README
¶
hdfd
Decred Overview
Decred is a blockchain-based cryptocurrency with a strong focus on community
input, open governance, and sustainable funding for development. It utilizes a
hybrid proof-of-work and proof-of-stake mining system to ensure that a small
group cannot dominate the flow of transactions or make changes to Decred without
the input of the community. A unit of the currency is called a hdfchain (HDF).
Latest Downloads
Core software:
- hdfd: a Decred full node daemon (this)
- hdfwallet: a CLI Decred wallet daemon
- hdfctl: a CLI client for hdfd and hdfwallet
Bundles:
- Decrediton: a GUI bundle for
hdfdandhdfwallet - CLI app suite:
a CLI bundle for
hdfdandhdfwallet
What is hdfd?
hdfd is a full node implementation of Decred written in Go (golang).
It acts as a fully-validating chain daemon for the Decred cryptocurrency. hdfd maintains the entire past transactional ledger of Decred and allows relaying of transactions to other Decred nodes around the world.
This software is currently under active development. It is extremely stable and has been in production use since February 2016.
It important to note that hdfd does NOT include wallet functionality. Users who desire a wallet will need to use hdfwallet(CLI) or Decrediton(GUI).
What is a full node?
The term 'full node' is short for 'fully-validating node' and refers to software that fully validates all transactions and blocks, as opposed to trusting a 3rd party. In addition to validating transactions and blocks, nearly all full nodes also participate in relaying transactions and blocks to other full nodes around the world, thus forming the peer-to-peer network that is the backbone of the Decred cryptocurrency.
The full node distinction is important, since full nodes are not the only type of software participating in the Decred peer network. For instance, there are 'lightweight nodes' which rely on full nodes to serve the transactions, blocks, and cryptographic proofs they require to function, as well as relay their transactions to the rest of the global network.
Why run hdfd?
As described in the previous section, the Decred cryptocurrency relies on having a peer-to-peer network of nodes that fully validate all transactions and blocks and then relay them to other full nodes.
Running a full node with hdfd contributes to the overall security of the network, increases the available paths for transactions and blocks to relay, and helps ensure there are an adequate number of nodes available to serve lightweight clients, such as Simplified Payment Verification (SPV) wallets.
Without enough full nodes, the network could be unable to expediently serve users of lightweight clients which could force them to have to rely on centralized services that significantly reduce privacy and are vulnerable to censorship.
In terms of individual benefits, since hdfd fully validates every block and transaction, it provides the highest security and privacy possible when used in conjunction with a wallet that also supports directly connecting to it in full validation mode, such as hdfwallet (CLI) and Decrediton (GUI).
Minimum Recommended Specifications (hdfd only)
- 12 GB disk space (as of April 2020, increases over time)
- 1GB memory (RAM)
- ~150MB/day download, ~1.5GB/day upload
- Plus one-time initial download of the entire block chain
- Windows 10 (server preferred), macOS, Linux
- High uptime
Getting Started
So, you've decided to help the network by running a full node. Great! Running hdfd is simple. All you need to do is install hdfd on a machine that is connected to the internet and meets the minimum recommended specifications, and launch it.
Also, make sure your firewall is configured to allow inbound connections to port 9108.
Installing and updating
Binaries (Windows/Linux/macOS)
Binary releases are provided for common operating systems and architectures. The easiest method is to download Decrediton from the link below, which will include hdfd. Advanced users may prefer the Command-line app suite, which includes hdfd and hdfwallet.
- How to verify binaries before installing: https://docs.clkj.ltd/advanced/verifying-binaries/
- How to install the CLI Suite: https://docs.clkj.ltd/wallets/cli/cli-installation/
- How to install Decrediton: https://docs.clkj.ltd/wallets/hdfchainiton/hdfchainiton-setup/
Build from source (all platforms)
Install Dependencies
-
Go 1.14 or 1.15
Installation instructions can be found here: https://golang.org/doc/install. Ensure Go was installed properly and is a supported version:
$ go version $ go env GOROOT GOPATHNOTE:
GOROOTandGOPATHmust not be on the same path. Since Go 1.8 (2016),GOROOTandGOPATHare set automatically, and you do not need to change them. However, you still need to add$GOPATH/binto yourPATHin order to run binaries installed bygo getandgo install(On Windows, this happens automatically).Unix example -- add these lines to .profile:
PATH="$PATH:/usr/local/go/bin" # main Go binaries ($GOROOT/bin) PATH="$PATH:$HOME/go/bin" # installed Go projects ($GOPATH/bin) -
Git
Installation instructions can be found at https://git-scm.com or https://gitforwindows.org.
