-
Notifications
You must be signed in to change notification settings - Fork 1
CronJob
Gregory Nickonov edited this page Mar 4, 2019
·
3 revisions
R.cron_job 'logs-cleaner' do
# Run midnight UTC every single day
schedule '0 0 * * *'
# Default restart policy is :OnFailure, see below
pod_template_spec.restart_policy :Never
container do
# Setup main container
end
endCronJob is one of the Pod Template Owners, so all methods, like pod_template and container are available.
By default, Pod's restart_policy is Always, but this value is invalid for CronJobs and leads to the error when creating the CronJob. To simplify development, in Sunstone default value for restart policy is OnFailure, so if you don't want the CronJob to be restarted in case of failure don't' forget to set the value to Never.
| Sunstone property | Kubernetes property | Type |
|---|---|---|
| metadata | metadata | Kubernetes Object Metadata |
| spec | spec | CronJobSpec |
A shortcut to spec.schedule, used to specify a run schedule in the Cron format.
| Sunstone property | Kubernetes property | Type |
|---|---|---|
| concurrency_policy | concurrencyPolicy | String |
| failed_jobs_history_limit | failedJobsHistoryLimit | Integer |
| job_template | jobTemplate | JobTemplateSpec |
| schedule | schedule | String |
| successful_jobs_history_limit | successfulJobsHistoryLimit | Integer |
| suspend | suspend | Boolean |
| Sunstone property | Kubernetes property | Type |
|---|---|---|
| metadata | metadata | Kubernetes Object Metadata |
| spec | spec | JobSpec |