Storage Provider Tutorial

This guide provides practical steps for the storage providers to start an es-node instance for connecting to the existing EthStorage testnet.

Before Starting

Minimum Hardware Requirements

  • CPU: A minimum of 4 cores and 8 threads

  • 8GB of RAM

  • Disk:

    • We recommend using an NVMe disk to support the full speed of sampling

    • At least 550 GB of available storage space for the runtime and sync of one data shard

  • Internet: At least 8MB/sec download speed

System Environment

  • MacOS Version 14+, Ubuntu 20.04+, or Windows with WSL (Windows Subsystem for Linux) version 2

  • (Optional) Docker 24.0.5+ (would simplify the process)

  • (Optional) Go 1.21+ and Node.js 16+ (can be installed following the steps)

ℹ️ Note: The steps assume the use of the root user for all command line operations. If using a non-root user, you may need to prepend some commands with "sudo".

Preparing miner and signer account

We recommend preparing two specific Ethereum accounts for this test.

  • The miner account will be set to receive rewards once the storage provider successfully submits a storage proof to the EthStorage contract. Each storage provider must use a unique miner account.

  • The signer account will act as a transaction signer and should contain a balance of test ETH. A signer account can be used by multiple storage providers.

ℹ️ Note: As Sepolia is used as L1 for the testnet, the test ETH can be requested from https://sepoliafaucet.com/.

⚠️ Warning: For safety reasons, we strongly suggest creating new wallets for the accounts to avoid the loss of any personal assets.

Remember to use the signer's private key (with ETH balance) to replace <private_key> in the following steps. And use the other address to replace <miner>.

Preparing Ethereum API endpoints

During the operation of the ES-Node, frequent Ethereum API calls are made, including at the execution layer and the consensus layer (the beacon chain). Therefore, we need you to prepare endpoints for two types of calls. We recommend using BlockPI for the execution layer endpoint and QuickNode for the beacon endpoint.

For details on the application process for endpoints, please refer to this section.

In the following tutorial, you will need to replace <el_rpc> for you execution layer endpoint, and <cl_rpc> for the beacon endpoint.

About run.sh and init.sh

The run.sh script serves as the entry point for launching the es-node with predefined parameters. By default, mining is enabled through the --miner.enabled flag in run.sh, which implies that you assume the role of a storage provider upon starting an es-node with the default settings.

However, before the es-node can be successfully launched, you must execute init.sh first. The primary function of this script is to verify the system environment, download and install dependencies, and initialize the data files in preparation for mining.

For specific usage and examples of the two scripts, refer to the steps outlined in Options for running es-node.

ℹ️ Note: Some of the flags/parameters used in run.sh are supposed to change over time. Refer to configuration for a full list.

Mining multiple shards

By default, only the first shard (shard 0) is mined using the default options in the scripts, but you have the choice to initialize and run your es-node in order to mine multiple selected shards by using additional options.

The flag --shard_index can be utilized multiple times with init.sh to generate data files for multiple shards on the es-node. For example,

env ES_NODE_STORAGE_MINER=0x123...ab ./init.sh --shard_index 1 --shard_index 2

Please take note of the following:

  • The shard files will be generated in the ./es-data directory with the naming convention shard-$(shard_index).dat by default settings in init.sh.

  • A shard will be omitted if its corresponding data file already exists.

  • shard-0.dat will be tried to create if no --shard_index is specified.

After initialization in this way, the run.sh script will attempt to operate on data files located in ./es-data/shard-*.dat. If you have relocated these data files or added additional files in another location, you can specify them using the --storage.files options repeatedly following ./run.sh.

About the option of zk prover implementation

The --miner.zk-prover-impl flag specifies the type of zkSNARK implementation. Its default value is 1, indicating the generation of zk proofs using snarkjs. The option 2 means to utilize go-rapidsnark. Since --miner.zk-prover-impl interacts closely with the environment, it is crucial to use the same configuration when running both init.sh and run.sh.

ℹ️ Note: If you have to run an es-node pre-built with --miner.zk-prover-impl 2 on Ubuntu 20.04, you will need to install extra packages.

Options for running es-node

You can run es-node from a pre-built executable, a pre-built Docker image, or from the source code.

From pre-built executables

Before running es-node from the pre-built executables, ensure that you have installed Node.js and snarkjs, unless --miner.zk-prover-impl flag is set to 2.

ℹ️ Note: Ensure that you run the executables on WSL 2 if you are using Windows, and both Node.js and snarkjs are installed on WSL instead of Windows.

