How to relaunch a Testnet node to use the new Testnet after a reset while using the Docker quickstart container?

Question

What are the recommended steps that should be taken to get a stellar testnet node using the quickstart docker reset on the new testnet?

Currently it is still using the old network by the looks of it, and the only information I can find is related to packages (https://github.com/stellar/packages#testnet-reset) and the commands do not apply to those of us using the Docker container. A set of similar commands to use with the docker would be great.

How can I get my stellar-core, horizon and postgres database working now that the Testnet has been reset?

I am running the quickstart container with a volume mounted on an Amazon EC2 instance with Amazon Linux 2.

The network was reset yesterday and is scheduled to happen 4 times a year so information on how to deal with this inevitable situation is greatly appreciated.

Answer 1

a) I guess the easiest way should be to run your testnet node in a non persistent ephemeral mode and simply restart the container.

b) You could as well open a shell into your container and manually recreate everything similar to the package way.

# open container shell
docker ps
docker exec -it [CONTAINER ID] /bin/bash

# stop core + horizon
supervisorctl stop stellar-core
supervisorctl stop horizon

# recreate databases
sudo -u postgres psql -c 'DROP DATABASE core' postgres
sudo -u postgres psql -c 'DROP DATABASE horizon' postgres
sudo -u postgres psql -c 'CREATE DATABASE core' postgres
sudo -u postgres psql -c 'CREATE DATABASE horizon' postgres

# remove buckets dir contents
rm -rf /opt/stellar/core/buckets/*

# init db schemes
/opt/stellar/horizon/bin/horizon db init
stellar-core --conf /opt/stellar/core/etc/stellar-core.cfg  new-db

# optional set "PREFERRED_PEERS_ONLY=true" to only connect to SDF nodes and thus mitigate stale SCP message issues
# vi /opt/stellar/core/etc/stellar-core.cfg

# start core + horizon
supervisorctl start stellar-core
supervisorctl start horizon

# check core status
curl localhost:11626/info
# check horizon status
curl localhost:8000

# optional unset "PREFERRED_PEERS_ONLY=true" once you are synced
# vi /opt/stellar/core/etc/stellar-core.cfg
# supervisorctl restart stellar-core