Blob Archiver

The Blob Archiver is a service to archive and allow querying of all historical blobs from the beacon chain. It consists of two components:

• Archiver - Tracks the beacon chain and writes blobs to a storage backend
• API - Implements the blob sidecars API, which allows clients to retrieve blobs from the storage backend

Storage

There are currently two supported storage options:

• On-disk storage - Blobs are written to disk in a directory
• S3 storage - Blobs are written to an S3 bucket (or compatible service)

You can control which storage backend is used by setting the BLOB_API_DATA_STORE and BLOB_ARCHIVER_DATA_STORE to either file or s3.

The s3 backend will also work with (for example) Google Cloud Storage buckets (instructions here).

Data Validity

Currently, the archiver and api do not validate the beacon node's data. Therefore, it's important to either trust the Beacon node, or validate the data in the client. There is an open issue to add data validation to the archiver and api.

Development

The Makefile contains a number of commands for development:

# Run the tests
make test
# Run the integration tests (will start a local S3 bucket)
make integration 

# Lint the project
make lint

# Build the project
make build

# Check all tests, formatting, building
make check

Run Locally

To run the project locally, you should first copy .env.template to .env and then modify the environment variables to your beacon client and storage backend of choice. Then you can run the project with:

docker compose up

You can see a full list of configuration options by running:

# API
go run api/cmd/main.go

# Archiver
go run archiver/cmd/main.go