Implementing Dry-Run Modes

Design worker logic to execute simulations without triggering external side effects for testing purposes.

In this lesson, you will learn how to design worker logic that allows you to simulate task execution without triggering irreversible side effects in your production systems.

The Problem of "Live" Workers

In complex background processing, your workers often interact with external systems—sending emails, charging credit cards, or updating database records. When debugging or testing new workflow logic, running these tasks in a production-like environment is dangerous because you cannot easily "undo" an API call or a payment. A "Dry-Run" mode allows you to validate that your worker logic, orchestration state, and data transformations are correct without actually moving the needle in the real world.

Implementing the Dry-Run Pattern

To implement a dry-run mode, you must decouple your Task Logic (what the worker computes) from your Side-Effect Drivers (how the worker interacts with the external world).

You can achieve this by passing an isDryRun flag into your worker’s execution context. Instead of calling external services directly, your workers should pass their intent through a wrapper or an abstraction layer that checks this flag:

async function processPayment(job, isDryRun = false) {
  const { amount, accountId } = job.data;

  // 1. Business Logic Execution
  const transaction = calculateFees(amount);

  // 2. Conditional Side Effect
  if (isDryRun) {
    console.log(`[DRY-RUN]: Would charge $${transaction} to account ${accountId}`);
    return { status: 'simulated', transaction };
  }

  // 3. Actual Side Effect
  return await paymentGateway.charge(accountId, transaction);
}

Strategic Integration

When designing your workers, ensure the dry-run flag is injectable, typically via the task payload or an environment variable. When running in dry-run mode, it is vital to log the entire intent of the operation. This allows you to verify that the job would have succeeded by inspecting logs, without having to risk the state of your production database. By centralizing these checks, you transform high-stakes background tasks into predictable, testable simulations.

Key Takeaway: Decouple your business logic from external effect execution and use a conditional flag to intercept side effects during simulation, ensuring you can validate complex workflows safely in any environment.