$ git version
Windows Example
PS> git clone https://github.com/hdfchain/hdfd $env:USERPROFILE\src\hdfd
PS> cd $env:USERPROFILE\src\hdfd
PS> go install . .\cmd\...
PS> hdfd -V
Run the hdfd executable now installed in "$(go env GOPATH)\bin".
Unix Example
This assumes you have already added $GOPATH/bin to your $PATH as described
in dependencies.
$ git clone https://github.com/hdfchain/hdfd $HOME/src/hdfd
$ git clone https://github.com/hdfchain/hdfctl $HOME/src/hdfctl
$ (cd $HOME/src/hdfd && go install . ./...)
$ (cd $HOME/src/hdfctl && go install)
$ hdfd -V
Run the hdfd executable now installed in $GOPATH/bin.
Docker
Running hdfd
You can run a hdfchain node from inside a docker container. To build the image yourself, use the following command:
docker build -t hdfchain/hdfd .
Or you can create an alpine based image (requires Docker 17.05 or higher):
docker build -t hdfchain/hdfd:alpine -f Dockerfile.alpine .
You can then run the image using:
docker run hdfchain/hdfd
You may wish to use an external volume to customize your config and persist the data in an external volume:
docker run --rm -v /home/user/hdfdata:/root/.hdfd/data hdfchain/hdfd
For a minimal image, you can use the hdfchain/hdfd:alpine tag. This is typically a more secure option while also being a much smaller image.
You can run hdfctl from inside the image. For example, run an image (mounting
your data from externally) with:
docker run --rm -ti --name=hdfd-1 -v /home/user/.hdfd:/root/.hdfd \
hdfchain/hdfd:alpine
And then run hdfctl commands against it. For example:
docker exec -ti hdfd-1 hdfctl getbestblock
Running Tests
All tests and linters may be run using the script run_tests.sh. Generally,
Decred only supports the current and previous major versions of Go.
./run_tests.sh
Contact
If you have any further questions you can find us at:
Issue Tracker
The integrated github issue tracker is used for this project.
Documentation
The documentation for hdfd is a work-in-progress. It is located in the docs folder.
License
hdfd is licensed under the copyfree ISC License.
Documentation
¶
Overview ¶
hdfd is a full-node Decred implementation written in Go.
The default options are sane for most users. This means hdfd will work 'out of the box' for most users. However, there are also a wide variety of flags that can be used to control it.
The following section provides a usage overview which enumerates the flags. An interesting point to note is that the long form of all of these options (except -C) can be specified in a configuration file that is automatically parsed when hdfd starts up. By default, the configuration file is located at ~/.hdfd/hdfd.conf on POSIX-style operating systems and %LOCALAPPDATA%\hdfd\hdfd.conf on Windows. The -C (--configfile) flag, as shown below, can be used to override this location.
