Run Kubernetes Job

What it does

This action creates one or more Pods and ensures that a specified number of them successfully terminate using Kubernetes Jobs. A simple case is to create one Job object in order to reliably run one Pod to completion. The Job object will start a new Pod if the first Pod fails or is deleted (for example due to a node hardware failure or a node reboot).

How it works

This action executes kubectl apply against a K8s cluster to run a job using provided YAML. Then, it runs kubectl describe to check if the job was completed and with what result (up to 90 minutes).

Action preview

SetupOptionsConditionAction

Kubernetes Run Job Action Features

• It’s possible to use Job YAML from the pipeline filesystem or define it in the action settings, so it can be generated by the action executed before the Run Job action
• Supported clusers: Private, Google Kubernetes Engine, Amazon EKS and Azure AKS clusters
• Supported authentication modes: Basic, Token, Client Certificates & Service Account
• Any version of Kubectl is supported including the latest
• In Options you can find the option Don’t remove jobs. If selected, the action will not perform spring cleaning of old jobs. Most often used to debug failed jobs.
• In Options you can find the option Don’t wait. If selected, the action won’t wait for the run job to complete.
• There are different Record modes supported: update the existing annotation value only if one already exists, Record the command and Don’t record the command
• Input fields: Server, Username, Password, YAML support Environment variables
• The action can be configured in YAML mode. See Run Kubernetes Job YAML parameters and Run Kubernetes Job YAML config example
• Buddy's Run Kubernetes Job API is supported
• The action supports conditional execution: it can be run on each execution, only if there are changes in the repository or specific paths since las execution or if an environment variable has a specific value
• Error handling: you can either stop the execution, mark action and pipeline as failed or proceed the execution, mark action as warning and jump to the next action

K8s/GKE Authorization:

• Host
• Port
• User
• Password
• API token
• Certificate