Operations, observability, backup and security projects that earn trust under real load.
11 long-form field guideskubernetes / kubernetes
The famous repository rewards a slower reading. Its real subject is a durable API for declaring operational intent, then reconciling clusters toward that intent through many cooperating control loops.
aquasecurity / trivy
One command can cover packages, operating-system vulnerabilities, application dependencies, secrets, configuration and SBOM output. The difficult part begins when a finding needs evidence, ownership and a repair decision.
apache / airflow
Airflow lets teams define workflows as Python DAGs, schedule task instances, track state in a metadata database, and inspect runs through its user interface and APIs. It is powerful for finite, dependency-driven batch work, but adopting the repository also means owning a scheduler, execution layer, database, provider set, and upgrade process.
prometheus / prometheus
Its pull-based collection and PromQL are only half the design. The other half is restraint: labels need boundaries, alerts need meaning and local storage needs an honest retention plan.
opentofu / opentofu
The fork preserved a familiar configuration, provider and state model, then placed its future under community governance. The result is useful only when teams treat plans and state as serious operational assets.
grafana / grafana
Connecting a data source is easy. The harder work is preserving metric meaning, query cost, permissions, alert ownership and licensing clarity as dashboards become shared operational infrastructure.
redis / redis
Redis makes counters, caches, queues, indexes, and ephemeral coordination feel immediate. The hard part is deciding which data deserves memory, what failure may lose, and which license actually covers the version and build you ship.
caddyserver / caddy
A short Caddyfile can front a serious service, but the repository is more than friendly syntax. It is a modular web server with a native JSON configuration, an admin API, certificate automation, and graceful runtime changes.
traefik / traefik
The proxy watches Docker, Kubernetes, files, and other providers, then hot-reloads routers, middlewares, and services. That responsiveness is valuable only when teams control who may publish the metadata it trusts.
temporalio / temporal
The server persists workflow history while user-owned workers replay deterministic code and execute side effects as activities. It can replace piles of retry and timer plumbing, provided teams accept a new programming and operating model.
grafana / loki
Grafana Loki is a horizontally scalable, multi-tenant log aggregation system inspired by Prometheus. It indexes a small set of labels for each log stream instead of the text of every line, and stores compressed logs in object storage. That keeps costs down for as long as the labels stay few and bounded.