Fablo supports:
- Environment: Docker
- RAFT and solo consensus protocols
- Multiple organizations and channels
- Chaincode installation and upgrade
- REST API client for CA and chaincodes (Fablo REST)
- Blockchain Explorer which can be enabled for each organization
Fablo is distributed as a single shell script which uses Docker image to generate the network config. You may keep the script in the root directory of your project or install it globally in your file system.
To install it globally:
sudo curl -Lf https://github.com/hyperledger-labs/fablo/releases/download/2.0.0/fablo.sh -o /usr/local/bin/fablo && sudo chmod +x /usr/local/bin/fablo
To get a copy of Fablo for a single project, execute in the project root:
curl -Lf https://github.com/hyperledger-labs/fablo/releases/download/2.0.0/fablo.sh -o ./fablo && chmod +x ./fablo
To create a local Hyperledger Fabric network with Node.js chaincode and REST API client, install Fablo and execute:
./fablo init node rest
./fablo up
After a few minutes the whole network will be set up and running.
You can check the running nodes via docker ps
or docker stats
, and you can query the network with command line (via cli.org1.example.com
container) or REST API client (via Fablo REST).
fablo up /path/to/fablo-config.json
The up
command creates initial configuration and starts Hyperledger Fabric network on Docker.
In this case network configuration is saved in $(pwd)/fablo-target
.
Then you can manage the network with other commands, for example stop
, start
, down
, prune
.
Provided Fablo configuration file describes network topology: root organization, other organizations, channels and chaincodes. See the samples or Fablo config section.
There are two basic use cases.
You may use Fablo to start and manage the network for development purposes, test different network topologies, run it in CI environment etc.
In this case you should keep fablo-target
directory intact and out of the version control.
Fablo will manage it locally.
On the other hand you can use Fablo to generate initial network configuration, keep it in version control and tweak for specific requirements.
In this case, however, you should use generated fablo-docker.sh
instead of fablo
script.
fablo init [node] [rest] [dev]
Creates simple network config file in current dir. Good step to start your adventure with Fablo or set up a fast prototype.
Fablo init
command takes three parameters (the order does not matter):
- Option
node
makes Fablo to generate a sample Node.js chaincode as well. - Option
rest
enables simple REST API with Fablo REST as standalone Docker container. - Option
dev
enables running peers in dev mode (so the hot reload for chaincode is possible).
Sample command:
fablo init node dev
Generated fablo-config.json
file uses single node Solo consensus and no TLS support.
This is the simplest way to start with Hyperledger Fabric, since Raft consensus requires TLS and TLS itself adds a lot of complexity to the blockchain network and integration with it.
fablo generate [/path/to/fablo-config.json|yaml [/path/to/fablo/target]]
Generates network configuration files in the given directory.
Default config file path is $(pwd)/fablo-config.json
or \$(pwd)/fablo-config.yaml
, default directory is $(pwd)/fablo-target
.
If you specify a different directory, you loose Fablo support for other commands.
If you want to use Fablo only to kick off the Hyperledger Fabric network, you may provide target directory parameter or just copy generated Fablo target directory content to desired directory and add it to version control. Note that generated files may contain variables with paths on your disk and generated crypto material for Hyperledger Fabric. Review the files before submitting to version control system.
fablo up [/path/to/fablo-config.json|yaml]
Starts the Hyperledger Fabric network for given Fablo configuration file, creates channels, installs and instantiates chaincodes.
If there is no configuration, it will call generate
command for given config file.
fablo [down | start | stop]
Downs, starts or stops the Hyperledger Fabric network for configuration in the current directory. This is similar to down, start and stop commands for Docker Compose.
fablo prune
Downs the network and removes fablo-target
directory.
fablo reset
fablo recreate [/path/to/fablo-config.json|yaml]
reset
-- down and up steps combined. Network state is lost, but the configuration is kept intact. Useful in cases when you want a fresh instance of network without any state.recreate
-- prunes the network, generates new config files and ups the network. Useful when you editedfablo-config
file and want to start newer network version in one command.
fablo validate [/path/to/fablo-config.json|yaml]
Validates network config. This command will validate your network config try to suggest necessary changes or additional tweaks.
Please note that this step is also executed automatically before each generate
to ensure that at least critical errors where fixed.
Fablo supports saving state snapshot (backup) of the network and restoring it. It saves all network artifacts, certificates, and the data of CA, orderer and peer nodes. Note the snapshot does not contain Fablo config file and chaincode source code, since both can be located outside Fablo working directory.
Snapshotting might be useful if you want to keep the current state of a network for future use (for testing, sharing the network state, courses and so on).
fablo snapshot <target-snapshot-path>
If you want to restore snapshot into current directory, execute:
fablo restore <source-snapshot-path>
Example:
- Assume you have a working network with some state.
