Blockfrost.io backend service

Blockfrost.io backend is an API service providing abstraction between you and Cardano blockchain data, taking away the burden of complexity, so you can focus on what really matters - developing your applications.

You can now Run-Your-Own.

Getting started • Running your own • Developing • 📌 Roadmap

Getting started

The backend is Node.js app written in Typescript using Fastify. To run it you need Node.js version 16 and higher (LTS is highly recommended). Blockchain data are queried from cardano-db-sync. Follow their documentation to learn more about running your own instance.

Running your own

We made it simple to run your own version of the Blockfrost backend.

System requirements

The system requirements for the basic Blockfrost backend stack (which means blockfrost-backend, cardano-node, cardano-db-sync and postgresql on the same machine) are:

64 GB of RAM or more
8 CPU cores or more
250 GB of disk space or more
SSD disk with at least 80k IOPS (measured as chunk size)

Configuration

There are several configuration files in config directory. Config file is picked based on a value in an environment variable NODE_ENV (value set in NODE_ENV must match the name of the config file). This environment variable is set automatically while running the backend via prepared yarn scripts.

If you are using an authenticated db connection that requires a password, you'd need to provide a PGPASSWORD environment variable and the postgres library will automatically pick it up.

Schema

{
  // Blockfrost backend settings
  server: {
    // Server listen address, you need to set this to 0.0.0.0 if running within docker
    listenAddress: 'localhost',
    // Server port
    port: 3000,
    // Whether to enable verbose logging, when disabled only ERRORs are printed to a console
    debug: true,
    // Whether to expose /prometheus endpoint
    prometheusMetrics: false,
  },
  // Cardano DB Sync SQL connection
  dbSync: {
    host: 'cdbsync-dev.mydomain.com',
    user: 'username',
    database: 'password',
  },
  // Cardano network - mainnet, testnet, preview, preprod
  network: 'mainnet',
  // Path to token registry directory (see next section for more details)
  tokenRegistryUrl: 'https://tokens.cardano.org',
}

💡 All config variables can be also set via environment variables which take precedence over values from a config file.

These values are BLOCKFROST_CONFIG_SERVER_PORT, BLOCKFROST_CONFIG_SERVER_DEBUG, BLOCKFROST_CONFIG_SERVER_PROMETHEUS_METRICS, BLOCKFROST_CONFIG_DBSYNC_HOST, BLOCKFROST_CONFIG_DBSYNC_USER, BLOCKFROST_CONFIG_DBSYNC_DATABASE, BLOCKFROST_CONFIG_DBSYNC_MAX_CONN, BLOCKFROST_CONFIG_NETWORK, BLOCKFROST_CONFIG_TOKEN_REGISTRY_URL.

Token registry

Blockfrost Backend uses token registry to provide off-chain metadata for native assets (eg. number of decimals). The token registry operated by Cardano Foundation and hosted at https://tokens.cardano.org is used by default. You can use self-hosted registry by setting tokenRegistryUrl in the config file.

Indices

Although it is possible to run RYO with vanilla db sync, Blockfrost usually queries hex values in the already encoded form.

Therefore, in order to speed up queries, it is recommended to create the following, custom, indices:

CREATE INDEX IF NOT EXISTS bf_idx_block_hash_encoded ON block USING HASH (encode(hash, 'hex'));
CREATE INDEX IF NOT EXISTS bf_idx_datum_hash ON datum USING HASH (encode(hash, 'hex'));
CREATE INDEX IF NOT EXISTS bf_idx_multi_asset_policy ON multi_asset USING HASH (encode(policy, 'hex'));
CREATE INDEX IF NOT EXISTS bf_idx_multi_asset_policy_name ON multi_asset USING HASH ((encode(policy, 'hex') || encode(name, 'hex')));
CREATE INDEX IF NOT EXISTS bf_idx_pool_hash_view ON pool_hash USING HASH (view);
CREATE INDEX IF NOT EXISTS bf_idx_redeemer_data_hash ON redeemer_data USING HASH (encode(hash, 'hex'));
CREATE INDEX IF NOT EXISTS bf_idx_scripts_hash ON script USING HASH (encode(hash, 'hex'));
CREATE INDEX IF NOT EXISTS bf_idx_tx_hash ON tx USING HASH (encode(hash, 'hex'));
CREATE UNIQUE INDEX IF NOT EXISTS bf_u_idx_epoch_stake_epoch_and_id ON epoch_stake (epoch_no, id);
CREATE INDEX IF NOT EXISTS bf_idx_reference_tx_in_tx_in_id ON reference_tx_in (tx_in_id);
CREATE INDEX IF NOT EXISTS bf_idx_collateral_tx_in_tx_in_id ON collateral_tx_in (tx_in_id);

Docker

We are hosting latest release of this software on Dockerhub. To run it using Docker:

docker run --rm \
  --name blockfrost-ryo \
  -p 3000:3000 \
  -e BLOCKFROST_CONFIG_SERVER_LISTEN_ADDRESS=0.0.0.0 \
  -v $PWD/config:/app/config \
  blockfrost/backend-ryo:latest

You can also generate a Docker image using Nix instead of the Dockerfile running

nix build .#dockerImage

Nix

To start the Blockfrost backend under nix, just run:

$(nix-build -A blockfrost-backend-ryo --no-out-link)/bin/blockfrost-backend-ryo

Or, since this repository is also a Nix flake, you can also run

nix run # inside this repo folder
# Otherwise, without manually cloning the repo before:
# nix run github:blockfrost/blockfrost-backend-ryo

Developing

This is an open-source project and anyone is welcome to contribute, please see CONTRIBUTING for more information.

garyblankenship/blockfrost-backend-ryo