Seraj Haqiqi
a36e72877d
- Add jarRef and jarPullSecret fields to FlinkJob CRD (jarUri/basicAuth deprecated) - OCI pull via go-containerregistry with dual auth (K8s pull secret + env vars) - Media type validation on pulled layers - Atomic status patches with runningJarRef/runningJarDigest tracking - NeedsUpgrade/RunningRef/RunningRefPatchData domain helpers - README with usage guide, pushing JARs, and GitHub Actions CI/CD workflow - CONTEXT.md domain glossary
1.5 KiB
Flink Kube Operator
Kubernetes operator that manages the lifecycle of Apache Flink jobs via CRDs.
Language
FlinkJob: A Kubernetes Custom Resource representing a desired Flink job. Contains spec (desired state) and status (observed state). Avoid: Job CR, CRD instance
ManagedJob: Runtime wrapper that drives a FlinkJob's lifecycle through its state machine (init → run → savepoint → upgrade). Avoid: Job controller, job runner
jarRef:
Full OCI reference identifying a JAR artifact in a registry (e.g. registry.io/org/app:1.0.0).
Avoid: jarUri, jar URL, artifact URL
jarDigest: SHA256 content hash of the JAR layer in the OCI manifest. Immutable identifier for the artifact bytes. Avoid: content hash, checksum
pullSecret:
A kubernetes.io/dockerconfigjson Kubernetes Secret containing registry credentials, referenced by name in the FlinkJob spec.
Avoid: imagePullSecret, registry secret
JarFile: Ephemeral artifact handle — a downloaded/pulled JAR on local disk awaiting upload to Flink, then deleted. Avoid: temp jar, local jar
LifeCycleStatus: Operator-managed lifecycle state of a FlinkJob (HEALTHY, RESTORING, GRACEFULLY_PAUSED, FAILED, etc.). Avoid: operator status, lifecycle state
JobStatus: Flink-reported job state (RUNNING, FAILED, FINISHED, etc.). Mirrored from the Flink REST API. Avoid: flink status, job state
Savepoint: A consistent snapshot of a Flink job's state, used for restart and upgrade. Avoid: checkpoint, snapshot