README
¶
Overview
For development and testing purposes, it's useful to be able to deploy the chain remotely where the validators are running on cloud infra. This directory contains setup scripts and Dockerfile that can be used to build Docker container images for running validator nodes.
Overall, the steps involve:
- Define validator node keys
- Build the app binary (i.e.
jinxprotocold) - Set up configuration files (i.e.
genesis.json,app.toml,app.toml, and etc.) - Setup the app data and configs (i.e. moving data and configs files into correct directory)
Currently, step 1 is a manual process and steps 2, 3 and 4 are automated within the scripts (local.sh, dev.sh, and staging.sh).
TODO(CORE-512): add info/resources around cloud setup. Doc
[Manual] Add new validator node keys to dev.sh or staging.sh
In order to add a new validator node to dev.sh or staging.sh, simply run make localnet-new-validator and copy the mnemonic, address, node_key, node_id in the their respective fields in dev.sh or staging.sh, and add a new moniker for this validator.
Running the local chain
The local chain can be run using the docker-compose.yml file at the root of this repository. Simply run make localnet-start to start the validator and make localnet-stop to stop it.
Running the dev or staging chains locally
You can run the testnet chain locally by running the following command from the v4 repository root.
It's necessary to specify the --home flag as this is how the container knows which validator to run as. A new home directory is created for each MONIKER defined in dev.sh or staging.sh. There is also an additional home directory for running the chain as a full-node (see instructions below).
# dev
docker build . --progress=plain --no-cache -f ./testing/testnet-dev/Dockerfile -t testnet && docker run testnet start --home /jinxprotocol/chain/.alice
# staging
docker build . --progress=plain --no-cache -f ./testing/testnet-staging/Dockerfile -t testnet && docker run testnet start --home /jinxprotocol/chain/.alice
Building and Pushing the Docker container image to ECR
Currently the staging container is pushed automatically to ECR when new changes are pushed to the main branch, however you can also manually deploy the container.
Run the ./build-push-ecr.sh script in this directory from this repository's root directory. You can specify either the "dev", "dev2", "dev3", "dev4", "dev5", or "staging" environment.
Ensure that your AWS_PROFILE is correct for the environment you want to deploy to (this can be found in ~/.aws/credentials), and can be selected using the AWS_PROFILE env var.
example: AWS_PROFILE=jinx-v4 ./scripts/build-push-ecr.sh staging
Opening a shell on Docker container images
It's useful to open up a shell on Docker container images to inspect or interact. Run the following command to do so:
$ docker run -it --entrypoint /bin/sh <image id>
Running a full-node
If you wish to run as a full-node instead of a validator, specify the --home flag as /jinxprotocol/chain/.full-node.
$ docker build . --progress=plain --no-cache -f ./testing/testnet-dev/Dockerfile -t testnet && docker run testnet start --home /jinxprotocol/chain/.full-node
Directories