MillerByte Logging

Why Traditional API Logging Falls Short

Traditional API logging scatters information across multiple log lines, making it difficult to understand what happened during a request. You end up with dozens of log statements per endpoint, and when something goes wrong, you're left piecing together fragments from different sources.

Traditional Approach

Scattered logs throughout your API:

// Traditional approach - scattered logs
_logger.LogInformation("Request started: {Method} {Path}", method, path);
_logger.LogInformation("User authenticated: {UserId}", userId);
_logger.LogInformation("Fetching user data");
_logger.LogInformation("Processing business logic");
_logger.LogInformation("Request completed with status {Status}", statusCode);

Wide API Events

Capture everything in one structured event:

[ApiLogging]
public async Task<IActionResult> GetUser(int id)
{
	var user = await _userRepository.GetByIdAsync(id);
	return Ok(user);
}

This creates a single MongoDB document containing the complete request context: endpoint info, user identity, request/response bodies, timing data, trace IDs, and any custom log messages you add during processing.

Key Benefits