Application Configuration
Bitcoin-S uses HOCON to configure various parts of the application the library offers. HOCON is a superset of JSON, that is, all valid JSON is valid HOCON.
All configuration for Bitcoin-S is under the bitcoin-s
key.
If you have a file application.conf
anywhere on your classpath when using bitcoin-s, the values there take precedence
over the ones found in our
reference.conf
. We also look for the file bitcoin-s.conf
in the current Bitcoin-S data directory.
The resolved configuration gets parsed by
AppConfig
.
AppConfig
is an abstract class that's implemented by corresponding case classes in the wallet
, chain
and node
projects. Here's some examples of how to construct a wallet configuration:
import org.apache.pekko.actor.ActorSystem
import org.bitcoins.wallet.config.WalletAppConfig
import com.typesafe.config.ConfigFactory
import java.nio.file.Paths
import scala.util.Properties
import scala.concurrent.ExecutionContext.Implicits.global
implicit val system: ActorSystem = ActorSystem("configuration-example")
// reads $HOME/.bitcoin-s/
val defaultConfig = WalletAppConfig.fromDefaultDatadir()
// reads a custom data directory
val customDirectory = Paths.get(Properties.userHome, "custom-bitcoin-s-directory")
val configFromCustomDatadir = WalletAppConfig(customDirectory, Vector.empty)
// reads a custom data directory and overrides the network to be testnet3
val customOverride = ConfigFactory.parseString("bitcoin-s.network = testnet3")
val configFromCustomDirAndOverride = WalletAppConfig(customDirectory, Vector(customOverride))
You can pass as many com.typesafe.config.Config
s as you'd like. If any keys appear multiple times the last one
encountered takes precedence.
Command Line Options
There are a few command line options available that take precedence over configuration file.
--datadir <directory>
datadir
sets the data directory instead of using the default$HOME/.bitcoin-s
--rpcbind <ip>
rpcbind
sets the interface the rpc server binds to instead of using the default127.0.0.1
--rpcport <port>
rpcport
sets the port the rpc server binds to instead of using the default9999
--force-recalc-chainwork
force-recalc-chainwork
will force a recalculation of the entire chain's chain work, this can be useful if there is an incompatible migration or if it got out of sync.-Dlogback.configurationFile=/path/to/config.xml
You can set a custom logback configuration. If you need help creating a custom logback file you can read the logback configuration documentation.
Internal configuration
Database connections are also configured by using HOCON. This is done
in reference.conf
inside the db-commons
module. The options exposed here are not intended to be used by users of Bitcoin-S, and are
internal only.
Database Migrations
All of our modules that require databases now have database migrations. The tool we use for these migrations is
called flyway. To find your projects migraitons, you need to look inside of the
[project-name]/src/main/resources/[database-name]/migration/
. For example, the chain projects migrations live under
the path chain/src/main/resources/chaindb/migration/V1__chain_db_baseline.sql
.
Migrations can be executed by calling
the DbManagement.migrate()
method. Migrations are applied by default on server startup, via
the AppConfig.start()
method.
These migrations are setup so that project's databases and migrations are independent of each other. Therefore if you
want to use the bitcoin-s-chain
project, but not the bitcoin-s-wallet
project, wallet migrations are not applied. It
should be noted if you are using a module as a library, you are responsible for configuring the database via
slick's configuration and calling
AppConfig.start()
to ensure the entire module is initialized correctly.
Example Configuration File
bitcoin-s {
# the network your bitcoin-s node is running on
network = "testnet3" # regtest, testnet3, mainnet, signet
# specify what backend you are using with bitcoin-s
# by default we do neutrino, but you can also connect
# bitcoind with the configuration settings in bitcoin-s.bitcoind-rpc
node.mode = neutrino # neutrino, bitcoind
# configurations for connecting to bitcoind
bitcoind-rpc {
# bitcoind rpc username
rpcuser = user
# bitcoind rpc password
# If your password contains the characters '$','{', '}', '[', ']', ':', '=', ',', '+', '#', '`', '^', '?', '!', '@', '*', '&', whitespace
# or the string "//", enclose it in double quotes
# rpcpassword = "password=" if the original password is password=, rpcpassword = "passwo//rd" if the original password is passwo//rd etc.
