kfp-kubernetes 2.16.1

Kubernetes platform configuration library and generated protos.

Kubeflow Pipelines SDK kfp-kubernetes API Reference

The Kubeflow Pipelines SDK kfp-kubernetes python library (part of the Kubeflow Pipelines project) is an addon to the Kubeflow Pipelines SDK that enables authoring Kubeflow pipelines with Kubernetes-specific features and concepts, such as:

• Secrets
• PersistentVolumeClaims
• ImagePullPolicies
• Ephemeral volumes
• Node selectors
• Tolerations
• Labels and annotations
• and more

Be sure to check out the full API Reference for more details.

Installation

The kfp-kubernetes package can be installed as a KFP SDK extra dependency.

pip install kfp[kubernetes]

Or installed independently:

pip install kfp-kubernetes

Getting started

The following is an example of a simple pipeline that uses the kfp-kubernetes library to mount a pre-existing secret as an environment variable available in the task's container.

Secret: As environment variable

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_secret():
    import os
    print(os.environ['SECRET_VAR'])

@dsl.pipeline
def pipeline():
    task = print_secret()
    kubernetes.use_secret_as_env(task,
                                 secret_name='my-secret',
                                 secret_key_to_env={'password': 'SECRET_VAR'})

Other examples

Here is a non-exhaustive list of some other examples of how to use the kfp-kubernetes library. Be sure to check out the full API Reference for more details.

Secret: As mounted volume

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_secret():
    with open('/mnt/my_vol') as f:
        print(f.read())

@dsl.pipeline
def pipeline():
    task = print_secret()
    kubernetes.use_secret_as_volume(task,
                                    secret_name='my-secret',
                                    mount_path='/mnt/my_vol')

Secret: As optional source for a mounted volume

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_secret():
    with open('/mnt/my_vol') as f:
        print(f.read())

@dsl.pipeline
def pipeline():
    task = print_secret()
    kubernetes.use_secret_as_volume(task,
                                    secret_name='my-secret',
                                    mount_path='/mnt/my_vol'
                                    optional=True)

ConfigMap: As environment variable

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_config_map():
    import os
    print(os.environ['CM_VAR'])

@dsl.pipeline
def pipeline():
    task = print_config_map()
    kubernetes.use_config_map_as_env(task,
                                 config_map_name='my-cm',
                                 config_map_key_to_env={'foo': 'CM_VAR'})

ConfigMap: As mounted volume

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_config_map():
    with open('/mnt/my_vol') as f:
        print(f.read())

@dsl.pipeline
def pipeline():
    task = print_config_map()
    kubernetes.use_config_map_as_volume(task,
                                       config_map_name='my-cm',
                                       mount_path='/mnt/my_vol')

ConfigMap: As optional source for a mounted volume

from kfp import dsl
from kfp import kubernetes

@dsl.component
def print_config_map():
    with open('/mnt/my_vol') as f:
        print(f.read())

@dsl.pipeline
def pipeline():
    task = print_config_map()
    kubernetes.use_config_map_as_volume(task,
                                       config_map_name='my-cm',
                                       mount_path='/mnt/my_vol',
				       optional=True)

PersistentVolumeClaim: Dynamically create PVC, mount, then delete

from kfp import dsl
from kfp import kubernetes

@dsl.component
def make_data():
    with open('/data/file.txt', 'w') as f:
        f.write('my data')

@dsl.component
def read_data():
    with open('/reused_data/file.txt') as f:
        print(f.read())

@dsl.pipeline
def my_pipeline():
    pvc1 = kubernetes.CreatePVC(
        # can also use pvc_name instead of pvc_name_suffix to use a pre-existing PVC
        pvc_name_suffix='-my-pvc',
        access_modes=['ReadWriteOnce'],
        size='5Gi',
        storage_class_name='standard',
    )

    task1 = make_data()
    # normally task sequencing is handled by data exchange via component inputs/outputs
    # but since data is exchanged via volume, we need to call .after explicitly to sequence tasks
    task2 = read_data().after(task1)

    kubernetes.mount_pvc(
        task1,
        pvc_name=pvc1.outputs['name'],
        mount_path='/data',
    )
    kubernetes.mount_pvc(
        task2,
        pvc_name=pvc1.outputs['name'],
        mount_path='/reused_data',
    )

    # wait to delete the PVC until after task2 completes
    delete_pvc1 = kubernetes.DeletePVC(
        pvc_name=pvc1.outputs['name']).after(task2)

PersistentVolumeClaim: Create PVC on-the-fly tied to your pod's lifecycle

from kfp import dsl
from kfp import kubernetes

@dsl.component
def make_data():
    with open('/data/file.txt', 'w') as f:
        f.write('my data')

@dsl.pipeline
def my_pipeline():
    task1 = make_data()
    # note that the created pvc will be autoamatically cleaned up once pod disappeared and cannot be shared between pods
    kubernetes.add_ephemeral_volume(
        task1,
        volume_name="my-pvc",
        mount_path="/data",
        access_modes=['ReadWriteOnce'],
        size='5Gi',
    )

Pod Metadata: Add pod labels and annotations to the container pod's definition

from kfp import dsl
from kfp import kubernetes


@dsl.component
def comp():
    pass


@dsl.pipeline
def my_pipeline():
    task = comp()
    kubernetes.add_pod_label(
        task,
        label_key='kubeflow.com/kfp',
        label_value='pipeline-node',
    )
    kubernetes.add_pod_annotation(
        task,
        annotation_key='run_id',
        annotation_value='123456',
    )

Kubernetes Field: Use Kubernetes Field Path as enviornment variable

from kfp import dsl
from kfp import kubernetes


@dsl.component
def comp():
    pass


@dsl.pipeline
def my_pipeline():
    task = comp()
    kubernetes.use_field_path_as_env(
        task,
        env_name='KFP_RUN_NAME',
        field_path="metadata.annotations['pipelines.kubeflow.org/run_name']"
    )

Timeout: Set timeout in seconds defined as pod spec's activeDeadlineSeconds

from kfp import dsl
from kfp import kubernetes

@dsl.component
def comp():
    pass

@dsl.pipeline
def my_pipeline():
    task = comp()
    kubernetes.set_timeout(task, 20)

ImagePullPolicy: One of "Always" "Never", "IfNotPresent".

from kfp import dsl
from kfp import kubernetes

@dsl.component
def simple_task():
    print("hello-world")

@dsl.pipeline
def pipeline():
    task = simple_task()
    kubernetes.set_image_pull_policy(task, "Always")

SecurityContext: Set security context for the task's container

from kfp import dsl
from kfp import kubernetes

@dsl.component
def my_task():
    print("running with custom security context")

@dsl.pipeline
def pipeline():
    task = my_task()
    kubernetes.set_security_context(
        task,
        run_as_user=1000,
        run_as_group=3000,
    )

Note: Cluster administrators can configure default runAsUser and runAsGroup values for all customer workload containers via the pipeline-install-config ConfigMap (defaultSecurityContextRunAsUser and defaultSecurityContextRunAsGroup). When admin defaults are configured, SDK-specified values for those fields are overridden.

ImagePullSecrets: Set secrets to authenticate image pulls

from kfp import dsl
from kfp import kubernetes

@dsl.container_component
def hello():
    return dsl.ContainerSpec(
        image='some-private-image:tag',
        command=['echo', 'hello']
    )

@dsl.pipeline
def pipeline():
    task = hello()
    kubernetes.set_image_pull_secrets(task, ['secret-name'])