Running a Supernode

We call Supernode a node that runs both L1+L2 nodes on the same ARM64 board. This is achieved by first syncing an L1 and later an L2 client.

As a first approach, we are setting up an L1 and an Optimism node and will be adding more L2s step-by-step guides.

Warning

In order to run a Supernode you will need a 32 GB RAM board. We currently support the Orange Pi 5 Plus device. Rock 5B 32 GB will be supported any time soon.

L1 sync

The first step is to sync an L1 node. You can choose any client combo. For instance:

sudo systemctl start prysm-beacon
sudo systemctl start geth

Once synced, you can start your L2 node.

Note

We recommend to wait for the Execution Client to finish the snapshot creation phase as it consumes a lot of resources. Check the logs and make sure the client is just importing blocks.

L2 sync

Once the L1 is up and running, we can start syncing the L2 node.

Optimism

First, we need to change a couple of configuration parameters.

As we are running a Supernode, we will be using localhost as the L1 provider. Let’s set localhost as our host:

sudo sed -i 's/l1ip/localhost/' /etc/ethereum/op-node.conf
sudo sed -i 's/l1beaconip/localhost/' /etc/ethereum/op-node.conf

Now we can start both clients: op-geth and op-node.

First, start op-geth:

sudo systemctl start op-geth

Note

As op-geth is configured to use snap sync, the client needs peers to download the info. So we need to forward the port 31303 to allow peers to connect.

Once op-geth is running, we can start op-node:

sudo systemctl start op-node

The snap sync process will start shortly (and will take ~10-15 hours). Once in sync, you will be running a SuperNode.