This guide contains steps to set up a node manually.
Please read about the sentry node architecture for the recommended production environment. Generally two nodes, one validator and one sentry / public node should be used.
The switcheod binary contains a full Switcheo TradeHub node as well as several other services.
switcheod startruns just the full nodeswitcheod rest-serverruns the cosmos-sdk REST server that allows for querying of chain data.switcheod persistenceruns a service that indexes trade transactions into postgresql and redis.switcheod interchainruns a service that writes transaction hashes for deposits and withdrawals into thepersistencedatabase - run this if you need that data.switcheod rest-apiruns a custom REST API server that serves queries from data indexed by thepersistenceservice.switcheod ws-apiruns a custom WS API server that streams data that is indexed by thepersistenceservice.switcheod oracleruns the oracle service - validators need to run this when trading begins, however this can be done on a separate node (such as a sentry node). Additionally,persistenceandrest-apineed to be running.switcheod liquidatorruns the liquidator service - only one validator needs to run this, but any validators that do will earn additional rewards. This can be done on a separate node (such as a sentry node).switcheod gasstationruns a service that helps users to create Ethereum wallets and deposit transactions - this only needs to be run by exchange operators.switcheod start -aruns all the above services.
The cosmovisor binary coordinates node upgrades, which automatically transition to the upgraded binary upon reaching the upgrade blockheight. You will run cosmovisor instead of switcheod and arguements passed to cosmovisor command will be passed to switcheod.
Nodes that serve public APIs (running persistence and reset-api, or ws-api) need the following databases installed.
Validator operators should run at least one full node with these installed for running the oracle service, but can skip installing these for the validator node itself.
- Redis
- Postgresql
Skip these for the validator node that are not running additional services.
Update package information:
sudo apt-get updateInstall and configure redis like so:
# install redis
sudo apt-get install -y redis-server
# use systemd to manage redis
sudo cp /etc/redis/redis.conf /etc/redis/redis.conf.$(date +%y%b%d-%H%M%S)
sudo sed -i -e 's/^supervised no/supervised systemd/g' /etc/redis/redis.conf
sudo systemctl restart redis.serviceInstall and configure postgresql like so:
# install psql
sudo apt-get install -y postgresql-10
# change authentication method to trust mode
# do not expose postgres on any public interfaces!
sudo sed -i -e '/^local all postgres peer$/d' \
-e 's/ peer/ trust/g' \
-e 's/ md5/ trust/g' \
/etc/postgresql/10/main/pg_hba.conf
# restart for config reload
sudo service postgresql restart
# modify this if different
POSTGRES_USER=ubuntu
# make user a psql superuser
psql postgres -tAc "SELECT 1 FROM pg_roles WHERE rolname='${POSTGRES_USER}'" | grep -q 1 || sudo -i -u postgres psql -c "CREATE USER ${POSTGRES_USER} SUPERUSER;"If you are running a public facing node, you should install nginx to serve requests:
# install nginx
sudo apt-get install -y nginx
# setup nginx proxy such that cosmos REST endpoints go to 1317
# while the rest goes to the custom off chain REST server at 5002.
# create nginx cache folder
sudo mkdir -p /var/cache/nginx
sudo bash -c 'cat <<EOT >> /etc/nginx/conf.d/switcheocli.conf
proxy_cache_path /var/cache/nginx/tradescan levels=1:2 keys_zone=tradescan_cache:10m max_size=10g inactive=2s use_temp_path=off;
proxy_cache_path /var/cache/nginx/cosmos_rest_server levels=1:2 keys_zone=cosmos_rest_server_cache:10m max_size=10g inactive=2s use_temp_path=off;
proxy_cache_path /var/cache/nginx/tendermint_api_server levels=1:2 keys_zone=tendermint_api_server_cache:10m max_size=10g inactive=2s use_temp_path=off;
server {
listen 5001 default_server;
listen [::]:5001 default_server;
server_name _;
proxy_cache tradescan_cache;
proxy_ignore_headers X-Accel-Expires Expires Cache-Control;
proxy_cache_valid any 1s;
proxy_cache_lock on;
location / {
proxy_pass http://127.0.0.1:5002;
}
location /node_info {
add_header 'Access-Control-Allow-Origin' '*';
proxy_pass http://127.0.0.1:1317;
}
location ~ ^/(staking|supply|slashing|distribution|auth|txs|subaccount|blocks|bank|gov)/ {
add_header 'Access-Control-Allow-Origin' '*';
proxy_pass http://127.0.0.1:1317;
}
}
server {
listen 1317 default_server;
listen [::]:1317 default_server;
server_name _;
proxy_cache cosmos_rest_server_cache;
proxy_ignore_headers X-Accel-Expires Expires Cache-Control;
proxy_cache_valid any 1s;
proxy_cache_lock on;
location / {
proxy_pass http://127.0.0.1:1318;
}
}
server {
listen 26657 default_server;
listen [::]:26657 default_server;
server_name _;
proxy_cache tendermint_api_server_cache;
proxy_ignore_headers X-Accel-Expires Expires Cache-Control;
proxy_cache_valid any 1s;
proxy_cache_lock on;
location / {
proxy_pass http://127.0.0.1:26659;
}
}
EOT
'
# restart nginx or reload config
sudo systemctl restart nginxRelease binaries can be found here. Unzip them to your home directory instructed in the main guide.
