPENSTACK_USER_DOMAIN_ID : ID of the OpenStack user domain (ex: default)
  • OPENSTACK_PROJECT_DOMAIN_ID : ID of the OpenStack project domain (ex: default)
  1. run the container : docker-compose up -d with the docker-compose.yml file
version: "3.9"
services:
  vdi-api:
    # build: .
    image: vdi-api:latest
    container_name: vdi-api
    restart: always
    ports:
      - 127.0.0.1:5001:5001
    environment:
      - URL_VNC=https://vnc.insa-cvl.com
      - URL_VDI=https://vdi.insa-cvl.com
      - URL_API=https://api.insa-cvl.com
      - DB_HOST=vdi-db
      - DB_NAME=vdi
      - DB_USER=vdi
      - DB_PASSWORD=azerty
      - OPENSTACK_PRIVATE_NETWORK_ID=0d49c37b-7077-4152-985c-f5a00ad20677
      - OPENSTACK_USERNAME=admin
      - OPENSTACK_PASSWORD=3pMrmW899b9y^2kiJa!6#Z#kE%@a2r
      - OPENSTACK_AUTH_URL=http://172.10.3.60:5000/v3
      - OPENSTACK_PROJECT_NAME=admin
      - OPENSTACK_USER_DOMAIN_ID=default
      - OPENSTACK_PROJECT_DOMAIN_ID=default
    depends_on:
      - vdi-db
    links:
      - vdi-db
  1. check the logs : docker logs vdi-api or docker-compose logs -f

Scheduler

  1. Build the image : docker build -t vdi-scheduler . or docker-compose build with the docker-compose.yml file
  2. Change the environment variables in the docker-compose.yml file
  • DB_HOST : Host of the database (ex: vdi-db)
  • DB_NAME : Name of the database (ex: vdi)
  • DB_USER : User of the database (ex: vdi)
  • DB_PASSWORD : Password of the database (ex: azerty)
  • OPENSTACK_PRIVATE_NETWORK_ID : ID of the private network
  • OPENSTACK_USERNAME : Username of the OpenStack user (ex: vdi)
  • OPENSTACK_PASSWORD : Password of the OpenStack user (ex: vdi)
  • OPENSTACK_AUTH_URL : URL of the OpenStack authentication (ex: http://172.10.3.60:5000/v3)
  • OPENSTACK_PROJECT_NAME : Name of the OpenStack project (ex: vdi)
  • OPENSTACK_USER_DOMAIN_ID : ID of the OpenStack user domain (ex: default)
  • OPENSTACK_PROJECT_DOMAIN_ID : ID of the OpenStack project domain (ex: default)
  1. Run the container : docker-compose up -d with the docker-compose.yml file
version: "3.9"
services:
  vdi-scheduler:
    # build: .
    image: vdi-scheduler:latest
    container_name: vdi-scheduler
    restart: always
    environment:
      - DB_HOST=vdi-db
      - DB_NAME=vdi
      - DB_USER=vdi
      - DB_PASSWORD=azerty
      - OPENSTACK_PRIVATE_NETWORK_ID=0d49c37b-7077-4152-985c-f5a00ad20677
      - OPENSTACK_USERNAME=admin
      - OPENSTACK_PASSWORD=3pMrmW899b9y^2kiJa!6#Z#kE%@a2r
      - OPENSTACK_AUTH_URL=http://172.10.3.60:5000/v3
      - OPENSTACK_PROJECT_NAME=admin
      - OPENSTACK_USER_DOMAIN_ID=default
      - OPENSTACK_PROJECT_DOMAIN_ID=default
    depends_on:
      - vdi-db
    links:
      - vdi-db
  1. check the logs : docker logs vdi-scheduler or docker-compose logs -f

APP

Github Repository

  1. Build the image : docker build -t vdi-app . or docker-compose build with the docker-compose.yml file.
  2. Change the environment variable URL_API to the API domain (ex: https://api.insa-cvl.com) in the docker-compose.yml file.
  3. Run the container : docker-compose up -d with the docker-compose.yml file : 
    version: "3.9"
services:
vdi-app:
    image: vdi-app:latest
    container_name: vdi-app
    ports:
        - "127.0.0.1:5000:5000"
    environment:
        - URL_API=https://api.insa-cvl.com
    restart: always
  4. Check the logs : docker logs vdi-app or docker-compose logs -f

Nginx Configuration

You can use Nginx to serve the API, the scheduler and the frontend. You can find all the configuration files in the Nginx folder.