Usage:
hdfd [OPTIONS]
Application Options:
-V, --version Display version information and exit
-A, --appdata= Path to application home directory
-C, --configfile= Path to configuration file
-b, --datadir= Directory to store data
--logdir= Directory to log output
--nofilelogging= Disable file logging
--dbtype= Database backend to use for the block chain
(default: ffldb)
--profile= Enable HTTP profiling on given [addr:]port --
NOTE: port must be between 1024 and 65536
--cpuprofile= Write CPU profile to the specified file
--memprofile= Write mem profile to the specified file
--testnet Use the test network
--simnet Use the simulation test network
--regnet Use the regression test network
-d, --debuglevel= Logging level for all subsystems {trace, debug,
info, warn, error, critical} -- You may also
specify
<subsystem>=<level>,<subsystem2>=<level>,... to
set the log level for individual subsystems --
Use show to list available subsystems (info)
--sigcachemaxsize= The maximum number of entries in the signature
verification cache (default: 100000)
--norpc Disable built-in RPC server -- NOTE: The RPC
server is disabled by default if no
rpcuser/rpcpass or rpclimituser/rpclimitpass is
specified
--rpclisten= Add an interface/port to listen for RPC
connections (default port: 9109, testnet: 19109)
-u, --rpcuser= Username for RPC connections
-P, --rpcpass= Password for RPC connections
--rpclimituser= Username for limited RPC connections
--rpclimitpass= Password for limited RPC connections
--rpccert= File containing the certificate file
--rpckey= File containing the certificate key
--tlscurve= Curve to use when generating the TLS keypair
(default: P-521)
--altdnsnames Specify additional dns names to use when
generating the rpc server certificate
[supports DCRD_ALT_DNSNAMES environment variable]
--notls Disable TLS for the RPC server -- NOTE: This is
only allowed if the RPC server is bound to
localhost
--rpcmaxclients= Max number of RPC clients for standard
connections (default: 10)
--rpcmaxwebsockets= Max number of RPC websocket connections (default:
25)
--rpcmaxconcurrentreqs= Max number of concurrent RPC requests that may be
processed concurrently (default: 20)
--proxy= Connect via SOCKS5 proxy (eg. 127.0.0.1:9050)
--proxyuser= Username for proxy server
--proxypass= Password for proxy server
--onion= Connect to tor hidden services via SOCKS5 proxy
(eg. 127.0.0.1:9050)
--onionuser= Username for onion proxy server
--onionpass= Password for onion proxy server
--noonion Disable connecting to tor hidden services
--torisolation Enable Tor stream isolation by randomizing user
credentials for each connection
-a, --addpeer= Add a peer to connect with at startup
--connect= Connect only to the specified peers at startup
--nolisten Disable listening for incoming connections --
NOTE: Listening is automatically disabled if the
--connect or --proxy options are used without
also specifying listen interfaces via --listen
--listen= Add an interface/port to listen for connections
(default all interfaces port: 9108, testnet:
19108)
--maxsameip= Max number of connections with the same IP -- 0
to disable (default: 5)
--maxpeers= Max number of inbound and outbound peers
(default: 125)
--dialtimeout= How long to wait for TCP connection completion
Valid time units are {s, m, h} Minimum 1 second
(default: 30s)
--peeridletimeout The duration of inactivity before a peer is timed
out. Valid time units are {s,m,h}. Minimum 15
seconds (default: 2m0s)
--noseeders Disable seeding for peer discovery
--nodnsseed DEPRECATED: use --noseeders
--externalip= Add an ip to the list of local addresses we claim
to listen on to peers
--nodiscoverip Disable automatic network address discovery of
local external IPs
--upnp Use UPnP to map our listening port outside of NAT
--nobanning Disable banning of misbehaving peers
--banduration= How long to ban misbehaving peers. Valid time
units are {s, m, h}. Minimum 1 second (default:
24h0m0s)
--banthreshold= Maximum allowed ban score before disconnecting
and banning misbehaving peers (default: 100)
--whitelist= Add an IP network or IP that will not be banned.
