Skip to content

Latest commit

 

History

History
350 lines (255 loc) · 11.3 KB

start-from-scratch.md

File metadata and controls

350 lines (255 loc) · 11.3 KB

Start From Scratch

This tutorial will step you through setting up a project from scratch with Dstack SDK. The following have a couple ways to get started with

  • Create a next.js application written in JavaScript
  • Build a Python backend application

{% hint style="info" %}

Before You Start

Make sure you have Docker Desktop or OrbStack started before you begin. {% endhint %}

{% tabs %} {% tab title="next.js" %}

Installation

For this example, we will be starting from scratch and will need to generate a new project. For simplicity, I will use a template for a next.js docker app.

npx create-next-app --example with-docker nextjs-docker

This should create a new project called nextjs-docker in the current working directory. Now let's add the @phala/dstack-sdk to the project along with viem and dotenv to get our project compatible with the TEE simulator.

npm install @phala/dstack-sdk viem dotenv

Make Calls to Dstack SDK Functions

Now that we have the packages installed, we can begin to configure our existing project to make calls to the deriveKey(path, data) and tdxQuote(data)functions in the Dstack SDK.

Edit Dockerfile

First lets make 1 small change to the Dockerfile and add the line to line 66 in the file.

ENV DSTACK_SIMULATOR_ENDPOINT="http://host.docker.internal:8090"

Next, lets create the API calls for both of the functions.

Create API Calls

Create 2 new files called derivekey.js tdxquote.js in the pages/api/ folder.

touch ./pages/api/derivekey.js
touch ./pages/api/tdxquote.js

You should now have 2 empty files in the directory, and run ls ./pages/api to verify the files were created.

ls pages/api 
total 24
drwxr-xr-x  5 hashwarlock  staff   160B Nov  7 14:51 .
drwxr-xr-x  5 hashwarlock  staff   160B Nov  7 14:47 ..
-rw-r--r--@ 1 hashwarlock  staff   892B Nov  7 15:01 derivekey.js
-rw-r--r--  1 hashwarlock  staff   169B Nov  7 14:00 hello.js
-rw-r--r--@ 1 hashwarlock  staff   604B Nov  7 15:05 tdxquote.js

Now that the API calls are created, let's implement the api/derivekey function first. We can copy and paste the following code snippet to generate a random ECDSA Keypair.

// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import {TappdClient} from '@phala/dstack-sdk'
import { privateKeyToAccount } from 'viem/accounts'
import {keccak256} from "viem";
import 'dotenv/config'

export const dynamic = 'force-dynamic'

const endpoint = process.env.DSTACK_SIMULATOR_ENDPOINT || 'http://localhost:8090'

export default async function derivekey(req, res) {
    // Get the Tappd client
    const client = new TappdClient(endpoint)
    const randomNumString = Math.random().toString();
    // Call the deriveKey function and pass in the root of trust to derive a key
    const randomDeriveKey = await client.deriveKey('/', randomNumString);
    // Hash the derivedKey uint8Array value
    const keccakPrivateKey = keccak256(randomDeriveKey.asUint8Array());
    // Get the private key account from the derived key hash
    const account = privateKeyToAccount(keccakPrivateKey);
    // Return derived key pair
    res.status(200).json({account: account.address, privateKey: keccakPrivateKey});
}

Let's move to implementing the api/tdxquote API call with the following snippet of code.

// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import {TappdClient} from '@phala/dstack-sdk'
import 'dotenv/config'

export const dynamic = 'force-dynamic'

const endpoint = process.env.DSTACK_SIMULATOR_ENDPOINT || 'http://localhost:8090'

export default async function tdxquote(req, res) {
    // Get Tappd client
    const client = new TappdClient(endpoint)
    const randomNumString = Math.random().toString();
    // Generate Remote Attestation Quote based on a random string of data
    const getRemoteAttestation = await client.tdxQuote(randomNumString);
    // Return Remote Attestation result
    res.status(200).json({ getRemoteAttestation });
}

Now that these API calls are done, we can start with the fun of testing this locally.

Testing Locally

With the basic implementation done, let's do a quick test to see if the code works by building the docker image and testing locally. 

docker build -t my-nextjs-app:latest .

Now that the docker image is built, let's start testing against the TEE Simulator. First we must pull the latest TEE Simulator in docker, and run it with the following commands.

docker pull phalanetwork/tappd-simulator:latest
docker run --rm -p 8090:8090 phalanetwork/tappd-simulator:latest

With the simulator up and running, we can now start our docker app to test out the functionality.

docker run --rm -p 3000:3000 my-nextjs-app:latest

Now we can go to http://localhost:3000 and see our deployed application.

Call the APIs

We can test to see if our functions work by calling the API calls with

We should see the results similar to the following screenshots.

/api/derivekey

/api/tdxquote

{% endtab %}

{% tab title="backend" %} For the python example, we will take an existing tutorial from Docker docs https://docs.docker.com/guides/python/containerize/ and add Dstack SDK to the project to make calls to the TEE SDK functions.

Installation

First step is to clone the python application repo.

git clone https://github.com/estebanx64/python-docker-example.git

Now that we have cloned a python repo, let's add the Dockerfile with the following contents.

