NEAR Protocol is a decentralized application platform that is secure enough to manage high value assets like money or identity and performant enough to make them useful for everyday people, putting the power of the Open Web in their hands.

Terraform is a tool by Hashicorp that allows developers to treat "infrastructure as code", easying the management and repeatibility of the infrastructure. Infrastructure and all kind of cloud resources are defined in modules, and Terraform creates/changes/destroys when changes are applied.

Inside the near folder you will find a module (and submodules) to create the setup for running a NEAR Validator on AWS. The next logic resources can be created:

• vpc module for setting up a VPC with a public and private subnet. bastion and validator nodes go in the public subnet(a private subnet is available to eventually allow for the validator and for sentry nodes in the public subnet).
• bastion module for an SSH bastion node. For security purposes, this is the only node that accepts external SSH traffic. All other nodes only accept SSH from the bastion.
• validator module for creating a Proxy connected to a validator.

All nodes run on the Ubuntu LTS 18.04 AMI. Running terraform apply will select the latest available AMI.

Near nodes will get provisioned with some recommended security settings from Celo cLab's Security Audit team. Most of this hardening is done in install-base.sh and final-hardening.sh.

Inside the example folder you can find an example tf to use the module. Use that tf as base file for your deployment, modifying the account variables used for your convenience. Alternatively you can take that tf files as base for customizing your deployment. Please take care specially about the VPC network configuration.

This guide assumes you already have a validator setup with an account and a staking pool contract deployed. If not, please visit NEAR DOCS for more information before continuing

git clone https://github.com/abellinii/near-terraform.git

cd near-terraform/example; cp example.secret.auto.tfvars secret.auto.tfvars

Populate the secret.auto.tfvars file with your near configuration details.

If it the the first time on the network and you do not have keys or a pool set initial_startup = true and ssh in to create your stake pool. Testnet documentation can be found here

This infracstructure setup is using AWS. You will need an AWS account and to have the AWS CLI installed and logged in to use this package. Information can be found here

Terraform init

Terraform Apply

To access the node you will first need to access the bastion as it is the only IP you shh into the validator. To do that:

ssh-add -K /path/to/your/key.pem

ssh -A ubuntu@<BastionIP>

ssh ubuntu@<validatorIP>

Nearup is NEAR's public scripts to launch near betanet and testnet node. This is installed on provisioning.

NEAR Shell is a fully featured NodeJS command line application that wraps near-api-js

Prometheus is installed on startup to expose metrics for the NEAR validating node and the system. Grafana is installed to help vsualize these metrics with a dashboard that is customized to show relevant metrics for NEAR validating nodes. The majority is configured and can be accessed by :3000 and default username and password is admin admin.

This was Near's Stake Wars Challenge 3 and a few great tutorials in various languages can be found here

Create's a warchest of staked tokens, and dynamically maintains no more than one seat. This is designed to monitor the minimum stake to become a validator, and dynamically manage your staking pool.

This was Near's Stake Wars Challenge 4 and utilizes a package built by eorituz(A Near community member) and more information can be found here

Note: Currently the user would have to ssh into the validator machine and login to NEAR near login for the Warchest bot to operate correctly

A CI Workflow that will check the latest release of nearcore and will build from source, test and create a docker image in a docker repository you create. Props go to masknetgoal634 for his implementation here.

Add these 3 secrets to your Gthub secrets(for a detailed look at whats happening please view the github link to the package):

DOCKER_IMAGE_NAME
DOCKER_USERNAME
DOCKER_PASSWORD