Frequently Asked Questions

Basic Usage

What is the difference between the global logger and a custom logger?

The global logger is used directly via package-level functions such as dd.Info(), suitable for simple scenarios. A custom logger is created via dd.New(), supporting independent configuration and lifecycle management.

// Global logger
dd.Info("global log")

// Custom logger
logger, _ := dd.New(dd.JSONConfig())
logger.Info("independent log")

How do I initialize the global logger at program startup?

func init() {
    err := dd.InitDefault(dd.JSONConfig())
    if err != nil {
        log.Fatal(err)
    }
}

Or via SetDefault:

cfg := dd.DefaultConfig()
cfg.Format = dd.FormatJSON
cfg.Targets = []dd.OutputTarget{
    dd.ConsoleOutput(),
    dd.FileOutput("logs/app.json"),
}
logger, _ := dd.New(cfg)
dd.SetDefault(logger)

What happens with Fatal-level logs?

Fatal / Fatalf / FatalWith call os.Exit(1) after emitting the log (deferred functions will not run; internally it first tries to Close() to flush pending logs, waiting up to 5 seconds). You can customize the exit behavior via FatalHandler; for resource cleanup, use ErrorWith + an explicit Shutdown(ctx) instead.

Configuration

How do I output to both console and file?

cfg := dd.DefaultConfig()
cfg.Targets = []dd.OutputTarget{
    dd.ConsoleOutput(),
    dd.FileOutput("logs/app.log"),
}
logger, _ := dd.New(cfg)

// Or JSON format
cfg = dd.DefaultConfig()
cfg.Format = dd.FormatJSON
cfg.Targets = []dd.OutputTarget{
    dd.ConsoleOutput(),
    dd.FileOutput("logs/app.json"),
}
logger, _ = dd.New(cfg)

How do I dynamically change the log level?

_ = logger.SetLevel(dd.LevelDebug)  // Modify at runtime (returns an error)
_ = dd.SetLevel(dd.LevelDebug)      // Modify the global logger's level

How do I configure the file-rotation strategy?

Configure via FileWriter:

fw, _ := dd.NewFileWriter("logs/app.log",
    dd.DefaultFileWriterConfig(),  // 100MB, 30 days, 10 backups
)

Performance

Will logging affect program performance?

DD is designed for high performance:

• Low-allocation optimization on hot paths
• Atomic level checks, lock-free
• Sensitive-data filtering for large inputs (>=10KB) runs in a separate goroutine with timeout protection; small inputs are processed synchronously
• Optional buffered writes to reduce I/O

How do I optimize for high-throughput scenarios?

1. Use BufferedWriter to reduce I/O
2. Check the level before building fields
3. Consider enabling log sampling
4. Avoid Any fields on hot paths

See Performance Tuning.

Security

How does sensitive-data filtering work?

SensitiveDataFilter uses regex pattern matching to automatically replace matching sensitive values with [REDACTED] before the log is written. Small inputs are processed synchronously; for large inputs the regex scan runs in a separate goroutine, but the caller blocks waiting for the result (the wait scales with input size — a fixed 50ms for inputs up to 32KB, then +200ms per additional 32KB; on timeout it returns [REDACTED]).

How do I customize sensitive-data patterns?

filter, _ := dd.NewCustomSensitiveDataFilter(
    `(?i)my_secret_field\s*[:=]\s*\S+`,
)

How do I ensure logs are not tampered with?

Use IntegritySigner to HMAC-sign logs:

cfg, _ := dd.DefaultIntegrityConfigSafe()
signer, _ := dd.NewIntegritySigner(cfg)
sig := signer.Sign(logMessage)
// Verify: signer.Verify(signedEntry)

Error Handling

Why does AddWriter return an error?

Possible reasons:

• ErrNilWriter -- a nil Writer was passed
• ErrLoggerClosed -- the logger is closed
• ErrMaxWritersExceeded -- the writer count exceeds the limit

How do I handle write failures?

logger.SetWriteErrorHandler(func(w io.Writer, err error) {
    // Custom handling
    metrics.WriteErrors.Inc()
})

Testing

How do I capture logs in tests?

Use LoggerRecorder:

rec := dd.NewLoggerRecorder()
logger, _ := rec.NewLogger()

logger.Info("test")

if !rec.ContainsMessage("test") {
    t.Error("expected log not found")
}

See Test Helper.

Next Steps

• Quick Start -- Getting started guide
• API Reference -- Complete API
• Production Checklist -- Pre-launch checks