Bee is the second official Ethereum Swarm implementation. This project is in very early stage and under active development.

No compatibility with the first Ethereum Swarm implementation is provided, mainly because the change in underlaying protocol from devp2p to libp2p. A bee node cannot join swarm network and vice versa.

API compatibility will be guaranteed when version 1.0 is released, but not before that.

Installing from source by checking the git repository:

git clone https://github.com/ethersphere/bee
cd bee
make binary
cp dist/bee /usr/local/bin/bee

Bee node provides CLI help that lists all available commands and flags. Every command has its own help.

bee -h
bee start -h

To run the node with the default configuration:

bee start

This command starts bee node including HTTP API for user interaction and P2P API for communication between bee nodes.

It will open an interactive prompt asking for a password which protects node private keys. Interactive prompt can be avoided by providing a CLI flag --password-file with path to the file that contains the password, just to pass it as the value to --password flag. These values are also possible to be set with environment variables or configuration file.

Configuration parameters can be passed to the bee node by:

• command line arguments
• environment variables
• configuration file

with that order of precedence.

Available command line parameters can be seen by invoking help flag bee start -h.

Environment variables are analogues to these flags and can be constructed with simple rules:

• remove -- prefix from the flag name
• capitalize all characters in the flag name
• replace all - characters with _
• prepend BEE_ prefix to the resulted string

For example, cli flag --api-addr has an analogues env variable BEE_API_ADDR

Configuration file path is by default $HOME/.bee.yaml, but it can be changed with --config cli flag or BEE_CONFIG environment variable.

Configuration file variables are all from the bee start -h help page, just without the -- prefix. For example, --api-addr and --data-dir can be specified with configuration file such as this one:

api-addr: 127.0.0.1:8085
data-dir: /data/bees/bee5

It is possible to start multiple, persistent, completely independent, bee nodes on a single running operating system. This can be achieved with less complexity with prepared configuration files for every node.

An example configuration for the first node would be:

api-addr: :8081
p2p-addr: :7071
debug-api-addr: :6061
enable-debug-api: true
data-dir: /tmp/bee/node1
password: some pass phze
verbosity: trace
tracing: true

Save a file named node1.yaml with this content, and create as many as you need by incrementing the number in in filename and api-addr, p2p-addr, debug-api-addr and data-dir.

For example, file node2.yaml should have this content:

api-addr: :8082
p2p-addr: :7072
debug-api-addr: :6062
enable-debug-api: true
data-dir: /tmp/bee/node2
password: some pass phze
verbosity: trace
tracing: true

The first node address will be used for other nodes to discover themselves. It is usually referred as the bootnode address.

bee --config node1.yaml start

When the node starts it will print out some of its p2p addresses in a multiaddress form like this: /ip4/127.0.0.1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM. This address is the bootnonde address.

Other nodes should be started by providing the bootnode address to them, so that they can connect to each other:

bee --config node2.yaml start --bootnode /ip4/127.0.0.1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM
bee --config node3.yaml start --bootnode /ip4/127.0.0.1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM
bee --config node4.yaml start --bootnode /ip4/127.0.0.1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM
...

Every node can provide its overlay address and underlay addresses by reading the logs when the node starts or through Debug API:

curl localhost:6060/addresses

It will return a response with addresses:

{
  "overlay": "7ecf777fdab7553fbc4db7f265a7bd80d27b171babe931bc61e9d7966974ef47",
  "underlay": [
    "/ip6/::1/udp/7071/quic/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM",
    "/ip4/127.0.0.1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM",
    "/ip4/127.0.0.1/udp/7071/quic/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM",
    "/ip6/::1/tcp/7071/p2p/16Uiu2HAm2LXfYsY9pXtgGdQ8oPb3bAkxwpfBE6AMzcscH1UkQLZM",
  ]
}

To check if two nodes are connected and to see the round trip time for message exchange between them, get the overlay address from one node, for example local node 2:

curl localhost:6062/addresses

And use that address in the API call on another node, for example, local node 1:

curl -XPOST localhost:8081/pingpong/d4440baf2d79e481c3c6fd93a2014d2e6fe0386418829439f26d13a8253d04f1

To process protocol buffer files and generate the Go code from it two tools are needed:

• protoc - https://github.com/protocolbuffers/protobuf/releases
• protoc-gen-gogofaster - https://github.com/gogo/protobuf

Makefile rule protobuf can be used to automate protoc-gen-gogofaster installation and code generation:

make protobuf

Please read the coding guidelines.

This library is distributed under the BSD-style license found in the LICENSE file.