POSDAO Testnet Prepare

Preparation scripts and a launcher of POSDAO testchain. It runs validator nodes, archive/rpc node, and netstat dashboard on a single server in docker containers. The number of validator nodes is configurable.

The setup is wrapped to docker image. The current version simultaneously uses Nethermind and OpenEthereum clients to run the nodes.

Build

git clone https://github.com/xdaichain/posdao-testnet-prepare
cd posdao-testnet-prepare
docker build -t poanetwork/posdao-testnet-prepare .

This will create the docker image to use it as described below.

Usage

The setup assumes a few steps to run the testchain:

• configuring and automatic creation of docker-compose.yml files for all nodes, preparing spec.json file
• running the testchain from scratch
• deploying staking token contract after the chain starts

1. Clone this repository and go to the compose directory:

git clone https://github.com/xdaichain/posdao-testnet-prepare
cd posdao-testnet-prepare/compose

2. Configure environment variables before starting the preparation scripts.

cp .env.example .env
nano .env # or use your favorite text editor

The config parameters are (default values can be left as is):

• NETWORK_NAME - testnet name written to spec.json file and used in autogenerated docker-compose.yml files for nodes.
• NETWORK_ID - network ID of the testchain.
• OWNER_BALANCE - number of coins available for the owner account (in 10**18 wei). This account will be generated automatically.
• DELEGATOR_MIN_STAKE - minimum number of staking tokens needed to be a POSDAO delegator (in 10**18 wei).
• CANDIDATE_MIN_STAKE - minimum number of staking tokens needed to be a POSDAO candidate (in 10**18 wei).
• STAKING_EPOCH_DURATION - the duration of POSDAO staking epoch in blocks.
• STAKE_WITHDRAW_DISALLOW_PERIOD - the duration period (in blocks) at the end of staking epoch during which participants are not allowed to stake/withdraw/order/claim their staking tokens.
• COLLECT_ROUND_LENGTH - the length of the randomness collection round (in blocks).
• VALIDATORS_NUMBER - the number of initial validators (up to 21 in the current version).
• SEQAPIKEY - an API key for Seq log collector. Used on Nethermind nodes to collect logs useful for debugging purposes.

3. Run preparation scripts.

This step will create nodes subdirectory in the compose directory.

pwd # make sure your are in the `compose` directory
docker-compose up

The appeared nodes directory will contain the following subdirectories and files:

• archive - the directory for archive/rpc node.
• ethstats - the directory for Netstats dashboard.
• validatorN - the directory for validator N node where N is the number of validator.
• spec.json - the spec file used by all nodes.
• run_all.sh - the script to run all nodes and Netstats dashboard (to up their docker containers).
• stop_all.sh - the script to stop all nodes and Netstats dashboard (to down their docker containers).

4. Start all nodes.

cd nodes
chmod +x run_all.sh
./run_all.sh

This will create and start all docker containers.

5. Deploy staking token contract and make initial stakes by validators.

This step must be launched until the first staking epoch is finished.

cd .. # go back to the `compose` directory
docker-compose start && docker-compose logs -f

6. Get owner's address and private key.

After the chain starts, only owner account will have positive balance in native coins.

To get owner's address:

pwd # make sure your are in the `compose` directory
docker-compose start && docker exec testnetprep cat /testnet/scripts/set-env.sh | grep OWNER=

To get owner's private key:

docker-compose start && docker exec testnetprep cat /testnet/keys/{OWNER_ADDRESS} # insert here the address got on the previous substep

7. Try to view Netstats dashboard and connect to RPC:

Netstat is available on http://[EXTERNAL_IP]:3000
RPC: http://[EXTERNAL_IP]:8545
WebSocket: ws://[EXTERNAL_IP]:8546/ws

Restarting nodes

If you need to restart a node, go to its subdirectory inside the compose/nodes directory, and run

docker-compose down
docker-compose up -d

Cleanup

If you need to stop all containers and remove them:

cd compose/nodes
chmod +x stop_all.sh
./stop_all.sh
cd .. # go back to the `compose` directory
docker-compose down
rm -rf nodes # to remove the testchain