9.1 KiB
Running nwaku
Nwaku binaries can be built and run on Linux and macOS. Windows support is experimental.
# Run with default configuration
./build/wakunode2
# See available command line options
./build/wakunode2 --help
Default configuration
By default a nwaku node will:
- generate a new private key and libp2p identities after every restart. See this tutorial if you want to generate and configure a persistent private key.
- listen for incoming libp2p connections on the default TCP port (
60000
) - enable
relay
protocol - subscribe to the default pubsub topic, namely
/waku/2/default-waku/proto
- enable
store
protocol, but only as a client. This implies that the nwaku node will not persist any historical messages itself, but can querystore
service peers who do so. To configurestore
as a service node, see this tutorial.
Note: The
filter
andlightpush
protocols are not enabled by default. Consult the configuration guide on how to configure your nwaku node to run these protocols.
Some typical non-default configurations are explained below. For more advanced configuration, see the configuration guide. Different ways to connect to other nodes are expanded upon in our connection guide.
Finding your listening address(es)
Find the log entry beginning with Listening on
.
It should be printed at INFO level when you start your node
and contains a list of all publically announced listening addresses for the nwaku node.
For example
INF 2022-05-11 16:42:30.591+02:00 Listening on topics="wakunode" tid=6661 file=wakunode2.nim:941 full=[/ip4/0.0.0.0/tcp/60000/p2p/16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H][/ip4/0.0.0.0/tcp/8000/ws/p2p/16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H]
indicates that your node is listening on the TCP transport address
/ip4/0.0.0.0/tcp/60000/p2p/16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H
and websocket address
/ip4/0.0.0.0/tcp/8000/ws/p2p/16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H
You can also query a running node for its listening addresses
using a get_waku_v2_debug_v1_info
JSON-RPC API call.
For example
curl -d '{"jsonrpc":"2.0","id":"id","method":"get_waku_v2_debug_v1_info", "params":[]}' --header "Content-Type: application/json" http://localhost:8545
returns a response similar to
{
"jsonrpc": "2.0",
"id": "id",
"result": {
"listenAddresses": [
"/ip4/0.0.0.0/tcp/60000/p2p/16Uiu2HAmLU5Nwng9dWFZwM2DgJ5QGcUuDnefJyHJiXUCVaprhgL4"
],
"enrUri": "enr:-IO4QDxToTg86pPCK2KvMeVCXC2ADVZWrxXSvNZeaoa0JhShbM5qed69RQz1s1mWEEqJ3aoklo_7EU9iIBcPMVeKlCQBgmlkgnY0iXNlY3AyNTZrMaEDdBHK1Gx6y_zv5DVw5Qb3DtSOMmVHTZO1WSORrF2loL2DdWRwgiMohXdha3UyAw"
}
}
Finding your discoverable ENR address(es)
A nwaku node can encode its addressing information in an Ethereum Node Record (ENR) according to 31/WAKU2-ENR
.
These ENR are most often used for discovery purposes.
ENR for DNS discovery
Find the log entry beginning with DNS: discoverable ENR
.
It should be printed at INFO level when you start your node with DNS discovery enabled
and contains an ENR that can be added to node lists discoverable via DNS.
For example
INF 2022-05-20 11:52:48.772+02:00 DNS: discoverable ENR topics="wakunode" tid=5182 file=wakunode2.nim:941 enr=enr:-Iu4QBZs5huNuEAjI9WA0HOAjzpmp39vKJAtYRG3HXH86-i3HGcxMgupIkyDBmBq9qJ2wFfgMiW8AUzUxTFMAzfJM5MBgmlkgnY0gmlwhAAAAACJc2VjcDI1NmsxoQN0EcrUbHrL_O_kNXDlBvcO1I4yZUdNk7VZI5GsXaWgvYN0Y3CC6mCFd2FrdTID
indicates that your node addresses are encoded in the ENR
enr=enr:-Iu4QBZs5huNuEAjI9WA0HOAjzpmp39vKJAtYRG3HXH86-i3HGcxMgupIkyDBmBq9qJ2wFfgMiW8AUzUxTFMAzfJM5MBgmlkgnY0gmlwhAAAAACJc2VjcDI1NmsxoQN0EcrUbHrL_O_kNXDlBvcO1I4yZUdNk7VZI5GsXaWgvYN0Y3CC6mCFd2FrdTID
ENR for Discovery v5
Find the log entry beginning with Discv5: discoverable ENR
.
It should be printed at INFO level when you start your node with Waku Discovery v5 enabled
and contains the ENR that will be discoverable by other peers.
