Telegram bot that allows you to add and manage torrents on your computer, remotely.
This project is being developed using Node.js, Node telegram bot API and Node-transmission.
Using the latter to manage the torrents from the server application, answers petitions from any authorized Telegram client. (Would be really cool to develop an Electron client to connect to it too)
The main objective of the project besides the functionality itself is to learn the proper way to work using DevOps concepts and how to successfully deploy and manage the service running on a cloud server.
Using Azure as IaaS to deploy, Vagrant to create the virtual machine, Ansible and Fabric to setup and manage the service.
Create Azure certificate
Using the following orders:
openssl req -x509 -nodes -days 3650 -newkey rsa:2048 -keyout <certificate_name>.pem -out <certificate_name>.pem
openssl x509 -inform pem -in <certificate_name>.pem -outform der -out <certificate_name>.cer
chmod 600 <certificate_name>.pem
After the creation, we upload the .cer file to Azure and make sure that the .pem file is present on the directory from we're going to deploy the bot.
Deploy on Azure
To automate the deployment you can simply execute the deploy.sh script and let it install and configure the dependencies. To do so, Vagrant will follow the ansible file instructions and then the following fabric methods to finish node's installation and setup.
After the execution, use the following order to launch the service:
BOT_TOKEN=<your_bot_token> DB_PASSWORD=<db_password> DB_USER=<db_user>
NODE_ENV=production fab -p '<vm_login_password>' -H
<vm_user>@<azure_app_name>.cloudapp.net <start/start_forever>
All environment variables can be specified on the fabric order or defined on the OS (on .bashrc for example).
The start method will launch the bot on the current terminal, which will end when the window is closed. To launch a persistent service use the start_forever method.
To kill all active bot processes you can make use of stop_forever.
You can find more information about the vagrant configuration and fabric methods on the Vagrantfile and the Fabric file.
Teletorrent is be deployed on a cloud server on Heroku, and communicate with Telegram's bot API services to interact with the users. During development it's features are being tested using a CI tool, following TDD practices.
For CI I'm currently using Travis CI, and launching the tests with mocha (using chai and supertest packages). At the moment, only response tests are written. I'm trying to figure out how to properly emulate a conversation with a user to be able to test all functionallity in the most realistic approach.
Last update involved general improvements on code style, adding configuration files for deploying the application and writing a torrentAPI, wrapping all functionality related to communicating with the torrent client. Using node-transmission Teletorrent is able to manage remotely any instance of transmission running on another machine.
MongoDB is being used to store downloads information, and it's currently deployed on mLab.
- Fork this repository
- Create account on Heroku and mLab.
- On mLab create a database, along with its user and password.
- On Heroku create the application instance and add the deployment method to Github, using your fork of the project.
- Optional: Enable automatic deploys and waiting for CI (Travis) to pass before each deploy.
- In Heroku, go to Settings and add the following configuration variables:
BOT_TOKEN : <your_bot_token>DB_PASSWORD: <your_db_password>DB_USER: <your_db_user>NODE_ENV: 'production'(This may be automatically set, but just to be sure.)
Teletorrent should be deployed and ready work, just make sure to add to config/transmission.js your transmission client information to be able to access it.
First, you need to create a telegram bot using BotFather:
- Clone the repository
- Execute
npm install
- Execute
BOT_TOKEN=<your_bot_token> npm start - Connect to your bot from any telegram client
npm test
Go to the project's website for more info.