🐕

# /etc/datadog-agent/datadog.yaml
api_key: YOUR_API_KEY
site: datadoghq.com

# Hostname
hostname: my-server

# Tags
tags:
  - env:production
  - service:web
  - team:backend

# Logs
logs_enabled: true

# APM
apm_config:
  enabled: true
  apm_dd_url: https://trace.agent.datadoghq.com

# Process monitoring
process_config:
  process_collection:
    enabled: true

from datadog import initialize, statsd

initialize(statsd_host="localhost", statsd_port=8125)

# Count
statsd.increment("page.views")
statsd.increment("user.signups", tags=["plan:free"])

# Gauge
statsd.gauge("queue.size", 42)
statsd.gauge("memory.used", 1024, tags=["host:web-1"])

# Histogram
statsd.histogram("request.duration", 0.5)

# Distribution
statsd.distribution("payment.amount", 99.99)

# Set (count unique values)
statsd.set("users.unique", user_id)

# Timing
from datadog import statsd

@statsd.timed("function.duration")
def my_function():
    pass

const StatsD = require("hot-shots");

const dogstatsd = new StatsD({
  host: "localhost",
  port: 8125,
  prefix: "myapp.",
  globalTags: { env: "production" },
});

// Count
dogstatsd.increment("page.views");
dogstatsd.increment("api.calls", 1, { endpoint: "/users" });

// Gauge
dogstatsd.gauge("queue.size", 42);

// Histogram
dogstatsd.histogram("request.duration", 150);

// Timer
const timer = dogstatsd.timer("function.duration");
// ... do work
timer.stop();

# Install: pip install ddtrace
# Run: ddtrace-run python app.py

from ddtrace import tracer, patch_all

# Auto-instrument libraries
patch_all()

# Manual span
with tracer.trace("my.operation", service="my-service") as span:
    span.set_tag("user.id", user_id)
    # Do work
    result = do_something()
    span.set_tag("result.count", len(result))

# Decorator
@tracer.wrap(service="my-service", resource="process_order")
def process_order(order_id):
    pass

// Install: npm install dd-trace
// Require at top of entry file
const tracer = require("dd-trace").init({
  service: "my-service",
  env: "production",
});

// Manual span
const span = tracer.startSpan("my.operation");
span.setTag("user.id", userId);
try {
  // Do work
} finally {
  span.finish();
}

// With scope
tracer.trace("my.operation", { service: "my-service" }, (span) => {
  span.setTag("custom.tag", "value");
  return doSomething();
});

# /etc/datadog-agent/conf.d/python.d/conf.yaml
logs:
  - type: file
    path: /var/log/myapp/*.log
    service: my-service
    source: python
    tags:
      - env:production

  - type: docker
    service: my-container
    source: docker

  - type: tcp
    port: 10514
    service: my-service
    source: custom

import logging
import json_log_formatter

# Setup JSON formatter
formatter = json_log_formatter.JSONFormatter()
handler = logging.StreamHandler()
handler.setFormatter(formatter)

logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.INFO)

# Log with attributes
logger.info("User logged in", extra={
    "user_id": user_id,
    "dd.trace_id": tracer.current_span().trace_id,
    "dd.span_id": tracer.current_span().span_id,
})

resource "datadog_monitor" "cpu_high" {
  name    = "High CPU Usage"
  type    = "metric alert"
  message = "CPU usage is high on {{host.name}}. @slack-alerts"

  query = "avg(last_5m):avg:system.cpu.user{env:production} by {host} > 80"

  thresholds = {
    critical = 80
    warning  = 70
  }

  notify_no_data    = true
  no_data_timeframe = 10

  tags = ["env:production", "team:backend"]
}

resource "datadog_monitor" "error_rate" {
  name    = "High Error Rate"
  type    = "query alert"
  message = "Error rate exceeded threshold. @pagerduty"

  query = "sum(last_5m):sum:trace.http.request.errors{env:production}.as_count() / sum:trace.http.request.hits{env:production}.as_count() * 100 > 5"

  thresholds = {
    critical = 5
    warning  = 2
  }
}