Menu
I have a service that deploys thousands of short lived jobs on Kubernetes each day. I am trying to get Kubernetes to delete these jobs after completion using the functionality described here:
https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/#clean-up-finished-jobs-automatically
The job completes, but the Job remains in the cluster after the time limit expressed in ttlSecondsAfterFinished.
I'm testing this on GKE and my master node is running 1.12.7-gke.25
I understand this is an alpha feature, but I haven't gotten a single Job to be deleted.
apiVersion: batch/v1
kind: Job
metadata:
name: ttl-test
spec:
ttlSecondsAfterFinished: 100
template:
metadata:
name: ttl-test
spec:
containers:
- name: test-container
image: myimage
608
Delete the job with kubectl (e.g. kubectl delete jobs/pi or kubectl delete -f ./job. yaml ). When you delete the job using kubectl , all the pods it created are deleted too.
How do you configure a Kubernetes Job so that Pods are retained after completion? - Configure the backofflimit parameter with a non-zero value. - Set a startingDeadlineSeconds value high enough to allow you to access the logs. - Set an activeDeadlineSeconds value high enough to allow you to access the logs.
To delete failed Jobs in GKE you will need to use following command: $ kubectl delete job $(kubectl get job -o=jsonpath='{. items[?(@. status.
Nope, Kubernetes no more reserves memory or CPU once Pods are marked completed.
In Kubernetes v1.21 the TTL Controller is in beta and enabled by default.
Until Kubernetes v1.20 it was an alpha feature. If you want to use it with a Kubernetes version older than v1.21, make sure alpha features are enabled on your GKE cluster. You can check this on the GCP console UI under Clusters / <your_cluster> / Details / Kubernetes alpha features.
99
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
