Skip to content

Latest commit

 

History

History
69 lines (51 loc) · 2.87 KB

README.md

File metadata and controls

69 lines (51 loc) · 2.87 KB
Raw

Next.js API to Tinybird

The goal of this project is to track API requests into Tinybird data pipeline, providing an easy way for your users to understand their activity/logs from your API.

Motivation

This is related to the Live Session for activity logs & usage for an application by Tinybird.

In this case I've implemented a decorator to track server-side analytics for api calls and get response times, status codes, path & method called.
Request & responses payloads could be added for tracking porpouses, but for security reasons should be avoided unless ofuscated treated to avoid any GDPR conflicts.

For user's UI activity we are using Google Anayltics alternative tracking based on Tinybird's snippet provided on the live session.

Setup

This is project was its a Next.js project bootstrapped with create-next-app.

Upstash

Upstash provide Serverless Kafka to easy configure your Event-Based architecture for you app.

  1. Create an account in Upstash
  2. Create a Kafka application
  3. Get the @upstash/kafka configuration and recover the variables UPSTASH_KAFKA_REST_URL, UPSTASH_KAFKA_REST_USERNAME & UPSTASH_KAFKA_REST_PASSWORD, and configure it on the .env.local.
  4. Create a topic inside the Kafka application called requests.
  5. Setup the project and start hitting the /api/users endpoints to create events on the topic.

Tinybird

  1. Create a Tinybird account, you can use the starter for google analytics.
  2. Configure a new DataSource from Kafka, its super simple to connect with your existing Upstash kafka topic.
  3. Setup the Bootstrap URL, Username & Password using SCRAM-SHA-256.
  4. Select the topic that you created, and will automatically generate a schema based on the messages available in the topic.
  5. Create a Pipe that will work as an API endpoint as well.
SELECT
 date,
 method,
 path,
 statusCode as "status_code",
 responseTime as "response_time",
 browser_name,
 os_name
FROM kafka_ds_523951
WHERE
  date >= today() - interval 7 day
  {% if defined(path) %}
    path = {{String(path)}}
  {% end %}
ORDER BY {{column(order_by, 'date')}} desc
LIMIT {{Int32(page_size, 100)}}
OFFSET {{Int32(page, 0) * Int32(page_size, 100)}}
  1. Configure the TINYBIRD_URL & TINYBIRD_TOKEN on your local environment.
    NOTE: Even though Tinybird can configure token's per customer, you could also recover specific user's parameters (user_id from authentication) on /api/requests and provide it directly from the backend as a query parameter.

Getting Started

First, run the development server:

npm run dev
# or
yarn dev

Open http://localhost:3000 with your browser to see the result.