- Execute
./fablo snapshot /tmp/my-snapshot
. It will create a file/tmp/my-snapshot.fablo.tar.gz
with the state of the network. It is not required to stop the network before making a snapshot. - Execute
./fablo prune
to destroy the current network. If the network was present, Fablo would not be able to restore the new one from backup. - Execute
./fablo restore /tmp/my-snapshot
to restore the network. - Execute
./fablo start
to start the restored network.
Typically, a snapshot of the network with little data will take less than 1 MB, so it is easy to share.
The script fabric-docker.sh
is generated among docker network configuration.
It does not support generate
command, however other commands work in same way as in fablo
.
Basically fablo
forwards some commands to this script.
If you want to use Fablo for network configuration setup only, then the fabric-docker.sh
file allows you to manage the network.
fablo chaincodes install
Install all chaincodes. Might be useful if for some reason, Fablo won't manage to do it by itself.
If you want to install a single chaincode defined in Fablo config file, execute:
fablo chaincode install <chaincode-name> <version>
fablo chaincode upgrade <chaincode-name> <version>
Upgrades chaincode with given name on all relevant peers. Chaincode directory is specified in Fablo config file.
Invokes chaincode with specified parameters.
fablo chaincode invoke <channel-name> <chaincode-name> <peers-domains-comma-separated> <command> [transient]
Sample command:
fablo chaincode invoke "my-channel1" "chaincode1" "peer0.org1.example.com" '{"Args":["KVContract:put", "name", "Willy Wonka"]}'
Gets the instantiated or installed chaincodes in the specified channel or peer.
fablo chaincodes list <peer> <channel>
Hyperledger Fabric allows to run peers in dev mode in order to allow simple develop of chaincodes. In this case chaincodes do not need to be upgraded each time, but they are run locally. This feature allows hot reload of chaincode code and speeds up the development a lot.
Fablo will run peers in dev mode when global.peerDevMode
is set to true
.
Note: in this case TLS has to be disabled, otherwise config validation fails.
The simplest way of trying Fablo with dev mode is as follows:
- Execute
fablo init node dev
. It will initialize Fablo config file with sample node chaincode and dev mode enabled. In this case Fablo config file hasglobal.peerDevMode
set totrue
, and thepackage.json
file for sample Node.js chaincode has a script for running chaincode in dev mode (start:dev
). - Start the network with
fablo up
. Because dev mode is enabled, chaincode containers don't start. Instead, Fablo approves and commits chaincode definitions from Fablo config file. - Npm install and start the sample chaincode with:
Now, when you update the chaincode source code, it will be automatically refreshed on Hyperledger Fabric Network.
(cd chaincodes/chaincode-kv-node && nvm use && npm i && npm run start:watch)
Our sample chaincode definition contains some scripts for running chaincode in dev mode:
"scripts": {
...
"start:dev": "fabric-chaincode-node start --peer.address \"127.0.0.1:8541\" --chaincode-id-name \"chaincode1:0.0.1\" --tls.enabled false",
"start:watch": "nodemon --exec \"npm run start:dev\"",
...
},
Worth considering:
- If you want chaincode to be running on multiple peers, you need to start it multiple times, specifying different
--peer.address
- In case of errors ensure you have the same
--chaincode-id-name
asCC_PACKAGE_ID
in Fablo output.
Feel free to update this scripts to adjust it to your chaincode definition.
fablo channel --help
Use it to list all available channel commands.
Commands are generated using fablo-config.json to cover all cases (queries for each channel. organization and peer)
fablo channel list org1 peer0
Lists all channels for given peer.
fablo channel getinfo channel_name org1 peer0
Prints channel info, like current block height for given peer
fablo channel fetch config channel_name org1 peer0 [file_name.json]
Fetches latest config block, decodes it and write to a JSON file.
fablo channel fetch <oldest|newest|block-number> channel_name org1 peer0 [file_name.json]
Fetches oldest, newest or a block with given number, and writes it to a file.
fablo version [--verbose | -v]
Prints current Fablo version.
With optional -v
or --verbose
flag it prints supported Fablo and Hyperledger Fabric versions as well.
fablo use
Lists all available Fablo versions.
fablo use <version-number>
Switches current script to selected version.
Fablo config is a single JSON or YAML file that describes desired Hyperledger Fabric network topology (network settings, CA, orderer, organizations, peers, channels, chaincodes).
It has to be compatible with the [schema].
You may generate a basic config with ./fablo init
command.
See the samples directory for more complex examples.
The basic structure of Fablo config file is as follows:
{
"$schema": "https://github.com/hyperledger-labs/fablo/releases/download/2.0.0/schema.json",
"global": { ... },
"orgs": [ ... ],
"channels": [ ... ],
"chaincodes": [ ... ]
}
Example:
"global": {
"fabricVersion": "2.4.2",
"tls": false,
"peerDevMode": false,
"monitoring": {
"loglevel": "debug"
},
"tools": {
"explorer": false
}
},
Example:
"orgs": [
{
"organization": {
"name": "Org1",
"domain": "org1.example.com"
},
"peer": {
"instances": 2,
"db": "LevelDb"
},
"orderers": [{
"groupName": "group1",
"type": "raft",
"instances": 3
}],
"tools": {
"fabloRest": true,
"explorer": true
}
},
...