For example
INF 2022-05-20 11:52:48.775+02:00 Discv5: discoverable ENR topics="wakunode" tid=5182 file=wakunode2.nim:905 enr=enr:-IO4QDxToTg86pPCK2KvMeVCXC2ADVZWrxXSvNZeaoa0JhShbM5qed69RQz1s1mWEEqJ3aoklo_7EU9iIBcPMVeKlCQBgmlkgnY0iXNlY3AyNTZrMaEDdBHK1Gx6y_zv5DVw5Qb3DtSOMmVHTZO1WSORrF2loL2DdWRwgiMohXdha3UyAw
indicates that your node addresses are encoded in the ENR
enr=enr:-IO4QDxToTg86pPCK2KvMeVCXC2ADVZWrxXSvNZeaoa0JhShbM5qed69RQz1s1mWEEqJ3aoklo_7EU9iIBcPMVeKlCQBgmlkgnY0iXNlY3AyNTZrMaEDdBHK1Gx6y_zv5DVw5Qb3DtSOMmVHTZO1WSORrF2loL2DdWRwgiMohXdha3UyAw
Typical configuration (relay node)
The typical configuration for a nwaku node is to run the relay
protocol,
subscribed to the default pubsub topic /waku/2/default-waku/proto
,
and connecting to one or more existing peers.
We assume below that running nodes also participate in Discovery v5
to continually discover and connect to random peers for a more robust mesh.
Connecting to known peer(s)
A typical run configuration for a nwaku node is to connect to existing peers with known listening addresses using the --staticnode
option.
The --staticnode
option can be repeated for each peer you want to connect to on startup.
This is also useful if you want to run several nwaku instances locally
and therefore know the listening addresses of all peers.
As an example, consider a nwaku node that connects to two known peers
on the same local host (with IP 0.0.0.0
)
with TCP ports 60002
and 60003
,
and peer IDs 16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H
and 16Uiu2HAmFBA7LGtwY5WVVikdmXVo3cKLqkmvVtuDu63fe8safeQJ
respectively.
The Discovery v5 routing table can similarly be bootstrapped using a static ENR.
We include an example below.
./build/wakunode2 \
--ports-shift:1 \
--staticnode:/ip4/0.0.0.0/tcp/60002/p2p/16Uiu2HAkzjwwgEAXfeGNMKFPSpc6vGBRqCdTLG5q3Gmk2v4pQw7H \
--staticnode:/ip4/0.0.0.0/tcp/60003/p2p/16Uiu2HAmFBA7LGtwY5WVVikdmXVo3cKLqkmvVtuDu63fe8safeQJ \
--discv5-discovery:true \
--discv5-bootstrap-node:enr:-JK4QM2ylZVUhVPqXrqhWWi38V46bF2XZXPSHh_D7f2PmUHbIw-4DidCBnBnm-IbxtjXOFbdMMgpHUv4dYVH6TgnkucBgmlkgnY0gmowhCJ6_HaJc2VjcDI1NmsxoQM06FsT6EJ57mzR_wiLu2Bz1dER2nUFSCpaFzCccQtnhYN0Y3CCdl-DdWRwgiMohXdha3UyDw
Tip:
--ports-shift
shifts all configured ports forward by the configured amount. This is another useful option when running several nwaku instances on a single machine and would like to avoid port clashes without manually configuring each port.
Connecting to the wakuv2.prod
network
See this explainer on the different networks and Waku v2 fleets.
You can use DNS discovery to bootstrap connection to the existing production network. Discovery v5 will attempt to extract the ENRs of the discovered nodes as bootstrap entries to the routing table.
./build/wakunode2 \
--ports-shift:1 \
--dns-discovery:true \
--dns-discovery-url:enrtree://ANEDLO25QVUGJOUTQFRYKWX6P4Z4GKVESBMHML7DZ6YK4LGS5FC5O@prod.wakuv2.nodes.status.im \
--discv5-discovery:true
Connecting to the wakuv2.test
network
See this explainer on the different networks and Waku v2 fleets.
You can use DNS discovery to bootstrap connection to the existing test network. Discovery v5 will attempt to extract the ENRs of the discovered nodes as bootstrap entries to the routing table.
./build/wakunode2 \
--ports-shift:1 \
--dns-discovery:true \
--dns-discovery-url:enrtree://AO47IDOLBKH72HIZZOXQP6NMRESAN7CHYWIBNXDXWRJRZWLODKII6@test.wakuv2.nodes.status.im \
--discv5-discovery:true
Typical configuration (relay and store service node)
Often nwaku nodes choose to also store historical messages
from where it can be queried by other peers who may have been temporarily offline.
For example, a typical configuration for such a store service node,
connecting to the wakuv2.test
fleet on startup,
appears below.
./build/wakunode2 \
--ports-shift:1 \
--store:true \
--persist-messages:true \
--db-path:/mnt/nwaku/data/db1/ \
--store-capacity:150000 \
--dns-discovery:true \
--dns-discovery-url:enrtree://AO47IDOLBKH72HIZZOXQP6NMRESAN7CHYWIBNXDXWRJRZWLODKII6@test.wakuv2.nodes.status.im \
--discv5-discovery:true
See our store configuration tutorial for more.
Interact with a running nwaku node
A running nwaku node can be interacted with using the Waku v2 JSON RPC API.
Note: Private and Admin API functionality are disabled by default. To configure a nwaku node with these enabled, use the
--rpc-admin:true
and--rpc-private:true
CLI options.