Custom Monitor

On this page

Custom monitors run reusable scenarios on shared Testable-hosted or self-hosted runners, or on per-check cloud VMs when that feature is enabled. Use them when a simple HTTP, ping, port, or heartbeat check is not enough to prove that a user flow or business process works.

Supported Scenario Types

TypeUse it for
PlaywrightModern browser journeys and end-to-end checks.
PuppeteerChrome-based browser automation and generated URL checks.
Load in ChromeCreate a generated Puppeteer scenario that loads a target URL.
SeleniumCross-browser or existing Selenium workflows.
Node.js ScriptCustom JavaScript checks and API workflows.
JMeterHTTP test plans and performance-oriented checks.
GatlingScala-based simulations and API workflows.
LocustPython-based user flows and service checks.
PhantomJSLegacy browser scripts.
HAR ReplayReplay captured HTTP traffic.
PostmanAPI collections and environment-driven checks.
JavaCustom Java checks.
MonitorReusable protocol-level HTTP, ping, port, and related endpoint scenarios.
AIAI-authored monitoring scenarios when enabled for the organization.

Configuration

Scenario Selection

Choose an existing scenario or create a new one from the monitor form. New scenarios are created in the selected folder and group, then edited inline with the embedded scenario editor.

Scenario Parameters

Scenarios can declare parameters so the same scenario definition can power multiple monitors. Required parameters are shown on the monitor form and saved with the monitor.

Scheduling, Regions, and Success Criteria

Custom monitors use the same schedule, region, notification, maintenance window, metrics, and success criteria panels as other runner-based monitors. A region can use a shared pool or a per-check runner. Success criteria can be configured globally, by monitor type, and by custom scenario type in organization monitor settings.

Run a Playwright Test Example

The Playwright Test monitor example includes a test spec, configuration, and a Bash script that uploads the project, creates a monitor, and reports the first completed check. The test follows a read-only navigation flow on the Testable website.

Install Node.js 22 or later, npm, Bash, curl, jq, and zip. You will also need a writable Testable API key for an organization with monitoring access and available browser execution capacity.

git clone https://github.com/testable/playwright-monitor-example.git
cd playwright-monitor-example
npm ci
./scripts/create-monitor.sh

The script prompts for your API key without displaying it. The key selects the organization, so no organization ID is required. By default, the script pauses the monitor after reporting the first completed check. See the README for local execution, configuration, and the option to keep the monitor running on a schedule.

Results

Custom monitor incidents show the execution that caused the incident and, when available, the execution that resolved it. This makes it possible to inspect steps, timings, assertions, logs, screenshots, and other scenario artifacts from the incident page.

Use Cases

  • Login, checkout, signup, or other browser user journeys.
  • Multi-step API flows that require authentication or state.
  • Business validations that combine UI, API, and custom assertions.
  • Private workflows that need self-hosted runners inside your network.
  • Reuse of existing Postman, JMeter, Gatling, Locust, Selenium, Playwright, or Puppeteer assets.

Best Practices

  • Keep scenarios focused on one user-visible behavior.
  • Use parameters for environment, usernames, paths, IDs, or other per-monitor differences.
  • Prefer stable selectors and explicit assertions for browser checks.
  • Choose runner regions close to the users or infrastructure you are validating.
  • Use success criteria to alert on degraded performance before a scenario fails completely.