New Relic Synthetics: Proactive Uptime and Scripted Monitoring

Synthetic monitors test your site from locations around the world

Synthetic monitoring runs automated checks against your endpoints on a schedule, from locations worldwide. It answers the question "is my site up and working right now?" before a single user complains.

Types of Synthetic Monitors

  • Ping — the lightest check: is the URL responding?
  • Simple Browser — loads the page in a real browser and verifies content
  • Scripted Browser — multi-step journeys (log in, search, check out)
  • API test — validates JSON responses and status codes from your APIs

A Scripted Check

Scripted monitors use JavaScript to simulate a user journey:

const assert = require('assert');
$webDriver.get('https://example.com/login');
$webDriver.findElement($driver.By.id('user')).sendKeys('demo');
// ... submit, then assert the dashboard loaded
$webDriver.findElement($driver.By.css('.dashboard'));

Failures from multiple locations confirm a real outage vs. a regional blip

Locations and Private Minions

Run checks from New Relic’s public locations to measure the public internet, or deploy private minions inside your network to monitor internal apps that aren’t exposed publicly.

Need Use
Public site uptime Public locations (multi-region)
Internal/staging apps Private minion in your VPC
Geo performance Several locations, compare latency

Avoiding False Alarms

  1. Require failures from multiple locations before alerting.
  2. Set a sensible frequency — every 1–5 minutes for critical flows.
  3. Keep scripts resilient to minor DOM changes with stable selectors.

A good synthetic suite is your early-warning system; tune it so every alert is worth waking up for.

What to Learn Next

  • Alerting on synthetic failures and duration
  • Combining synthetics with APM to confirm root cause
  • Monitoring SSL certificate expiry with API checks

Arivanandhan Chitheshwaran