The celery-low Fix Was a Configuration-Contract Failure

Connecting missing type, missing service port, and command fixes into one root lesson.

Symptom

I fixed celery-low in three small steps: the deployment command and values, the missing service.port, and then the missing celery.type.

The symptom pointed at the wrong layer. The failures were not one mysterious Kubernetes bug; they were multiple gaps between the chart template and the values contract.

My first rule: identify the stage of failure

producer -> queue/broker -> worker class -> task execution -> side effect

What I checked

I checked queue depth, oldest-job age, task execution time, worker availability, retry/error rate before changing anything.

Where the problem actually was

What this came down to was this: The failures were not one mysterious Kubernetes bug; they were multiple gaps between the chart template and the values contract. From that I kept one rule: When a workload is special, give it an explicit schema and test the rendered command and manifest together.

Prevention checklist

Test process commands exactly as rendered in the pod spec.
Measure queue wait separately from execution time.
Confirm workers consume the intended queue.
Scale from workload pressure, not CPU alone.
Keep scheduler semantics separate from worker semantics.