Skip to content

Configuration

Handler Parameters

The DatadogHTTPHandler accepts the following configuration parameters:

Required Parameters

  • api_key (str): Your Datadog API key

Optional Parameters

  • service (str): Service name for your logs (default: "python-app")
  • environment (str): Environment name (default: "production")
  • version (str): Application version (default: "1.0.0")
  • site (str): Datadog site (default: "datadoghq.com")
  • tags (str): Comma-separated tags (default: "")
  • batch_size (int): Number of logs to batch before sending (default: 10)
  • flush_interval (float): Maximum time in seconds before flushing logs (default: 5.0)
  • max_retries (int): Maximum number of retry attempts (default: 3)
  • timeout (float): Request timeout in seconds (default: 30.0)

Configuration Examples

Basic Configuration

import logging
from datadog_http_handler import DatadogHTTPHandler

handler = DatadogHTTPHandler(
    api_key="your-datadog-api-key"
)

logger = logging.getLogger(__name__)
logger.addHandler(handler)

Advanced Configuration

handler = DatadogHTTPHandler(
    api_key="your-datadog-api-key",
    service="my-web-service",
    environment="staging",
    version="2.1.0",
    site="datadoghq.eu",  # EU site
    tags="team:backend,component:auth",
    batch_size=25,
    flush_interval=10.0,
    max_retries=5,
    timeout=45.0
)

Environment Variables

You can also configure the handler using environment variables:

export DATADOG_API_KEY="your-api-key"
export DATADOG_SERVICE="my-service"
export DATADOG_ENV="production"
export DATADOG_VERSION="1.0.0"

import os
from datadog_http_handler import DatadogHTTPHandler

handler = DatadogHTTPHandler(
    api_key=os.environ["DATADOG_API_KEY"],
    service=os.environ.get("DATADOG_SERVICE", "python-app"),
    environment=os.environ.get("DATADOG_ENV", "production"),
    version=os.environ.get("DATADOG_VERSION", "1.0.0")
)

Logging Configuration

Log Levels

The handler respects standard Python logging levels:

logger.debug("Debug message")     # DEBUG level
logger.info("Info message")       # INFO level  
logger.warning("Warning message") # WARNING level
logger.error("Error message")     # ERROR level
logger.critical("Critical message") # CRITICAL level

Log Format

You can customize log formatting:

import logging
from datadog_http_handler import DatadogHTTPHandler

# Create handler
handler = DatadogHTTPHandler(api_key="your-key")

# Set custom format
formatter = logging.Formatter(
    '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
)
handler.setFormatter(formatter)

# Add to logger
logger = logging.getLogger(__name__)
logger.addHandler(handler)

Performance Tuning

Batch Size

Larger batch sizes reduce API calls but increase memory usage:

# For high-volume applications
handler = DatadogHTTPHandler(
    api_key="your-key",
    batch_size=50,  # Send 50 logs per batch
    flush_interval=2.0  # Flush every 2 seconds
)

# For low-volume applications
handler = DatadogHTTPHandler(
    api_key="your-key", 
    batch_size=5,   # Send 5 logs per batch
    flush_interval=10.0  # Flush every 10 seconds
)

Timeout Settings

Adjust timeouts based on your network conditions:

# For reliable networks
handler = DatadogHTTPHandler(
    api_key="your-key",
    timeout=10.0,  # 10 second timeout
    max_retries=2
)

# For unreliable networks
handler = DatadogHTTPHandler(
    api_key="your-key",
    timeout=60.0,  # 60 second timeout
    max_retries=5
)