# If it contains '\' or '"', escape it with '\'
# rpcpassword = "pass\\word" if the original password is pass\word, rpcpassword = "pass\"word" if the original password is pass"word
rpcpassword = password
# Binary location of bitcoind
binary = ${HOME}/.bitcoin-s/binaries/bitcoind/bitcoin-0.20.1/bin/bitcoind
# bitcoind datadir
datadir = ${HOME}/.bitcoin
# bitcoind network host
connect = localhost
# bitcoind p2p port
port = 8333
# bitcoind rpc host
rpcconnect = localhost
# bitcoind rpc port
rpcport = 8332
# bitcoind zmq raw tx
zmqpubrawtx = "tcp://127.0.0.1:28332"
# bitcoind zmq raw block
zmqpubrawblock = "tcp://127.0.0.1:28333"
# bitcoind zmq hash tx
zmqpubhashtx = "tcp://127.0.0.1:28330"
# bitcoind zmq raw block
zmqpubhashblock = "tcp://127.0.0.1:28331"
#If you have a bitcoind instance that is running remotely on another machine, you should set it to true
isRemote = false
}
# settings if you are using a neutrino node in bitcoin-s
node {
# a list of peer addresses in form "hostname:portnumber"
# Port number is optional, the default value is 8333 for mainnet,
# 18333 for testnet and 18444 for regtest.
peers = [""]
# try to connect to peers from dns seeds, database, addr messages etc
enable-peer-discovery = true
# number of persistent peer connections to maintain for node use
maxConnectedPeers = 2
# time interval for trying next set of peers in peer discovery
try-peers-interval = 12 hour
# the delay until we start attempting to connect to peers
try-peers-start-delay = 1 second
# wait time for queries like getheaders etc before switching to another
query-wait-time = 120s
hikari-logging = true
hikari-logging-interval = 10 minute
# whether to have p2p peers relay us unconfirmed txs
relay = false
# how often we run health checks for our peers
health-check-interval = 1 minutes
# if the peer does not send us a message within this duration
# we disconnect it for inactivity
peer-timeout = 20 minute
# how long we wait until we attempt to re-connect to a peer we have
# in our database that we have connected to previously
connection-attempt-cool-down-period = 5 minutes
}
proxy {
# You can configure SOCKS5 proxy to use Tor for outgoing connections
enabled = false
socks5 = "127.0.0.1:9050"
}
# tor settings
tor {
# You can enable Tor for incoming connections
enabled = false
control = "127.0.0.1:9051"
# Tor daemon can be provided by the node operator.
# If this parameter set to true, bitcoin-s will connect the provided Tor daemon.
# Otherwise bitcoin-s will start its own pre-packaged daemon.
provided = false
# This parameter allows to use random port numbers for pre-packaged Tor daemon,
# which is useful if another Tor daemon instance already bound SOCKS5 and control ports.
# In this case bitcoin-s.tor.control and bitcoin-s.proxy.socks5
# addresses will be automatically changed to "localhost:<random port>"
use-random-ports = true
# The password used to arrive at the HashedControlPassword for the control port.
# If provided, the HASHEDPASSWORD authentication method will be used instead of
# the SAFECOOKIE one.
# password = securePassword
# The path to the private key of the onion service being created
# privateKeyPath = /path/to/priv/key
# Optonal Tor targets. If empty all hidden serices will be created at localhost.
targets = []
