Skip to content

Latest commit

 

History

History

indexer

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
resources
resources
 
 
src
src
 
 
CHANGELOG.md
CHANGELOG.md
 
 
Cargo.lock
Cargo.lock
 
 
Cargo.toml
Cargo.toml
 
 
README.md
README.md
 
 

README.md

Track and trace indexer

There are two binaries in this project. An indexer that indexes data into a database and a server that serves data from the database.

The easiest way to run the indexer and server is to use docker-compose as described in the Track and Trace project's main README.md file.

Prerequisites

  • PostgreSQL installed or running it in a docker container: https://www.postgresql.org/download/
  • A database connection to PostgreSQL: If you use the default connection host=localhost dbname=indexer user=postgres password=password port=5432, make sure you have created the database indexer as follows:

Open a terminal-based front-end to PostgreSQL:

sudo -u postgres psql

Create the indexer database:

CREATE DATABASE indexer;

Alternatively, you can run the Postgres database in a docker container. The command below will create an indexer db automatically:

docker run -p 5432:5432 -e POSTGRES_PASSWORD=password -e POSTGRES_DB="indexer" --rm postgres

Build the indexer and server

To build the tools make sure you have the repository submodules initialized

git submodule update --init --recursive

The tool can be built by running

cargo build --release

This will produce two binaries (indexer and server) in the target/release directory.

The indexer binary

It is a tool for indexing event data from the track and trace contract into a postgres database. The database is configured with the tables from the file ../resources/schema.sql. The monitored events ItemStatusChangedEvent and ItemCreatedEvent are indexed in their respective tables. A third table settings exists to store global configurations (e.g.: the contract address, latest block processed, and the genesis block hash).

The global configurations are set when the indexer is started for the first time. Re-starting the indexer will check if its current settings are compatible will the stored indexer settings to prevent corrupting the database. In addition, the settings can be queried by the front end to check compatibility.

When the indexer is started for the first time, it will look up when the smart contract instance was created and use that block as the starting block. When the indexer is re-started with the same database settings, it resumes indexing from the latest_processed_block_height+1 as stored in the database.

All monitored events in a block are atomically added in one database transaction to postgres. This ensures a simple recovery process since we always process the complete block or roll back the database to the beginning of the block. In addition, the indexer has a re-try logic and will try to re-connect to the database pool and re-submit any failed database transaction.

Each event can be uniquely identified by the transaction_hash and event_index. The event_index is the index from the array of logged events in a transaction.

Run the indexer

cargo run --bin indexer -- --node https://grpc.testnet.concordium.com:20000 --contract "<8901,0>" --log-level debug

Configure the indexer

There are a few options to configure the indexer:

  • --node is the endpoint to the Concordium node grpc v2 API. If not specified, the default value https://grpc.testnet.concordium.com:20000 is used.

  • --contract is the contract index of the track-and-trace smart contract, e.g. <8901,0>.

  • --db-connection should specify your postgreSQL database connection. If not specified, the default value host=localhost dbname=indexer user=postgres password=password port=5432 is used.

  • --log-level specifies the maximum log level. Possible values are: trace, debug, info, warn, and error. If not specified, the default value info is used.

The server binary

You have to build the front end in the folder ../frontend before running this command.

Run the server

cargo run --bin server

Configure the server

There are a few options to configure the server:

  • --listen-address is the listen address where the server will be listen on. If not specified, the default value 0.0.0.0:8080 is used.

  • --frontend is the path to the directory where the frontend assets are located. If not specified, the default value ../frontend/dist is used.

  • --db-connection should specify your postgreSQL database connection. If not specified, the default value host=localhost dbname=indexer user=postgres password=password port=5432 is used.

  • --log-level specifies the maximum log level. Possible values are: trace, debug, info, warn, and error. If not specified, the default value info is used.

The following option are also available, which are forwarded to the frontend:

  • --node specifies the gRPC interface of a Concordium node. (Defaults to https://grpc.testnet.concordium.com:20000)

  • --network specifies the network to use, i.e., mainnet or testnet. Defaults to testnet.

  • --contract-address specifies the contract address of the track and trace contract (format is <1234,0>).

  • --sponsored-transaction-backend specifies the endpoint to the sponsored transaction backend. (Defaults to http://localhost:8000).

An example of running the service with basic settings and testnet node would be:

cargo run --bin server  -- --contract-address <YOUR_CONTRACT_ADDRESS>

An example to run the service with some filled in example settings would be:

cargo run --bin server  -- --contract-address "<8901,0>"