Copy switcheod and switcheocli to /usr/local/bin
cd install-mainnet && sudo cp bin/{switcheod,switcheocli} /usr/local/bin && cd -Add these to .bashrc or otherwise ensure these env variables are applied to the context of the node.
export APP_ENV=production
export CHAIN_ID=switcheo-tradehub-1 # or "switcheochain" for testnet
export TENDERMINT_SERVER_HOST=localhost
export TENDERMINT_SERVER_PORT=26659 # default
export COSMOS_REST_SERVER_HOST=0.0.0.0
export COSMOS_REST_SERVER_PORT=1318 # make sure this matches nginx
export API_REST_PORT=5002 # make sure this matches nginx
export POSTGRES_HOST=localhost
export POSTGRES_PORT=5432
export POSTGRES_DB=switcheochain
export POSTGRES_USER=ubuntu # modify if different
export REDIS_HOST=localhost
export REDIS_PORT=6379
export LOG_TO_FILE=1
export LOG_LEVEL=info
export SWTH_UPGRADER=cosmos1flqfs2dzzkrf49aj5pg0nj340jdqzgje02smww
export SEND_ETH_TXNS=1
export WALLET_PASSWORD=
export DAEMON_NAME=switcheod
export DAEMON_HOME=${HOME}/.switcheod
export PATH=${HOME}/.switcheod/cosmovisor/current/bin:${HOME}/.switcheod/cosmovisor/genesis/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/binReload shell:
exec $SHELL# initialize validators's and node's configuration files. Change <moniker> to your identifier.
switcheod init <moniker> --chain-id $CHAIN_ID
# configure switcheocli
switcheocli config chain-id $CHAIN_ID
switcheocli config output json
switcheocli config indent true
switcheocli config trust-node true
# configure node timeout to 1s,:
sed -i -e 's/timeout_commit = "5s"/timeout_commit = "'"1s"'"/g' ~/.switcheod/config/config.toml
# configure node pruning settings to keep-recent 100 keep-every 10000 interval 10:
sed -i -e 's/pruning-keep-recent = "0"/pruning-keep-recent = "100"/g' ~/.switcheod/config/app.toml
sed -i -e 's/pruning-keep-every = "0"/pruning-keep-every = "10000"/g' ~/.switcheod/config/app.toml
sed -i -e 's/pruning-interval = "0"/pruning-interval = "10"/g' ~/.switcheod/config/app.toml
# if this is a public node, listen on public interfaces and disable CORS:
sed -i -e 's/cors_allowed_origins = \[\]/cors_allowed_origins = \["*"\]/g' ~/.switcheod/config/config.toml
sed -i -e 's/laddr = "tcp:\/\/127.0.0.1:26657"/laddr = "tcp:\/\/0.0.0.0:26659"/g' ~/.switcheod/config/config.toml
sed -i -e 's/addr_book_strict = true/addr_book_strict = false/g' ~/.switcheod/config/config.toml
