Importing the rainbow tables

You will need an already initialized graph-node database. You will also need to be able to run psql and connect to the graph-node database. If your database is sharded, this data needs to be imported into the primary database

Download the SQL dump file from Goolge cloud storage:

• Compressed with gzip (5.9 GB): ens_names.sql.gz (sha256)
• Compressed with zstd (5.6 GB): ens_names.sql.zst (sha256)

Import the dump:

The psql command for the import has to be run as the same user as the user that graph-node uses to connect to the database.

zcat ens_names.sql.gz | psql graph

or

zstdcat ens_names.sql.zst | psql graph

Data preparation

These are the steps to generate the ens_names.sql.gz file, and are not needed for just importing that data.

Convert ENS's rainbow table data into a SQL script that we can ingest with psql, similar to a plain text dump from pg_dump.

Download input for rainbow tables via gsutil cp gs://ens-files/preimages/* .

Run this as cat preimages-* | cargo run --release | gzip > data.sql.

Takes a while (10 minutes) on my machine and results in a 6GB file (133M entities)

Exporting the prepared data:

pg_dump -c -x -O --if-exists --no-tablespaces -t ens_names -f /var/tmp/ens_names.sql.gz -Z9