How To Run Your Own Wallet Server
This guide will walk you through the process of setting up a LBRY wallet server. This involves provisioning a web server and setting up some services (docker, lbrycrd, and the wallet server). At the end, you’ll have your own connection to the LBRY network.
note: This is early-stage stuff. You may encounter unexpected issues. Please be patient and don’t hesitate to reach out for help.
Start With A Fresh Server
We recommend a quad-core server with at least 16GB RAM, 200GB disk, and a fresh Ubuntu 18.04 install. Memory usage is flexible. 32 GB works best, but 16 GB is enough for a few clients.
Make sure your firewall has ports 9246 and 50001 open. 9246 is the port lbrycrd uses to communicate to other nodes. 50001 is the wallet server RPC port.
Install lbrycrd
Download and setup
Download the latest release of lbrycrd.
Then, create a folder on your home directory called .lbrycrd and save the following to .lbrycrd/lbrycrd.conf:
txindex=1
server=1
daemon=1
rpcuser=lbry
rpcpassword=lbry
dustrelayfee=0.00000001
Feel free to change the rpcuser or rpcpassword. If you do, you’ll have to update the DAEMON_URL variable later on (in the docker-compose.yml file) to match the user/password you chose.
Create a service (optional)
You can run lbrycrdd directly using ./lbrycrdd. However, we recommend creatinga systemd service to manage the process for you.
Create a file at /etc/systemd/system/lbrycrdd.service with the following contents:
[Unit]
Description="LBRYcrd daemon"
After=network.target
[Service]
ExecStart=/home/<your_user>/lbrycrdd -datadir="/home/<your_user>/.lbrycrd" -pid="/run/lbrycrdd/lbrycrdd.pid"
# Creates /run/lbrycrdd
RuntimeDirectory=lbrycrdd
Type=Forking
PIDFile=/run/lbrycrdd/lbrycrdd.pid
Restart=on-failure
# hardening
PrivateTmp=true
ProtectSystem=full
NoNewPrivileges=true
PrivateDevices=true
MemoryDenyWriteExecute=true
[Install]
WantedBy=multi-user.target
Then run sudo systemctl daemon-reload.
Now you can start and stop lbrycrd with sudo service lbrycrdd start and sudo service lbrycrdd stop.
You can watch the lbrycrd log with tail -f ~/.lbrycrd/debug.log
Set Up Docker
Install Docker & Docker Compose
sudo apt install -y apt-transport-https ca-certificates curl software-properties-common && \
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add - && \
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" && \
sudo apt install -y docker-ce docker-ce-cli containerd.io && \
sudo systemctl enable docker && sudo systemctl start docker && \
sudo curl -L "https://github.com/docker/compose/releases/download/1.24.1/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose && \
sudo chmod +x /usr/local/bin/docker-compose
sudo usermod -aG docker $USER
Download our example docker-compose.yml
You can see it here.
curl -L "https://raw.githubusercontent.com/lbryio/lbry-sdk/master/docker/docker-compose-wallet-server.yml" -o docker-compose.yml
Make sure the user and password in the DAEMON_URL variable (the lbry@lbry part) in this docker-compose.yml matches the user/password in your ~/.lbrycrd/lbrycrd.conf file.
Download snapshots for elasticsearch and the wallet server (optional)
You can skip the initial sync by starting from a snapshot. The following will download a snapshot of the elasticsearch volume and move it into the default location for docker volumes on ubuntu, on other systems you may need to adjust the path used here. Note: snapshot heights must be the same. The tars can be deleted after setting the volumes up.
SNAPSHOT_HEIGHT="1049658"
ES_VOLUME_PATH="/var/lib/docker/volumes/${USER}_es01"
ES_SNAPSHOT_TAR_NAME="es_snapshot_${SNAPSHOT_HEIGHT}.tar"
ES_SNAPSHOT_URL="https://snapshots.lbry.com/hub/${ES_SNAPSHOT_TAR_NAME}"
wget $ES_SNAPSHOT_URL
echo "decompressing elasticsearch snapshot"
tar -xf $ES_SNAPSHOT_TAR_NAME
sudo chown -R $USER:root "snapshot_es_${SNAPSHOT_HEIGHT}"
sudo chmod -R 775 "snapshot_es_${SNAPSHOT_HEIGHT}"
sudo mkdir -p $ES_VOLUME_PATH
sudo rm -rf "${ES_VOLUME_PATH}/_data"
sudo mv "snapshot_es_${SNAPSHOT_HEIGHT}" "${ES_VOLUME_PATH}/_data"
The following will download the wallet server docker volume and move it into place as well.
echo "fetching wallet server snapshot"
SNAPSHOT_HEIGHT="1049658"
HUB_VOLUME_PATH="/var/lib/docker/volumes/${USER}_wallet_server"
SNAPSHOT_TAR_NAME="wallet_server_snapshot_${SNAPSHOT_HEIGHT}.tar"
SNAPSHOT_URL="https://snapshots.lbry.com/hub/${SNAPSHOT_TAR_NAME}"
wget $SNAPSHOT_URL
tar -xf $SNAPSHOT_TAR_NAME
sudo mkdir -p $HUB_VOLUME_PATH
sudo rm -rf "${HUB_VOLUME_PATH}/_data"
sudo chown -R 999:999 "snapshot_${SNAPSHOT_HEIGHT}"
sudo mv "snapshot_${SNAPSHOT_HEIGHT}" "${HUB_VOLUME_PATH}/_data"
Turn It On
Start the servers
docker-compose up --detach
Check that everything worked
The first time you start the wallet server, it will take a few minutes to download a recent snapshot of the database and extract it. You can follow the progress with
docker-compose logs --follow
After the wallet server has caught up, it will bind to port 50001 and start responding to requests. You can check if this happened by running
sudo netstat -tlpn | grep 50001
If there is no output, the port is ont bound yet and the server is still catching up. Check the logs for more info.
After the wallet server is ready, check that it responds to basic RPC calls:
echo '{"id":1,"method":"server.version"}' | timeout 1 curl telnet://localhost:50001
You should see a response like {"jsonrpc": "2.0", "result": ["0.46.1", "0.0"], "id": 1}. If you do, congratulations! You’ve set up your own wallet server.
To check Elastic search, there are two commands you can use:
curl localhost:9200 # get Elastic status
curl localhost:9200/claims/_count # check how many claims have been synced to Elastic
Maintenance
Stopping and Restarting
Use the usual docker-compose commands (start, stop, pause, etc) to control the servers. Run docker-compose --help to see the
options.
Updating
To update to the latest wallet server release, run the following:
docker pull lbry/wallet-server:latest-release
docker-compose down
docker-compose up --detach
Resyncing
From time to time, we’ll release an update that requires recreating one of the databases from scratch. Most of the time we will try to ensure there is an automatic migration, but even then, if you think the server has invalid data you can also try a resync.
The process is similar to an update, but causes the server to be down for much longer.
Main database
Holds the raw blockchain data and takes several days to resync from scratch, so be sure to have a snapshot or try that last.
docker pull lbry/wallet-server:latest-release
docker-compose down
docker volume rm "$(whoami)_wallet_server"
WALLET_SERVER_SNAPSHOT_URL= docker-compose up --detach
Elasticsearch
ES does the indexing of claims from the main database. It should take around 6 hours to resync on a fast machine.
docker pull lbry/wallet-server:latest-release
docker-compose down
docker volume rm "$(whoami)_es01"
docker-compose up --detach
Get in touch
Whether you got to the end without a hiccup or you got stuck along the way, we want to hear from you. Join our Discord to get help, stay updated, and talk to other wallet server operators.
Edit this page