],
The other available parameters for entries in orgs
array are:
organization.mspName
(default:organization.name + 'MSP'
)ca.prefix
(default:ca
)ca.db
(default:sqlite
, other:postgres
)peer.prefix
(default:peer
)peer.anchorPeerInstances
(default:1
)orderers
(defaults to empty:[]
)tools.explorer
- whether run Blockchain Explorer for the organization (default:false
)tools.fabloRest
- whether run Fablo REST for the organization (default:false
)
- may be
LevelDb
(default) orCouchDb
.
- is optional as some organizations may have orderer defined, but some don't.
- At least one orderer group is required to run Fabric network (requirement is validated before run).
- If you want to spread orderers in group between many organizations use same
groupName
in every group definition. - The property
orderers.type
may besolo
orraft
. We do not support the Kafka orderer.
Example:
"channels": [
{
"name": "my-channel1",
"groupName": "group1",
"orgs": [
{
"name": "Org1",
"peers": [
"peer0",
"peer1"
]
},
{
"name": "Org2",
"peers": [
"peer0"
]
}
]
},
...
],
- Property
groupName
is optional (defaults to first orderer group found). If you want to handle channel with different orderer group define it inorgs
and pass it's name here.
Example:
"chaincodes": [
{
"name": "chaincode1",
"version": "0.0.1",
"lang": "node",
"channel": "my-channel1",
"directory": "./chaincodes/chaincode-kv-node",
"privateData": {
"name": "org1-collection",
"orgNames": ["Org1"]
}
},
{
"name": "chaincode2",
"version": "0.0.1",
"lang": "java",
"channel": "my-channel2"
}
]
The other available parameters for entries in chaincodes
array are:
init
- initialization arguments (for Hyperledger Fabric below 2.0; default:{"Args":[]}
)initRequired
- whether the chaincode requires initialization transaction (for Hyperledger Fabric 2.0 and greater; default:false
)endorsement
- the endorsement policy for the chaincode (in case of missing value for Hyperledger Fabric 2.0 and greater there is no default value - Hyperledger by default will take the majority of organizations; for Hyperledger Fabric below 2.0 Fablo generates an endorsement policy where all organizations need to endorse)
The property lang
may be golang
, java
or node
.
The privateData
parameter is optional. You don't need to define the private data collection for the chaincode. By default there is none (just the implicit private data collection in Fabric 2.x).
Hooks in Fablo are Bash commands to be executed after a specific event.
Right now Fablo supports only one kind of hook: postGenerate
.
It will be executed each time after the network config is generated -- after ./fablo generate
command (executed separately or automatically by ./fablo up
).
The following hook example will change MaxMessageCount
to 1 in generated Hyperledger Fabric config:
"hooks": {
"postGenerate": "perl -i -pe 's/MaxMessageCount: 10/MaxMessageCount: 1/g' \"./fablo-target/fabric-config/configtx.yaml\""
}
Genrated Hooks are saved in fablo-target/hooks
.
---
"$schema": https://github.com/hyperledger-labs/fablo/releases/download/2.0.0/schema.json
global:
fabricVersion: 2.4.2
tls: false
orgs:
- organization:
name: Orderer
domain: root.com
orderers:
- groupName: group1
prefix: orderer
type: solo
instances: 1
- organization:
name: Org1
domain: org1.example.com
tools:
fabloRest: true
explorer: true
peer:
instances: 2
- organization:
name: Org2
domain: org2.example.com
peer:
instances: 1
channels:
- name: my-channel1
orgs:
- name: Org1
peers:
- peer0
- peer1
- name: Org2
peers:
- peer0
chaincodes:
- name: and-policy-chaincode
version: 0.0.1
lang: node
channel: my-channel1
init: '{"Args":[]}'
endorsement: AND('Org1MSP.member', 'Org2MSP.member')
directory: "./chaincodes/chaincode-kv-node"
privateData:
- name: org1-collection
orgNames:
- Org1
TODO
Fablo will generate the connection profiles for each organization defined in the configuration.
You can find them in fablo-target/fablo-config/connection-profiles
directory in json
and yaml
format.
Fablo is integrated with simple REST API for CA and chaincodes, supported by Fablo REST.
If you want to use it, provide for your organization "tools": { "fabloRest": true }
.
Visit the Fablo REST project for more documentation.
Fablo can run Blockchain Explorer for you.
Provide for your organization "tools": { "explorer": true }
, if you want to use it per organization, or provide the same value in global
section of the config, if you want to use one global Explorer for all organizations.
We'd love to have you contribute! Please refer to our contribution guidelines for details.
Fablo was originally created at SoftwareMill by @Hejwo and @dzikowski. In December 2021, Fablo joined Hyperledger Labs.