# if this is a validator node behind a sentry node, disable pex:
sed -i -e "s/pex.*/pex = false/g" ~/.switcheod/config/config.toml
# set the initial peers and trusted node to sync genesis file from
# note that this configuration may differ if you are running in sentry configuration!
## testnet
persistent_peers=026b889e51c31c5370c4a6b43d7193d583efa4a2@54.255.42.175:26656
node_url=54.255.42.175:26657
## mainnet
persistent_peers=d363e17a3d4c7649e5c59bcd33176a476433108c@54.179.34.89:26656,ca1189045e84d2be5db0a1ed326ce7cd56015f11@54.255.5.46:26656,d233cd97f8c81b7705ea16b962a4101063875151@175.41.151.35:26656
node_url=54.255.5.46:26657
# set persistent_peers
cmd="s/persistent_peers = \"\"/persistent_peers = \"${persistent_peers}\"/g"
sed -i -e "${cmd}" ~/.switcheod/config/config.toml
# load genesis file
apt-get install jq -y # next cmd uses jq
curl -s "${node_url}/genesis" | jq '.result.genesis' > ~/.switcheod/config/genesis.json
# initialise supporting directories
mkdir -p ~/.switcheod/logs/old ~/.switcheod/migrations/migrate ~/.switcheod/config/oracles
# initialise db
# NOTE: skip this if not running postgresql!
createdb switcheochainCopy switcheod and switcheocli to ~/.switcheod/cosmovisor, from release downloaded previously.
Remove switcheod/switcheocli from /usr/local/bin as we would be using the binaries from ~/.switcheod/cosmovisor.
cd install-mainnet && sudo cp -r cosmovisor ~/.switcheod/ && sudo rm /usr/local/bin/{switcheod,switcheocli} && cd - && rm -rf install-mainnetIf running the oracle and liquidator services, ensure WALLET_PASSWORD is set in ~/.env_switcheo.
The start-all -a command will attempt to run these subservices and load oraclewallet and liquidator accounts from the file-based keyring if it is present.
# run node only (for validator that is running oracle / liquidator service separately)
bash -c "source ~/.env_switcheo && cosmovisor start"
# run node with public apis /and oracle and liquidator services
bash -c "source ~/.env_switcheo && cosmovisor start -a"You can find the logs from switcheod here:
tail -f ~/.switcheod/logs/*❗ WARNING ❗
You should check that the node has caught up to latest block by running switcheocli status before continuing! If your node is still syncing (sync_info.catching_up: true), it will not be able to validate new blocks and you will end up getting slashed / jailed.
Ensure your node is healthy to avoid getting your stake slashed. You can find info on your node at:
curl localhost:26657/abci_info
You can check that your wallets have sufficient SWTH after starting through:
curl http://localhost:5001/auth/accounts/<address>
You may need some time for your node to sync to see updated information.
Promote your node to validator with this command:
# get node public key on validator machine
# testnet: switcheod node tendermint show-validator
$ switcheod tendermint show-validator
swthvalconspub1zcjduepqqsuvl3xj58qmfv49je....
# on operator machine
switcheocli tx staking create-validator --amount <amountToStakeInSatoshis>swth --moniker <yourmoniker> --pubkey <pubkey> --commission-max-change-rate 0.010000000000000000 --commission-max-rate 0.200000000000000000 --commission-rate 0.100000000000000000 --min-self-delegation 1 --fees 100000000swth -b block --from val --keyring-backend file -yYou should strongly consider setting up these services to maintain the well-being of your node.
You can use systemctl to ensure your node remains up. Here is an example systemd configuration.
[Unit]
Description=switcheod service
After=network.target
StartLimitIntervalSec=0
[Service]
Type=simple
Restart=always
RestartSec=5
User=root
ExecStart=/bin/bash -c "source ~/.env_switcheo; killall -w -s 9 switcheod switcheocli; cosmovisor start-all -a"
LimitNOFILE=64000
[Install]
Enable switcheod service to auto start after boot.
sudo systemctl enable switcheod.serviceImportant: When running the
oracleandliquidatorservice, theoraclewalletandliquidatorwallet must be in the file-based keyring of the same machine, and theWALLET_PASSWORDenvironment must be set. Otherwise,WALLET_PASSWORDshould not be set.
Switcheo TradeHub outputs quite alot of logs, and it may be neccessary to keep them. Use logrotate to manage the disk usage of logs.
sudo bash -c 'cat <<EOT >> /etc/logrotate.d/switcheo_logs
$HOME/.switcheod/logs/*.log ${HOME}/.switcheod/logs/*.err {
olddir ${HOME}/.switcheod/logs/old
daily
missingok
rotate 30
compress
copytruncate
notifempty
}
EOT