}
# settings for the chain module
chain {
neutrino {
filter-header-batch-size.default = 2000
filter-header-batch-size.regtest = 10
# You can set a network specific filter-header-batch-size
# by adding a trailing `.networkId` (main, test, regtest)
# It is recommended to keep the main and test batch size high
# to keep the sync time fast, however, for regtest it should be small
# so it does not exceed the chain size.
filter-batch-size = 1000
}
hikari-logging = true
hikari-logging-interval = 10 minute
websocket {
# don't emit block processed events over the websocket
# until IBD is complete. This is an optimization for the
# the UI so it doesn't have to handle hundreds of thousands of
# events while IBD is going on.
block-processed-ibd = false
}
}
# settings for wallet module
wallet {
# You can have multiple wallets by setting a different
# wallet name for each of them. They will each have
# their own unique seed and database or schema,
# depending on the database driver.
# The wallet name can contain letters, numbers, and underscores '_'.
# walletName = MyWallet0
defaultAccountType = segwit # legacy, segwit, nested-segwit
bloomFalsePositiveRate = 0.0001 # percentage
# the number of consecutive addresses that we do not
# discover funds in before we mark as rescan as exhausted
# this is needed because we can never truely tell how many addresses
# the wallet has used when executing a rescan from a seed
addressGapLimit = 100
# the number of addresses that get generated everytime
# we need to rescan. If a match occurs within the addressGapLimit
# we generate another discoveryBatchSize addresses and then rescan again
discoveryBatchSize = 100
requiredConfirmations = 6
# Expected average fee rate over the long term
# in satoshis per virtual byte
longTermFeeRate = 10
# How big the address queue size is before we throw an exception
# because of an overflow
addressQueueSize = 10
# How long we attempt to generate an address for
# before we timeout
addressQueueTimeout = 5 seconds
# Allow external payout and change addresses in DLCs
# By default all DLC addresses are generated by the wallet itself
allowExternalDLCAddresses = false
# How often the wallet will rebroadcast unconfirmed transactions
rebroadcastFrequency = 4 hours
hikari-logging = true
hikari-logging-interval = 10 minute
}
keymanager {
# You can optionally set a BIP 39 password
# bip39password = "changeMe"
# Password that your seed is encrypted with
# aesPassword = changeMe
# At least 16 bytes of entropy encoded in hex
# This will be used as the seed for any
# project that is dependent on the keymanager
# entropy = ""
}
# Bitcoin-S provides manny different fee providers
# You can configure your server to use any of them
# Below is some examples of different options
fee-provider {
# name = mempoolspace # Uses mempool.space's api
# The target is optional for mempool.space
# It refers to the expected number of blocks until confirmation
# target = 6
# name = bitcoinerlive # Uses bitcoiner.live's api
# The target is optional for Bitcoiner Live
# It refers to the expected number of blocks until confirmation
# target = 6
# name = bitgo # Uses BitGo's api
# The target is optional for BitGo
# It refers to the expected number of blocks until confirmation
# target = 6
# name = constant # A constant fee rate in sats/vbyte
# target = 1 # Will always use 1 sat/vbyte
}
dlcnode {
# The address we are listening on for incoming connections for DLCs
# Binding to 0.0.0.0 makes us listen to all incoming connections
# Consider using 127.0.0.1 listen address if Tor is enabled.
# listen = "0.0.0.0:2862"
# The address our peers use to connect to our node.
# By default it's the same as the listen address,
# or if Tor is enabled, the hidden service's onion address.
# You can specify a port number like this "192.168.0.1:12345",
# The default port number is the same as in the listen adrress
# external-ip = "192.168.0.1"
}
server {
# The port we bind our rpc server on
rpcport = 9999
# The ip address we bind our server too
rpcbind = "127.0.0.1"
# The port we bind our websocket server on
wsport = 19999
# The ip address we bind the websocket server too
wsbind = "127.0.0.1"
# The basic auth password. It must me must be non empty.
password = topsecret
}
oracle {
# The port we bind our rpc server on
rpcport = 9998
# The ip address we bind our server too
rpcbind = "127.0.0.1"