Download the pre-built package suitable for your platform:

  • Linux x86-64 or WSL:

curl -L https://github.com/ethstorage/es-node/releases/download/v0.1.16/es-node.v0.1.16.linux-amd64.tar.gz | tar -xz

  • MacOS x86-64:

curl -L https://github.com/ethstorage/es-node/releases/download/v0.1.16/es-node.v0.1.16.darwin-amd64.tar.gz | tar -xz

  • MacOS ARM64:

curl -L https://github.com/ethstorage/es-node/releases/download/v0.1.16/es-node.v0.1.16.darwin-arm64.tar.gz | tar -xz

In folder es-node.v0.1.16, init es-node by running:

env ES_NODE_STORAGE_MINER=<miner> ./init.sh --l1.rpc <el_rpc>

Run es-node

env ES_NODE_STORAGE_MINER=<miner> ES_NODE_SIGNER_PRIVATE_KEY=<private_key> ./run.sh --l1.rpc <el_rpc> --l1.beacon <cl_rpc>

From a Docker image

First init an es-node environment with the following command (If you are using Windows, execute the command in WSL):

docker run --rm \
          -v ./es-data:/es-node/es-data \
          -v ./zkey:/es-node/build/bin/snark_lib/zkey \
          -e ES_NODE_STORAGE_MINER=<miner> \
          --entrypoint /es-node/init.sh \
          ghcr.io/ethstorage/es-node:v0.1.16 \
          --l1.rpc <el_rpc>

Then start an es-node container:

docker run --name es -d \
          -v ./es-data:/es-node/es-data \
          -v ./zkey:/es-node/build/bin/snark_lib/zkey \
          -e ES_NODE_STORAGE_MINER=<miner> \
          -e ES_NODE_SIGNER_PRIVATE_KEY=<private_key> \
          -p 9545:9545 \
          -p 9222:9222 \
          -p 30305:30305/udp \
          --entrypoint /es-node/run.sh \
          ghcr.io/ethstorage/es-node:v0.1.16 \
          --l1.rpc <el_rpc> \
          --l1.beacon <cl_rpc>

After launch, you can check docker logs using the following command:

docker logs -f es

Mount data location using Docker volume option

Docker volumes (-v) are a mechanism for storing data outside containers. In the above docker run command , you have the flexibility to modify the data file location on your host machine, ensuring that the disk space requirements are fulfilled. For example:

docker run --name es  -d  \
          -v /host/path/with/large/space:/es-node/es-data \
          ...

ℹ️ Note: The absolute host path does not function well on Windows, for more details please refer here

From source code

You will need to install Go to build es-node from source code.

If you intend to build es-node on Ubuntu, be sure to verify some dependencies.

Just like running a pre-built, if you plan to utilize the default zkSNARK implementation, ensure that you have installed Node.js and snarkjs.

Now download source code and switch to the latest release branch:

git clone https://github.com/ethstorage/es-node.git
cd es-node
git checkout v0.1.16

Build es-node:

make

Init es-node

env ES_NODE_STORAGE_MINER=<miner> ./init.sh --l1.rpc <el_rpc>

Start es-node

env ES_NODE_STORAGE_MINER=<miner> ES_NODE_SIGNER_PRIVATE_KEY=<private_key> ./run.sh --l1.rpc <el_rpc> --l1.beacon <cl_rpc>

With source code, you also have the option to build a Docker image by yourself and run an es-node container:

# init
env ES_NODE_STORAGE_MINER=<miner> docker-compose run --rm --entrypoint "/es-node/init.sh" node

# start
env ES_NODE_STORAGE_MINER=<miner> ES_NODE_SIGNER_PRIVATE_KEY=<private_key> docker-compose up

If you want to run Docker container in the background and keep all the logs:

env ES_NODE_STORAGE_MINER=<miner> ES_NODE_SIGNER_PRIVATE_KEY=<private_key> ./run-docker.sh

Applying for Ethereum API endpoints

Applying for a free Sepolia execution layer endpoint from BlockPI

Go to the BlockPI website, click Get Started. After signing in, you will get your Free Package Gift. Click Generate API Key, and remember to select Sepolia, and you will get your API endpoint.

Finally, access the detailed page of your API endpoint and activate the Archive Mode under the Advanced Features section.

ℹ️ Note: The free plan of BlockPI provides sufficient usage as execution layer RPC for es-node, but it cannot be used as a beacon endpoint.

Applying for a free Sepolia beacon endpoint from QuickNode

Go to the QuickNode website, click Get started for free. After signing in, you can create an endpoint. Remember to select Ethereum and Sepolia to continue. In the Compliance & Safety category, select Endpoint Armor, and select the free plan. After completing the required information, you will receive the endpoint along with an API key.

