This repository contains the code and configuration files of the metadata oracle for Cardano.
Key Features • Installation • Usage • Configuration
- Data source definition templates that are easy to share within the community
- Fetching and scraping data with error handling and retries
- Transaction construction, including dynamic fee estimation
- Support for remote API Blockfrost.io
- Support for full local Cardano node (using
cardano-cli) - Prometheus style endpoint for monitoring and Grafana integration
If you are not running NixOS, you need to at least install Nix
$ curl -L https://nixos.org/nix/install | shAfter installing Nix, add the following to /etc/nix/nix.conf to take advantage of the cache, so you do not have to rebuild everything. After adding it, you need to restart the nix-daemon service.
substituters = https://cache.nixos.org
trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY=If you are using NixOS, you probably know how to add these as your binary caches.
First, build the docker container and then run this within the repository:
$ docker build . -t cardano-metadata-oracleAfter building your container, just run it:
$ docker run cardano-metadata-oracle --helpWe're also hosting the image directly on Docker Hub. Run it direclty by using:
$ docker run fivebinaries/cardano-metadata-oracle --help
Unable to find image 'fivebinaries/cardano-metadata-oracle:latest' locally
latest: Pulling from fivebinaries/cardano-metadata-oracle
USAGE
$ cardano-metadata-oracleFor lightweight Cardano metadata node, you need to use a remote API such as Blockfrost.io. Go to their website and retrieve a project_id for the specific network you want to use (Mainnet or Testnet).
export BLOCKFROST_PROJECT_ID=your_blockfrost_project_id
To post your datapoint to the Cardano network, you need to provide at least origin-file and seed-file of the wallet containing the funds to do posting. You can use files in the test folder.
When using the Cardano full node, make sure you set the environmental variable CARDANO_NODE_SOCKET_PATH to the socket file of the node.
The metadata oracle node needs a list of origins from where to fetch the data. The source of this file is specified using--seed-file. It is a simple YAML file, using the ticker as the top element with a list sources for given ticker. In the example below, we define two sources for ticker ADAUSD. The path proprieties specifies the position within the json document.
There are many possible expressions you can use in path to extract the data you need. For full syntax and more examples head to jsonpath documentation.
ADAUSD:
- name: coinGecko
url: https://api.coingecko.com/api/v3/coins/cardano?localization=false&tickers=false&market_data=true
path: '$.market_data.current_price.usd'
abort_on_failure: true
- name: cryptoCompare
url: https://min-api.cryptocompare.com/data/pricemulti?fsyms=ADA&tsyms=USDD&api_key=6e49d40fcabeb5e
path: '$.ADA.USD'
Have a look in the test/data/ directory for inspiration.
The Cardano metadata node by default tries to sign and post the transaction to the blockchain. In order to do this, we need a wallet with funds. This is being defined by the --seed-file switch pointing to the ED25519-BIP32 seed word phrase.
Keep sane permissions on this file as it contains sensitive information. Also, it is best practice to keep only a small amount of funds on this wallet and top it up from time to time.
If you want to construct only a transaction and sign it offline, use the --address switch.
By default, the node uses Cardano mainnet network. To post to testnet network, use the network=testnet switch.
By default, we're using matadata label 1968 as per nut.link specification. If you have a specific use case, it might be a good idea to use a different label.
You have the possibility to export prometheus-style file with the current address balance, the time of last run and the execution time. This will help you a lot with monitoring of your oracle.