How to install, run and hack DAppBoard.
Table of content:
The DAppBoard project is a suite of open source tools used to capture and analyze Ethereum blockchain data. It is o into 4 repositories:
- ETL: The tools that extract and transform the blockchain data and inject into our database.
- Web: A backend and frontend based on SailsJS. This is what you see when you visit dappboard.com.
- Environment: Contains script and environment variable you will need to run DAppBoard ETL and Web.
- Documentation: How to set up and run your own DAppBoard.
Running DAppBoard requires few resources
- Ubuntu 18.04 server: any other distribution should work but this is what we use.
- PostgreSQL database: or any compatible system like Amazon Aurora.
We are currently hosting our pipeline at DigitalOcean and would like to thanks them for their support.
On a fresh Ubuntu server, clone the environment installation script and run it as an administrator. This will install all the tools needed to run the ETL and Web interface.
cd ~ && git clone https://github.com/DAppBoard/dappboard-environment.git && cd dappboard-environment && sudo ./install_server.sh
You will then need to fill the environment file, the sample one is available here dappboard-environment/env_sample:
# Database connection information
export DAPPBOARD_PSQL_HOST=""
export DAPPBOARD_PSQL_DB=""
export DAPPBOARD_PSQL_USER=""
export DAPPBOARD_PSQL_PASSWORD=""
# The connection string for your RPC node or Infura
export DAPPBOARD_NODE_URL=""
# Only necessary for scripts that are fetching data from Etherscan (ABI/Sourcecode)
export DAPPBOARD_ETHERSCAN_API=""
Once you have filled the information, source it and add it to your general environment.
source env_sample && sudo cat env_sample >> /etc/environment
For installing the ETL, we just need to clone the repo and install NPM dependencies.
cd ~ && cd dappboard-etl/etl && npm install
You will then need to create all the tables and indexes. For this, connect to your PostgreSQL database and run the SQL queries located in the dappboard-etl/schemas/tables/ folder or here.
Installing the Web part is optional as it's only valuable if you want to replicate DAppBoard. We publish it for learning and transparency purpose.
cd ~ && cd dappboard-web/ && npm install
We use PM2 in order to run and monitor node script: it was installed with the server environment. In the home folder run:
pm2 start etl.ecosystem.config.js
This will start all the ETL scripts:
- live is responsible for getting new blocks and ingesting them.
- past is responsible for going back in time for ingesting the blocks.
- scrape_tokens is running every hour and get information about the tokens that emitted Transfer events.
In the web folder, run:
sails lift
Everything you see (sources, documentation) is published under the MIT License.
We invite you to use our work and contribute. If you use our tools or some of our data, feel free to tell it to the world!