# The basic auth password. It must me must be non empty.
password = supersecret
hikari-logging = true
hikari-logging-interval = 10 minute
db {
path = ${bitcoin-s.datadir}/oracle/
}
}
dbDefault = {
dataSourceClass = slick.jdbc.DatabaseUrlDataSource
profile = "slick.jdbc.SQLiteProfile$"
db {
# for information on parameters available here see
# https://scala-slick.org/doc/3.3.1/api/index.html#slick.jdbc.JdbcBackend$DatabaseFactoryDef@forConfig(String,Config,Driver,ClassLoader):Database
path = ${bitcoin-s.datadir}/${bitcoin-s.network}/
driver = org.sqlite.JDBC
user = ""
password = ""
host = localhost
port = 5432
# this needs to be set to 1 for SQLITE as it does not support concurrent database operations
# see: https://github.com/bitcoin-s/bitcoin-s/pull/1840
numThreads = 1
queueSize=5000
connectionPool = "HikariCP"
registerMbeans = true
}
hikari-logging = false
hikari-logging-interval = 10 minute
}
testkit {
pg {
#enabled postgres backend database for all test cases
enabled = false
}
}
}
pekko {
loglevel = "OFF"
stdout-loglevel = "OFF"
http {
client {
# The time after which an idle connection will be automatically closed.
# Set to `infinite` to completely disable idle connection timeouts.
# some requests potentially take a long time, like generate and prune
idle-timeout = 5 minutes
}
server {
# The amount of time until a request times out on the server
# If you have a large payload this may need to be bumped
# https://doc.akka.io/docs/akka-http/current/common/timeouts.html#request-timeout
request-timeout = 10s
}
}
actor {
debug {
# enable DEBUG logging of all AutoReceiveMessages (Kill, PoisonPill etc.)
autoreceive= off
# enable function of LoggingReceive, which is to log any received message at
# DEBUG level
receive = on
# enable DEBUG logging of unhandled messages
unhandled = off
# enable DEBUG logging of actor lifecycle changes
lifecycle = off
event-stream=off
}
}
}
Database configuration
By default, bitcoin-s uses Sqlite to store its data. It creates three Sqlite databases
in ~/.bitcoin-s/${network}
: chain.sqlite
for chain
project,
node.sqlite
for node
project and wallet.sqlite
the wallet. This is the default configuration, it doesn't require
additional changes in the config file.
bitcoin-s
also supports PostgreSQL as a database backend. In order to use a PostgreSQL database for all project you
need to add following into your config file:
bitcoin-s {
common {
profile = "slick.jdbc.PostgresProfile$"
db {
driver = org.postgresql.Driver
# these 3 options will result into a jdbc url of
# "jdbc:postgresql://localhost:5432/database"
name = database
host = localhost
port = 5432
user = "user"
password = "topsecret"
numThreads = 5
# http://scala-slick.org/doc/3.3.3/database.html
connectionPool = "HikariCP"
registerMbeans = true
}
}
chain.profile = ${bitcoin-s.common.profile}
chain.db = ${bitcoin-s.common.db}
chain.db.poolName = "chain-connection-pool"
node.profile = ${bitcoin-s.common.profile}
node.db = ${bitcoin-s.common.db}
node.db.poolName = "node-connection-pool"
wallet.profile = ${bitcoin-s.common.profile}
wallet.db = ${bitcoin-s.common.db}
wallet.db.poolName = "wallet-connection-pool"
oracle.profile = ${bitcoin-s.common.profile}
oracle.db = ${bitcoin-s.common.db}
oracle.db.poolName = "oracle-connection-pool"
}
The database driver will create a separate SQL namespace for each sub-project: chain
, node
and wallet
.
Also you can use mix databases and drivers in one configuration. For example, This configuration file enables Sqlite
for node
project (it's default, so its configuration is omitted), and walletdb
and chaindb
PostgreSQL databases
for wallet
and chain
projects:
bitcoin-s {
chain {
profile = "slick.jdbc.PostgresProfile$"
db {
driver = org.postgresql.Driver
name = chaindb
host = localhost
port = 5432
user = "user"
password = "topsecret"
}
}
wallet {
profile = "slick.jdbc.PostgresProfile$"
db {
driver = org.postgresql.Driver
name = walletdb
host = localhost
port = 5432
user = "user"
password = "topsecret"
}
}
}