Pipelines#
Pipelines are how the operator’s code reaches production without hand-walking it through. Continuous integration tests every change, continuous delivery promotes the resulting artifact through environments, and GitOps reconciles declared state into the cluster. The same machinery continuously ships mission-package updates, software updates, and the other software and data the mission depends on. See Environments for the tiers a change is promoted through.
For the operator this is more than convenience. A pipeline is the audit trail, every artifact deployed to a redirector, range, or defended host is traceable back to a commit, a reviewer, and a test result. Skip it and the operator gives up reproducibility and accountability in one move.
Practices#
Continuous integration tests every change; continuous delivery promotes the artifact through environments.
Declarative state in Git plus a controller that reconciles the running system.
Tools#
GitHub’s in-platform CI/CD. YAML workflows; the default when the code lives in GitHub.
GitLab’s pipeline engine. Broadest single-vendor DevSecOps surface.
The long-running CI server. Plugin-heavy, self-hosted, in production at most Fortune 500s.
The SaaS CI that pioneered Docker-first builds. Orbs, generous parallelism.
SaaS control plane, operator-owned runners. SaaS ergonomics without the data-egress problem.
Kubernetes-native CI/CD framework. Pipelines and tasks as CRDs.
GitOps continuous-delivery controller for Kubernetes. The de-facto GitOps CD tool.
The other CNCF-graduated GitOps controller. Composable CRDs the operator wires together.
Container-native CI engine. Lightweight alternative to Jenkins.
Azure DevOps CI/CD. Deep integration with Azure resources and .NET stacks.