Installation

If you want to run an instance of the indexer, first make sure to have Node.js in version 17 or higher, as well as docker installed. Pull the latest code from the main branch on GitHub (we'll soon be starting to version releases).

git clone git@github.com:teztok/indexer.git

Running the indexer in dev/local mode

Install the Node.js dependencies by running:

npm install

Create a `.env` file by making a copy of `.env.sample` and modifying it to your needs.

cp .env.sample .env

Start Postgres and Hasura by running:

docker-compose up -d

Now run this command to create the database schema:

npm run init:db

To start the indexer run:

npm start

It will start indexing from block 1365000 (end of February 2021).

Build and run the indexer through docker-compose:

Create a

.env.prod

file by making a copy of `.env.sample` and modifying it to your needs.

cp .env.sample .env.prod

Now create a custom docker-compose file by copying `docker-compose.prod.yml` to `docker-compose.custom.yml` and modifying it to your needs.

cp docker-compose.prod.yml docker-compose.custom.yml

It would take a very long time to index everything from scratch, which is why we strongly recommend downloading and importing a recent database dump from https://backups.teztok.com/daily/. This dump also contains the Hasura configuration and data from tzprofiles.

Here is an example of how such an SQL dump can be imported into a docker container running Postgres.

# Start Postgres
docker-compose -f docker-compose.custom.yml up -d teztok_database

# Import a recent dump (can take quite some time)
gunzip < teztok-20220520-000000.sql.gz | docker exec -i teztok_database psql -U teztok -d teztok

# Stop Postgres
docker-compose -f docker-compose.custom.yml down

Now build the docker image:

docker-compose -f docker-compose.custom.yml build

And then start the indexer:

docker-compose -f docker-compose.custom.yml up

It's recommended to download and import an SQL dump from our backup server every once in a while in order to get fixes. This way you also don't need to bother with re-indexing.