Node Setup Manual
How to Operate Tellor Node.
Recommended Hardware Specs
Operating a node for a personal RPC can be done usin most modern computers. Even modest hardware should be fine for this purpose.
Operating a validator:
Modern cpu with at least 8 cores / threads
ram: 32 gb + (16gb swap space recommended)
storage: 1000gb+ @ NVME gen4
network: 500mb/s DL, 100mb/s UL (the faster the better)
Software Prerequisites
jq, yq, sed, curl, wget, make, and Go are required for running the various commands and config scripts and commands in this guide:
sudo apt install jq yq sed curl wget build-essentialInstall go if you want to build binaries from the layer repo:
Go version 1.22: Use the default install instructions here.
jq, yq, sed, curl, wget, make, and Go are required for running the various commands and config scripts and commands in this guide:
brew install jq yq sed wget && xcode-select --installGo ≥ 1.22: Use the default install instructions here.
Commands shown should just work while logged in as a user (using root is not recommended).
If you are using an older Mac with an intel chip, the linux versions (amd64) in step 1 below may be used. (just remember to use the mac commands!)
If on raspberry pi or similar, use the binary downloads for "arm64".
Sync Method
These instructions use state sync: your node downloads a recent chain state snapshot from peers. This is the fastest way to get online, but you will not be able to query block info (like transactions) for blocks produced before the day of your sync.
1. Download and Organize the layerd Binary
As you progress through the steps, be sure to select the tabs that work for your setup! You will get errors if you use the linux commands on mac and vice-versa.
First, download the binary from the Tellor Github.
Initialize .layer folder in your home directory:
2. Set System Variables
A Layer node uses the following variables:
TOKEN_BRIDGE_V2_ADDRESS: the token bridge contract address.
Set the environment variables so that they are set in new terminal windows by default. Open your .bashrc or .zshrc file with a text editor like nano:
Add these lines to the bottom of the file. Remember to replace the example ETH_RPC_URL with your actual Ethereum RPC url, and if you're going to run a reporter, replace the REPORTERS_VALIDATOR_ADDRESS with your own as well.
Exit nano with ctrl^x then enter y to save the changes.
Load the new variables:
3. Edit Chain Configuration for Tellor.
We need to open up the tellor layer config files and change some variables. You can use any local text editor like code, vim, or nano.
Note: All other variables can be safely left as is.
In ~/.layer/config/app.toml:
In ~/.layer/config/client.toml:
In ~/.layer/config/config.toml:
4. Sync the Node
Before starting your node, it's a good idea to think about how you want to run it so that the process does not get killed accidentally. This is not obvious for beginners. Try GNU screen or tmux. More advanced setups can be achieved using systemd services.
We need to make a few more config edits to make sure your state sync goes smoothly.
To find a good trusted height to use for a snapshot sync, we need to find the height of a snapshot available from
https://mainnet.tellorlayer.com/rpc/. Copy and paste this entire block of commands into a terminal and hit enter:
The output should be something like:
Edit config.toml:
Open your config file:
Scroll or search (ctrl^w) the file and edit the state sync variables shown here to match the trusted height and trusted hash you found above:
Be sure to replace the trust_height and trust_hash with the block number and hash from the curl command above.
Exit nano with ctrl^x then enter y to save the changes.
Start your node:
The node should start up quickly and begin downloading a state snapshot from peers.
Some errors related to peer connections can be expected even if the snapshot sync is working properly. (e.g. "we need more peers", or "Failed to reconnect")
To check if the node is fully synced, open a separate terminal window and run:
You should see a JSON formatted list of information about your running node. If you see catching_up":false that means that you're node is fully synced and ready to use!
Last updated
Was this helpful?