# syntax=docker/dockerfile:1

# Comments are provided throughout this file to help you get started.
# If you need more help, visit the Dockerfile reference guide at
# https://docs.docker.com/go/dockerfile-reference/

# Want to help us make this template better? Share your feedback here: https://forms.gle/ybq9Krt8jtBL3iCk7

ARG PYTHON_VERSION=3.11.4
FROM python:${PYTHON_VERSION}-slim AS base

# Prevents Python from writing pyc files.
ENV PYTHONDONTWRITEBYTECODE=1

# Keeps Python from buffering stdout and stderr to avoid situations where
# the application crashes without emitting any logs due to buffering.
ENV PYTHONUNBUFFERED=1

WORKDIR /app

# Create a non-privileged user that the app will run under.
# See https://docs.docker.com/go/dockerfile-user-best-practices/
ARG UID=10001
RUN adduser \
    --disabled-password \
    --gecos "" \
    --home "/nonexistent" \
    --shell "/sbin/nologin" \
    --no-create-home \
    --uid "${UID}" \
    appuser

# Download dependencies as a separate step to take advantage of Docker's caching.
# Leverage a cache mount to /root/.cache/pip to speed up subsequent builds.
# Leverage a bind mount to requirements.txt to avoid having to copy them into
# into this layer.
RUN --mount=type=cache,target=/root/.cache/pip \
    --mount=type=bind,source=requirements.txt,target=requirements.txt \
    python -m pip install -r requirements.txt

# Switch to the non-privileged user to run the application.
USER appuser

# Copy the source code into the container.
COPY . .

# Expose the port that the application listens on.
EXPOSE 3000

# Dstack simulator endpoint for local deployments
ENV DSTACK_SIMULATOR_ENDPOINT="http://host.docker.internal:8090"

# Run the application.
CMD python3 -m uvicorn app:app --host=0.0.0.0 --port=3000

You can also add a .dockerignore file with the following contents.

# Include any files or directories that you don't want to be copied to your
# container here (e.g., local build artifacts, temporary files, etc.).
#
# For more help, visit the .dockerignore file reference guide at
# https://docs.docker.com/go/build-context-dockerignore/

**/.DS_Store
**/__pycache__
**/.venv
**/.classpath
**/.dockerignore
**/.env
**/.git
**/.gitignore
**/.project
**/.settings
**/.toolstarget
**/.vs
**/.vscode
**/*.*proj.user
**/*.dbmdl
**/*.jfm
**/bin
**/charts
**/docker-compose*
**/compose.y*ml
**/Dockerfile*
**/node_modules
**/npm-debug.log
**/obj
**/secrets.dev.yaml
**/values.dev.yaml
LICENSE
README.md

Now that we have the Dockerfile built, let's add the Python Dstack SDK to our python project in requirements.txt

dstack-sdk==0.1.2
asyncio>=3.4.3
fastapi==0.111.0

Make Calls to Dstack SDK Functions

Let's make 2 new API call within the app.py file for derivekey and tdxquote.

import os
from dstack_sdk import AsyncTappdClient, DeriveKeyResponse, TdxQuoteResponse
from fastapi import FastAPI

app = FastAPI()

@app.get("/")
async def root():
    return {"message": "The World! Call /derivekey or /tdxquote"}

@app.get("/derivekey")
async def derivekey():
    client = AsyncTappdClient()
    deriveKey = await client.derive_key('/', 'test')
    assert isinstance(deriveKey, DeriveKeyResponse)
    asBytes = deriveKey.toBytes()
    assert isinstance(asBytes, bytes)
    limitedSize = deriveKey.toBytes(32)
    return {"deriveKey": asBytes.hex(), "derive_32bytes": limitedSize.hex()}
    
@app.get("/tdxquote")
async def tdxquote():
    client = AsyncTappdClient()
    tdxQuote = await client.tdx_quote('test')
    assert isinstance(tdxQuote, TdxQuoteResponse)
    return {"tdxQuote": tdxQuote}

Now that these API calls are done, we can start with the fun of testing this locally.

Testing Locally

With the basic implementation done, let's do a quick test to see if the code works by building the docker image and testing locally. 

docker build -t my-python-app:latest .

Now that the docker image is built, let's start testing against the TEE Simulator. First we must pull the latest TEE Simulator in docker, and run it with the following commands.

docker pull phalanetwork/tappd-simulator:latest
docker run --rm -p 8090:8090 phalanetwork/tappd-simulator:latest

With the simulator up and running, we can now start our docker app to test out the functionality.

docker run --rm -p 3000:3000 my-nextjs-app:latest

Now we can go to http://localhost:3000 and see our deployed application.

Call the APIs

We can test to see if our functions work by calling the API calls with

We should see the results similar to the following screenshots.

/derivekey

/tdxquote

{% endtab %} {% endtabs %}

Conclusion

Now you have the skills to start building on the Dstack SDK from scratch. Checkout the other tutorials where we start from a template or integrate the SDK into an existing project. For more info on how to setup your project to be deployed on the real TEE hardware, reach out to the Phala Team for more information.