Skip to content

Latest commit

 

History

History
248 lines (178 loc) · 7.82 KB

deploy-testnet.md

File metadata and controls

248 lines (178 loc) · 7.82 KB

Deploy your own testnet

This document describes 3 ways to setup a network of terrad nodes, each serving a different usecase:

  1. Single-node, local, manual testnet
  2. Multi-node, local, automated testnet
  3. Multi-node, remote, automated testnet

Supporting code can be found in the networks directory and additionally the local or remote sub-directories.

NOTE: The remote network bootstrapping may be out of sync with the latest releases and is not to be relied upon.

Single-node, local, manual testnet

This guide helps you create a single validator node that runs a network locally for testing and other development related uses.

Requirements

Create genesis file and start the network

# You can run all of these commands from your home directory
cd $HOME

# Initialize the genesis.json file that will help you to bootstrap the network
terrad init --chain-id=testing testing

# Create a key to hold your validator account
terracli keys add validator

# Add that key into the genesis.app_state.accounts array in the genesis file
# NOTE: this command lets you set the number of coins. Make sure this account has some coins
# with the genesis.app_state.staking.params.bond_denom denom, the default is staking
terrad add-genesis-account $(terracli keys show validator -a) 1000uluna,1000usdr

# Generate the transaction that creates your validator
terrad gentx --name validator --amount 100uluna

# Add the generated bonding transaction to the genesis file
terrad collect-gentxs

# Now its safe to start `terrad`
terrad start

This setup puts all the data for terrad in ~/.terrad. You can examine the genesis file you created at ~/.terrad/config/genesis.json. With this configuration terracli is also ready to use and has an account with tokens (both staking and custom).

Multi-node, local, automated testnet

From the networks/local directory:

Requirements

Build

Build the terrad binary (linux) and the tendermint/terradnode docker image required for running the localnet commands. This binary will be mounted into the container and can be updated rebuilding the image, so you only need to build the image once. Go to the main project directory for Terra Core, and run:

# Build the linux binary in ./build
make build-linux

# Build tendermint/terradnode image
make build-docker-terradnode

Run your testnet

To start a 4 node testnet run:

make localnet-start

This command creates a 4-node network using the terradnode image. The ports for each node are found in this table:

Node ID P2P Port RPC Port
terranode0 26656 26657
terranode1 26659 26660
terranode2 26661 26662
terranode3 26663 26664

To update the binary, just rebuild it and restart the nodes:

make build-linux build-docker-terradnode localnet-start

Configuration

The make localnet-start creates files for a 4-node testnet in ./build by calling the terrad testnet command. This outputs a handful of files in the ./build directory:

$ tree -L 2 build/
build/
├── terracli
├── terrad
├── gentxs
│   ├── node0.json
│   ├── node1.json
│   ├── node2.json
│   └── node3.json
├── node0
│   ├── terracli
│   │   ├── key_seed.json
│   │   └── keys
│   └── terrad
│       ├── ${LOG:-terrad.log}
│       ├── config
│       └── data
├── node1
│   ├── terracli
│   │   └── key_seed.json
│   └── terrad
│       ├── ${LOG:-terrad.log}
│       ├── config
│       └── data
├── node2
│   ├── terracli
│   │   └── key_seed.json
│   └── terrad
│       ├── ${LOG:-terrad.log}
│       ├── config
│       └── data
└── node3
    ├── terracli
    │   └── key_seed.json
    └── terrad
        ├── ${LOG:-terrad.log}
        ├── config
        └── data

Each ./build/nodeN directory is mounted to the /terrad directory in each container.

Logging

Logs are saved under each ./build/nodeN/terrad/terra.log. You can also watch logs directly via Docker, for example:

docker logs -f terradnode0

Keys & Accounts

To interact with terracli and start querying state or creating txs, you use the terracli directory of any given node as your home, for example:

terracli keys list --home ./build/node0/terracli

Now that accounts exists, you may create new accounts and send those accounts funds!

::: tip Note: Each node's seed is located at ./build/nodeN/terracli/key_seed.json and can be restored to the CLI using the terracli keys add --restore command :::

Special binaries

If you have multiple binaries with different names, you can specify which one to run with the BINARY environment variable. The path of the binary is relative to the attached volume. For example:

# Run with custom binary
BINARY=terrafoo make localnet-start

Multi-node, remote, automated testnet

The following should be run from the networks directory.

Terraform & Ansible

Automated deployments are done using Terraform to create servers on AWS then Ansible to create and manage testnets on those servers.

Prerequisites

export AWS_ACCESS_KEY_ID="2345234jk2lh4234"
export AWS_SECRET_ACCESS_KEY="234jhkg234h52kh4g5khg34"
export TESTNET_NAME="remotenet"
export CLUSTER_NAME= "remotenetvalidators"
export SSH_PRIVATE_FILE="$HOME/.ssh/id_rsa"
export SSH_PUBLIC_FILE="$HOME/.ssh/id_rsa.pub"

These will be used by both terraform and ansible.

Create a remote network

SERVERS=1 REGION_LIMIT=1 make validators-start

The testnet name is what's going to be used in --chain-id, while the cluster name is the administrative tag in AWS for the servers. The code will create SERVERS amount of servers in each availability zone up to the number of REGION_LIMITs, starting at us-east-2. (us-east-1 is excluded.) The below BaSH script does the same, but sometimes it's more comfortable for input.

./new-testnet.sh "$TESTNET_NAME" "$CLUSTER_NAME" 1 1

Quickly see the /status endpoint

make validators-status

Delete servers

make validators-stop

Logging

You can ship logs to Logz.io, an Elastic stack (Elastic search, Logstash and Kibana) service provider. You can set up your nodes to log there automatically. Create an account and get your API key from the notes on this page, then:

yum install systemd-devel || echo "This will only work on RHEL-based systems."
apt-get install libsystemd-dev || echo "This will only work on Debian-based systems."

go get github.com/mheese/journalbeat
ansible-playbook -i inventory/digital_ocean.py -l remotenet logzio.yml -e LOGZIO_TOKEN=ABCDEFGHIJKLMNOPQRSTUVWXYZ012345

Monitoring

You can install the DataDog agent with:

make datadog-install