Administration and Deployment

This section contains information about deploying DAGs into production and the administration of airflow deployments.

• Production Deployment
  • Database backend
  • Multi-Node Cluster
  • Logging
  • Configuration
  • Scheduler Uptime
  • Production Container Images
  • Helm Chart for Kubernetes
  • Kerberos-authenticated workers
  • Secured Server and Service Access on Google Cloud
• Security
  • Access Control
  • API
  • Audit Logs in Airflow
  • Flower
  • Kerberos
  • Webserver
  • Workload
  • Secrets
  • Reporting Vulnerabilities
• Logging & Monitoring
  • Logging and Monitoring architecture
  • Logging for Tasks
  • Metrics
  • Callbacks
  • Checking Airflow Health Status
  • Error Tracking
  • Tracking User Activity
• Kubernetes
  • Helm Chart for Kubernetes
• Lineage
  • Shorthand notation
  • Lineage Backend
• Listeners
  • Lifecycle Events
  • TaskInstance State Change Events
  • Usage
• DAG Serialization
  • Dag Serialization Settings
  • Limitations
  • Using a different JSON Library
• Modules Management
  • How package/modules loading in Python works
  • Typical structure of packages
  • Built-in PYTHONPATH entries in Airflow
  • Best practices for module loading
  • Inspecting your PYTHONPATH loading configuration
  • Adding directories to the PYTHONPATH
  • Creating a package in Python
• Scheduler
  • DAG File Processing
  • Triggering DAG with Future Date
  • Running More Than One Scheduler
  • Fine-tuning your Scheduler performance
• Pools
  • Using multiple pool slots
• Cluster Policies
  • How do define a policy function
  • Available Policy Functions
  • Examples
• Priority Weights