Avançar para o conteúdo principal
https://catalogartifact.azureedge.net/publicartifacts/dcassociatesgroupinc.airflow-ubuntu-2404-95a55e4a-5e33-44f7-b8bf-a44ddfd897f3/image0_dcacdiamond216.png

Apache Airflow on Ubuntu 24.04 LTS

de Derek Coleman & Associates Corporation

Apache Airflow 2.10 on Ubuntu 24.04 LTS, workflow orchestration platform.

Apache Airflow 2.10 on Ubuntu 24.04 LTS for Azure Marketplace

Apache Airflow is the open-source platform to author, schedule, and monitor data workflows as code. This image ships Airflow 2.10 pre-installed in a dedicated virtualenv, pre-configured, and hardened on Ubuntu 24.04 LTS, ready to run pipelines on first boot inside your Azure subscription.

Why this image: Standing up Airflow on Azure without a managed service means more than a pip install. You also pin the version against the matching constraints file, wire separate systemd units for the webserver and scheduler, generate a Fernet key, and patch current Ubuntu CVEs. This image does all of that so your team ships a working orchestration platform without the manual configuration work.

Who this is for: Data engineers building and scheduling ETL and ELT pipelines. Platform teams who need self-managed workflow orchestration in their own Azure subscription. Analytics teams automating recurring data jobs as code.

Target use cases:

  • ETL and ELT orchestration — schedule and monitor multi-step data pipelines defined as Python DAGs
  • Data warehouse loading — coordinate extract, transform, and load jobs with dependencies and retries
  • Machine-learning pipelines — orchestrate training, evaluation, and deployment steps on a schedule
  • Recurring job automation — replace fragile cron jobs with observable, retryable workflows

What is pre-installed and configured:

  • Apache Airflow 2.10 in a virtualenv — installed under /opt/airflow with the matching upstream constraints file; admin user and Fernet key generated at first boot, never baked in
  • Webserver on port 8080 — the Airflow UI for DAG monitoring and management
  • Scheduler as a separate systemd unit — webserver and scheduler each start on boot and restart on failure
  • 33 hardening steps applied — automated audit covering Trusted Launch, SSH key injection, walinuxagent, Defender for Cloud, and cloud-init idempotency
  • Trusted Launch ready — Gen2 image with vTPM and Secure Boot support per Microsoft requirements
  • Azure Linux Agent pre-installed — custom-script extension, run-command, and managed identity work on first boot
  • Monthly patch cadence — rebuilt from upstream Ubuntu and project security advisories within days of each release

Recommended deployment: Standard_D2as_v5 (2 vCPU, 8 GB RAM) for development and light DAGs. Standard_D4as_v5 (4 vCPU, 16 GB RAM) for production scheduling. Attach a Premium SSD data disk and use an external metadata database for heavier workloads.

System requirements: Minimum 2 vCPU, 8 GB RAM. Gen2 VM. Open TCP 8080 (web UI) inbound in your Network Security Group.

Azure integration: Azure Monitor Agent, Defender for Cloud, Azure Backup, and Update Manager extensions install without conflicts. Assign a Managed Identity to let your workloads retrieve secrets from Azure Key Vault without embedding credentials in configuration files.

Licensing: Apache Airflow is Apache License 2.0 open-source software. No per-seat license or royalty required. Only standard Azure compute and storage rates apply.

Support: Email support@dcassociatesgroup.com for deployment assistance.