(eg. 192.168.1.0/24 or ::1)
--nocheckpoints Disable built-in checkpoints. Don't do this
unless you know what you're doing
--dumpblockchain= Write blockchain as a flat file of blocks for use
with addblock, to the specified filename
--minrelaytxfee= The minimum transaction fee in HDF/kB to be
considered a non-zero fee (default: 0.0001)
--limitfreerelay= Limit relay of transactions with no transaction
fee to the given amount in thousands of bytes per
minute (default: 15)
--norelaypriority Do not require free or low-fee transactions to
have high priority for relaying
--maxorphantx= Max number of orphan transactions to keep in
memory (default: 100)
--blocksonly Do not accept transactions from remote peers
--acceptnonstd Accept and relay non-standard transactions to
the network regardless of the default settings
for the active network
--rejectnonstd Reject non-standard transactions regardless of
the default settings for the active network
--allowoldvotes Enable the addition of very old votes to the
mempool
--generate Generate (mine) bitcoins using the CPU
--miningaddr= Add the specified payment address to the list of
addresses to use for generated blocks -- At least
one address is required if the generate option is
set
--blockminsize= Minimum block size in bytes to be used when
creating a block
--blockmaxsize= Maximum block size in bytes to be used when
creating a block (default: 375000)
--blockprioritysize= Size in bytes for high-priority/low-fee
transactions when creating a block (default:
20000)
--miningtimeoffset= Offset the mining timestamp of a block by this
many seconds (positive values are in the past)
--nonaggressive Disable mining off of the parent block of the
blockchain if there aren't enough voters
--nominingstatesync Disable synchronizing the mining state with other
nodes
--allowunsyncedmining Allow block templates to be generated even when
the chain is not considered synced on networks
other than the main network. This is
automatically enabled when the simnet option is
set. Don't do this unless you know what you're
doing
--txindex Maintain a full hash-based transaction index
which makes all transactions available via the
getrawtransaction RPC
--droptxindex Deletes the hash-based transaction index from the
database on start up and then exits
--addrindex Maintain a full address-based transaction index
which makes the searchrawtransactions RPC
available
--dropaddrindex Deletes the address-based transaction index from
the database on start up and then exits
--noexistsaddrindex Disable the exists address index, which tracks
whether or not an address has even been used
--dropexistsaddrindex Deletes the exists address index from the
database on start up and then exits
--nocfilters (Deprecated) Disable compact filtering (CF)
support
--dropcfindex (Deprecated) Deletes the index used for compact
filtering (CF) support from the database on start
up and then exits
--piperx= File descriptor of read end pipe to enable parent
-> child process communication
--pipetx= File descriptor of write end pipe to enable
parent <- child process communication
--lifetimeevents Send lifetime notifications over the TX pipe
Help Options:
-h, --help Show this help message
Source Files
¶
Directories
¶
| Path | Synopsis |
|---|---|
|
blockchain
module
|
|
|
stake
Module
|
|
|
certgen
module
|
|
|
chaincfg
module
|
|
|
chainhash
Module
|
|
|
cmd
|
|
|
crypto
|
|
|
blake256
Module
|
|
|
ripemd160
Module
|
|
|
database
module
|
|
|
dcrec
module
|
|
|
edwards
Module
|
|
|
secp256k1
Module
|
|
|
dcrjson
module
|
|
|
gcs
module
|
|
|
blockcf
Module
|
|
|
internal
|
|
|
fees
Package fees provides hdfchain-specific methods for tracking and estimating fee rates for new transactions to be mined into the network.
|
Package fees provides hdfchain-specific methods for tracking and estimating fee rates for new transactions to be mined into the network. |
|
fees/cmd/dumpfeedb
Tool dumpfeedb can be used to dump the internal state of the buckets of an estimator's feedb so that it can be externally analyzed.
|
Tool dumpfeedb can be used to dump the internal state of the buckets of an estimator's feedb so that it can be externally analyzed. |
|
limits
Package limits allows some process limits to be raised.
|
Package limits allows some process limits to be raised. |
|
mempool
Package mempool provides a policy-enforced pool of unmined Decred transactions.
|
Package mempool provides a policy-enforced pool of unmined Decred transactions. |
|
mining
Package mining includes all mining and policy types, and will house all mining code in the future.
|
Package mining includes all mining and policy types, and will house all mining code in the future. |
|
mining/cpuminer
Package cpuminer provides facilities for solving blocks (mining) using the CPU.
|
Package cpuminer provides facilities for solving blocks (mining) using the CPU. |
|
rpcserver
Package rpcserver includes all RPC server interfaces, types, and pieces of code pertaining to implementing the RPC server.
|
Package rpcserver includes all RPC server interfaces, types, and pieces of code pertaining to implementing the RPC server. |
|
version
Package version provides a single location to house the version information for hdfd and other utilities provided in the same repository.
|
Package version provides a single location to house the version information for hdfd and other utilities provided in the same repository. |
|
rpc
|
|
|
jsonrpc/types
Module
|
|
|
Package rpctest provides a hdfd-specific RPC testing harness crafting and executing integration tests by driving a `hdfd` instance via the `RPC` interface.
|
Package rpctest provides a hdfd-specific RPC testing harness crafting and executing integration tests by driving a `hdfd` instance via the `RPC` interface. |
|
Package sampleconfig provides a single function that returns the contents of the sample configuration file for hdfd.
|
Package sampleconfig provides a single function that returns the contents of the sample configuration file for hdfd. |
|
wire
module
|