Controlling cron job on kubernates

Question

I have a job in kubernetes that usually takes 2 minutes, and on certain occasions lasts up to 20 minutes. I'd rather run it every 5 minutes, but I can't run more than one job, no parallelism. How can I coordinate the runs? Of course I can run it every 20 minutes, but I need it to run more frequently.

One solution is to run it from a shell script, but then I'll need to kill it after 20 minutes, which is too much; we have kubernets.

Is there a way to coordinate the runs intelligently? (Policies may vary, but that's a different story.)

Answer 1

Use a cron job and set the concurrencyPolicy to Forbid.

Forbid: The CronJob does not allow concurrent runs; if it is time for a new job run and the previous job run hasn't finished yet, the CronJob skips the new job run.

apiVersion: batch/v1
kind: CronJob
metadata:
  name: hello
spec:
  schedule: "*/5 * * * *"
  concurrencyPolicy: Forbid
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hello
            image: busybox:1.28
            imagePullPolicy: IfNotPresent
            command:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure