Mina Indexer

The Mina Indexer is a redesigned version of the software collectively called the "Mina archive node."

Note: As the project is in active development, be aware that the public APIs and functionalities are subject to change.

Motivation

Working with the Mina Archive Node can be complex and time-consuming. It requires the exact configuration of multiple components — a Mina Node, an Archive Node, and a Postgres database — and an in-depth knowledge of SQL and the Mina blockchain.

Additionally, even with a proper setup, the Archive Node system is prone to missing blocks, creating gaps in the data. These gaps require manual intervention for resolution, which adds layers of complexity to the system's management.

A major problem with the Archive Node is its reliance on a pg_dump from a currently active node for initial setup. This approach centralizes data, necessitating trust from the operator's side.

Solution

The Mina Indexer addresses this by simplifying the initial configuration by using precomputed blocks as the source of truth, bypassing the need for external database dumps.

We designed the Mina Indexer to be a comprehensive replacement for the Mina Archive Node, providing an easier-to-use platform with native support for the Rosetta API. We aim to streamline blockchain interaction for developers within the Mina ecosystem by providing developers and operators with a better toolset optimized for the Mina ecosystem.

Getting Started

Development Prerequisites

1. Install Nix (with Flakes) using The Determinate Nix Installer.
2. Install and configure Direnv.

Execution Environment

Invoking 'mina-indexer database create' on very large blocks directories will ingest the blocks at very high speed. This requires setting ulimit -n to a larger-than-default value (for many Linux distros). 1024 is not sufficient. See your distro's documentation on how to set ulimit -n (max open files) to 4096 or more.

Normal operation of the 'mina-indexer' does not require this change.

Building the Project

Run just check to verify that the project compiles. This will compile the mina-indexer binary in debug mode.

Storage

The default storage location is on /mnt because the testing code may download large volumes of test data, and placing on /mnt gives an opportunity to use different storage volumes from one's build directory.

Set the VOLUMES_DIR environment variable if you want to replace /mnt with another path.

Testing

Unit Tests

Execute unit tests to validate code functionality with:

just test-unit

Regression Tests

To invoke the regression test suite, the directory /mnt/mina-indexer-dev must exist.

To quickly perform regression tests, which check for new bugs in existing features after updates, use:

just bt

To perform the unit tests in addition to that, use:

just tt

To perform the test battery that the (tier-1) CI runs, use:

just tier1

More Tests

To invoke a more comprehensive regression test suite, the directory /mnt/mina-indexer-test must exist.

Invoke:

just tier2

Or, for even more testing:

just tier3

Deployment

To deploy a mina-indexer locally, the directory /mnt/mina-indexer-prod must exist.

Run just deploy-local-prod to start an instance.

Generating OCI Images With Nix

Building the OCI (Docker) image from Nix must happen from an x86-64-linux machine.

Issue the following command to build the image and load it into Docker.

just build-image

License

Copyright 2022-2024 Granola Systems Inc.

This software is licensed under the Apache License, Version 2.0.

Contributing

This project uses C4(Collective Code Construction Contract) process for contributions.