Skip to content

Latest commit

 

History

History
74 lines (64 loc) · 4.01 KB

podtemplates.md

File metadata and controls

74 lines (64 loc) · 4.01 KB
Raw

Pod templates

A pod template specifies a subset of PodSpec configuration that will be used as the basis for the Task pod.

This allows for customizing some Pod specific field per Task execution, TaskRun or for all Task executions part of a PipelineRun.

Alternatively, you can also define a default pod template in Tekton config, see Customizing basic execution parameters When a pod template is specified for a PipelineRun or TaskRun, the default pod template is ignored, i.e. both templates are NOT merged, it's always one or the other.

The current fields supported are:

  • nodeSelector: a selector which must be true for the pod to fit on a node, see Assigning pods to Nodes.
  • tolerations: allow (but do not require) the pods to schedule onto nodes with matching taints.
  • affinity: allow to constrain which nodes your pod is eligible to be scheduled on, based on labels on the node.
  • securityContext: pod-level security attributes and common container settings, like runAsUser or selinux.
  • volumes: list of volumes that can be mounted by containers belonging to the pod. This lets the user of a Task define which type of volume to use for a Task volumeMount
  • runtimeClassName: the name of a runtime class to use to run the pod.
  • automountServiceAccountToken: whether the token for the service account being used by the pod should be automatically provided inside containers at a predefined path. Defaults to true.
  • dnsPolicy: the DNS policy for the pod, one of ClusterFirst, Default, or None. Defaults to ClusterFirst. Note that ClusterFirstWithHostNet is not supported by Tekton as Tekton pods cannot run with host networking.
  • dnsConfig: additional DNS configuration for the pod, such as nameservers and search domains.
  • enableServiceLinks: whether services in the same namespace as the pod will be exposed as environment variables to the pod, similar to Docker service links. Defaults to true.
  • priorityClassName: the name of the priority class to use when running the pod. Use this, for example, to selectively enable preemption on lower priority workloads.
  • schedulerName the name of the scheduler to use when dispatching the Pod. This can be used when workloads of specific types need specific schedulers, e.g.: If you are using volcano.sh for Machine Learning Workloads, you can pass the schedulerName and have Tasks be dispatched by the volcano.sh scheduler.
  • imagePullSecret the name of the secret used when pulling the image if specified.
  • hostNetwork: set this to true if the host network namespace should be used. Defaults to false.

A Pod template can be specified for TaskRun or PipelineRun resources. See Specifying a Pod template for a TaskRun example or Specifying a Pod template for a PipelineRun example using Pod templates.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License.