--- name: azure-waf-operational-excellence metadata: category: WellArchitectedFramework description: >- Generates guidance for operating Azure workloads, based on the Operational Excellence pillar of the Azure Well-Architected Framework. Use when designing DevOps practices, infrastructure as code, deployment pipelines, observability, incident management or automation for an Azure workload. --- # Azure Well-Architected Framework skill for the Operational Excellence pillar ## Overview The Operational Excellence pillar supports repeatable, reliable and safe deployments of infrastructure and code. It starts with a DevOps approach, automates what is routine, keeps people for what benefits from their judgement, and uses observability both to watch health and to test the design. ## Core principles The recommendations below follow the published framework. Written for this agent from Microsoft's published design review checklist. Codes are Microsoft's. - **Define standard practices** (OE:01): Align specialties to clear practices, with accountability and a blameless culture. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/devops-culture - **Standardize operations** (OE:02): Make routine, ad-hoc and emergency tasks consistent and predictable. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/formalize-operations-tasks - **Formalize development** (OE:03): Make the lifecycle from idea to delivery transparent. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/formalize-development-practices - **Standardize tools** (OE:04): Set source control, design patterns, documentation and style. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/tools-processes - **Use infrastructure as code** (OE:05): Prefer declarative, modular code with quality assurance. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/infrastructure-as-code-design - **Build a supply chain** (OE:06): Promote changes through automated, tested pipelines and quality gates. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/workload-supply-chain - **Design monitoring** (OE:07): Capture telemetry, metrics and logs from infrastructure and code. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/observability - **Manage incidents** (OE:08): Set roles, procedures and an architecture for fast detection, diagnosis and recovery. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/incident-response - **Test to business objectives** (OE:09): Adopt testing practices that uphold quality standards. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/testing - **Automate reliably** (OE:10): Automate repetitive tasks with clear return, keeping automation secure and maintainable. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/enable-automation - **Deploy safely** (OE:11): Release small increments with gates and progressive exposure, with a plan for emergencies. Grounding document: https://learn.microsoft.com/azure/well-architected/operational-excellence/safe-deployments ## Relevant products The following are _examples_ of products relevant to this pillar: - **Delivery**: GitHub Actions, Azure Pipelines, deployment slots, Azure Deployment Environments - **Infrastructure as code**: Terraform, Bicep - **Observability**: Azure Monitor, Log Analytics, Application Insights, Managed Grafana - **Operations**: Azure Automation, Update Manager ## Workload assessment questions Ask appropriate questions to understand the requirements and constraints of the workload and the organization. Choose from this list: - Who owns the workload in production, and how is on-call organized? - How do changes reach production, and how are they rolled back? - Is all infrastructure defined as code? - What logs, metrics and traces are collected, and where? - How are incidents declared, run and reviewed? - Which routine tasks are still manual? - How are environments kept consistent? ## Validation checklist Use this checklist to evaluate the architecture against this pillar: - Ownership and on-call are stated. - All infrastructure is deployed from code in source control. - Pipelines test and promote changes through environments. - Deployments are incremental, with a tested rollback. - Telemetry is central, and dashboards and alerts exist. - Incident roles and runbooks exist, with blameless reviews. - Routine tasks are automated where it pays.