Uptime

Description

monitoring RESTful API server that allows authenticated users to monitor URLs, and get detailed uptime reports about their availability, average response time, and total uptime/downtime.

Overview

• Signup with email verification.
• CRUD operations for URL checks (GET, PUT and DELETE can be called only by the user user who created the check).
• Authenticated users can receive a notification whenever one of their URLs goes down or up again:
  • Email.
  • Webhook (optional).
• Authenticated users can get detailed uptime reports about their URLs availability, average response time, and total uptime/downtime.
• Authenticated users can group their checks by tags and get reports by tag.

Decisions I took

using linux CRON job over Nodejs setInterval()

• I think setInterval wouldn't be as accurate as the CRON job if the node main thread is busy with somthing else due the whole event loops life cycle, thus I delegate polling request every interval of time to a CRON which is more reliable and accurate and making use of linux OS

Features

• APIs should be consuming and producing application/json.
• Authenication should be stateless.
• Each URL check may have the following options:
  • name: The name of the check.
  • url: The URL to be monitored.
  • protocol: The resource protocol name HTTP, HTTPS, or TCP.
  • path: A specific path to be monitored (optional).
  • port: The server port number (optional).
  • webhook: A webhook URL to receive a notification on (optional).
  • timeout (defaults to 5 seconds): The timeout of the polling request (optional).
  • interval (defaults to 10 minutes): The time interval for polling requests (optional).
  • threshold (defaults to 1 failure): The threshold of failed requests that will create an alert (optional).
  • authentication: An HTTP authentication header, with the Basic scheme, to be sent with the polling request (optional).
    • authentication.username
    • authentication.password
  • httpHeaders: A list of key/value pairs custom HTTP headers to be sent with the polling request (optional).
  • assert: The response assertion to be used on the polling response (optional).
    • assert.statusCode: An HTTP status code to be asserted.
  • tags: A list of the check tags (optional).
  • ignoreSSL: A flag to ignore broken/expired SSL certificates in case of using the HTTPS protocol.
• Each report may have the following information:
  • status: The current status of the URL.
  • availability: A percentage of the URL availability.
  • outages: The total number of URL downtimes.
  • downtime: The total time, in seconds, of the URL downtime.
  • uptime: The total time, in seconds, of the URL uptime.
  • responseTime: The average response time for the URL.
  • history: Timestamped logs of the polling requests.

Run Local Development environment via docker

docker compose --env-file .env.dev up

Test

Running unit tests

# unit tests
 pnpm run test

# test coverage
 pnpm run test:cov

Running e2e tests

requires pnpm and docker

basically what running e2e tests do:

1. it spawns a test db container
2. create a an instance of the app
3. run e2e tests
4. close/destory the app
5. teardown test db

 pnpm test:e2e

Swagger (OpenAPI) documentation

accessed via the endpoint /api