Operations
Once your Pulumi program is written and first deployed, “operations” is everything that happens next: rolling out changes safely, running Pulumi from CI/CD, diagnosing failures, and recovering from incidents.
These pages cover the Day 2 workflows for running Pulumi at scale.
Sections
Stack management - Day-to-day stack operations: targeted updates, update plans, editing state files when recovery is needed, and configuring a self-managed (DIY) state backend.
Continuous delivery - Integrate Pulumi with your CI/CD platform of choice. Reference setups for Argo CD, GitHub Actions, GitLab CI/CD, Jenkins, and more.
Troubleshooting - Diagnose and recover from common failures including update conflicts, destroy failures, interrupted updates, and CI/CD pipeline issues.
Debugging - Attach a debugger, capture verbose logs, and trace performance for your Pulumi programs.
Docker images - Pulumi’s officially published container images: what’s in each, where to pull from, release cadence, and how to use them as a base for custom images.
Container image build caching and deployment reproducibility - Build caching, digest-based tagging, and a build/deploy stack split for container images built with the Docker or Docker Build providers.
Least privilege - Apply least-privilege security across your Pulumi platform: cloud provider credentials, IaC and ESC access, and CI/CD. A recommended next step once your operational workflows are in place.