ℹ️ Note: The free plan of QuickNode is adequate for use as a beacon endpoint (<cl_rpc>) for running es-node, but is NOT sufficient as execution layer endpoint.

Install dependencies

ℹ️ Note: Not all steps in this section are required; they depend on your choice.

Install Go

Download a stable Go release, e.g., go1.21.4.

curl -OL https://golang.org/dl/go1.21.4.linux-amd64.tar.gz

Extract and install

tar -C /usr/local -xf go1.21.4.linux-amd64.tar.gz

Update $PATH

echo "export PATH=$PATH:/usr/local/go/bin" >> ~/.profile
source ~/.profile

Install Node.js

Install Node Version Manager

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.3/install.sh | bash

Close and reopen your terminal to start using nvm or run the following to use it now:

export NVM_DIR="$HOME/.nvm"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh"  # This loads nvm
[ -s "$NVM_DIR/bash_completion" ] && \. "$NVM_DIR/bash_completion"  # This loads nvm bash_completion

Choose a Node.js version above 16 (e.g. v20.*) and install

nvm install 20

Activate the Node.js version

nvm use 20

Install snarkjs

npm install -g snarkjs

Install RapidSNARK dependencies

Check if build-essential and libomp-dev packages are installed on your Ubuntu system:

dpkg -l | grep build-essential
dpkg -l | grep libomp-dev

Install the build-essential and libomp-dev packages if no information printed:

apt update
apt install build-essential
apt install libomp-dev

Install libc6_2.35

This installation is intended for scenarios where you encounter errors like this while running the pre-built es-node on Ubuntu 20.04:

/lib/x86_64-linux-gnu/libc.so.6: version 'glibc_2.32' not found
/lib/x86_64-linux-gnu/libc.so.6: version 'glibc_2.34' not found

To prevent the error, add the following line to your /etc/apt/sources.list:

deb http://security.ubuntu.com/ubuntu jammy-security main

Next, install libc6_2.35 by running the following commands:

apt update
apt install -y libc6

Check the status after launching the es-node

It's important to monitor the node closely until it successfully submits its first storage proof. Typically, the process encompasses three main stages.

Data sync phase

The es-node will synchronize data from other peers in the network. You can check from the console the number of peers the node is connected to and, more importantly, the estimated syncing time.

During this phase, the CPUs are expected to be fully occupied for data processing. If not, please refer to the FAQ for performance tuning on this area.

A typical log entry in this phase appears as follows:

INFO [01-18|09:13:32.564] Storage sync in progress progress=85.50% peerCount=2 syncTasksRemain=1@0 blobsSynced=1@128.00KiB blobsToSync=0 fillTasksRemain=30 emptyFilled=3,586,176@437.77GiB emptyToFill=608,127   timeUsed=1h48m7.556s  eta=18m20.127s

Sampling phase

Once data synchronization is complete, the es-node will enter the sampling phase, also known as mining.

A typical log entry of sampling during a slot looks like this:

INFO [01-19|05:02:23.210] Mining info retrieved                    shard=0 LastMineTime=1,705,634,628 Difficulty=4,718,592,000 proofsSubmitted=6
INFO [01-19|05:02:23.210] Mining tasks assigned                    shard=0 threads=64 block=10,397,712 nonces=1,048,576
INFO [01-19|05:02:26.050] Sampling done with all nonces            samplingTime=2.8s shard=0 block=10,397,712

When you see "Sampling done with all nonces", it indicates that your node has successfully completed all the sampling tasks within a slot. The "samplingTime" value informs you of the duration, in seconds, it took to complete the sampling process.

If the es-node doesn't have enough time to complete sampling within a slot, the log will display "Mining tasks timed out". For further actions, please refer to the FAQ.

Proof submission phase

Once the es-node calculates a valid storage proof, it will submit the proof to the EthStorage contract and receive the rewards.

A typical log entry of submitting proof looks like this:

INFO [01-19|05:02:23.210] Calculated a valid hash                  shard=0 thread=3 block=5,437,371 nonce=58101
INFO [01-19|05:05:23.210] Mining transaction confirmed"            txHash=0x7afa13e5211c403a7024bdf0a6880203d54698355679be9aab1aa0ecef78eecd
INFO [01-19|05:05:23.210] Mining transaction success! √            miner=0xBB9D13efa21c0a053eCFE622e2AbfAF0D7573f50

Once you see this log, congratulations on completing the entire process as a storage provider. You can also check how many storage proofs you've submitted, your ETH profit, and your ranking on the dashboard.

PreviousStorage Provider GuideNextConfiguration

Last updated