This repository is a DevOps upgraded version for the successful freqtrade bot. Including data collection & visualization, using Elasticsearch, logstash & Kibana

In order to run the freqtrade-buhtik, you'll have to have these requirements installed:

• Docker
• Docker-compose

What are the changes from the normal Freqtrade project? We've added:

• Data handler container (written in Python) - getting relevant data from the freqtrade API and saves it in a log file. Information regarding the data is detailed below
• A data collector & analyze - Logstash & Elasticsearch in 2 different containers
• Visualization - using Kibana, you can take all that data and make dashboards from them

The data collected from the freqtrade API and delivered to Elasticsearch & Kibana includes:

• Current BTC in account
• Current value in dollars
• Current best pair of trading
• Average duration for trades
• Latest trade done (date)
• Total trades done
• Current open trades

In order to allow full control for the bot & the ELK (Elasticsearch, Logstash, Kibana) services,you should inject the credentials to the containers. The best way to do so is to create a credentials file called .env (include it in .gitignore) so git won't save it. Set the following environment variables inside this file:

• BINANCE_API_KEY
• BINANCE_API_SECRET_KEY
• FREQ_API_PWD
• ELASTIC_PASSWORD

The user for logging in to the Kibana GUI: elastic

These are the default configurations I put for the trading freqtrade bot, you can change it on the config.json:

• stake_currency: BTC
• stake_amount: unlimited
• tradable_balance_ratio: 1

The coins pairs that are allowed to be trade by freqtrade are:

• EOS/BTC
• ETH/BTC
• LINK/BTC
• NEO/BTC
• XRP/BTC
• XTZ/BTC
• MATIC/BTC
• VET/BTC
• ADA/BTC
• BAND/BTC
• DOGE/BTC
• XLM/BTC
• ATOM/BTC
• LEND/BTC
• ERD/BTC
• ZIL/BTC
• ONT/BTC
• ANKR/BTC
• SOL/BTC
• COMP/BTC
• STORJ/BTC
• KAVA/BTC

Firstly, build the docker images locally by:

docker-compose build

Secondly, run the docker-compose:

docker-compose up