pykube-ng 0.20

Features

• HTTP interface using requests using kubeconfig for authentication

• Python native querying of Kubernetes API objects

Installation

To install pykube, use pip:

pip install pykube-ng

Interactive Console

The pykube library module can be run as an interactive console locally for quick exploration. It will automatically load ~/.kube/config to provide the api object, and it loads pykube classes (Deployment, Pod, ..) into local context:

python3 -m pykube
>>> [d.name for d in Deployment.objects(api)]

Usage

Query for all ready pods in a custom namespace:

import operator
import pykube

api = pykube.HTTPClient(pykube.KubeConfig.from_file("~/.kube/config"))
pods = pykube.Pod.objects(api).filter(namespace="gondor-system")
ready_pods = filter(operator.attrgetter("ready"), pods)

Access any attribute of the Kubernetes object:

pod = pykube.Pod.objects(api).filter(namespace="gondor-system").get(name="my-pod")
pod.obj["spec"]["containers"][0]["image"]

Selector query:

pods = pykube.Pod.objects(api).filter(
    namespace="gondor-system",
    selector={"gondor.io/name__in": {"api-web", "api-worker"}},
)
pending_pods = pykube.objects.Pod.objects(api).filter(
    field_selector={"status.phase": "Pending"}
)

Watch query:

watch = pykube.Job.objects(api, namespace="gondor-system")
watch = watch.filter(field_selector={"metadata.name": "my-job"}).watch()

# watch is a generator:
for watch_event in watch:
    print(watch_event.type) # 'ADDED', 'DELETED', 'MODIFIED'
    print(watch_event.object) # pykube.Job object

Create a Deployment:

obj = {
    "apiVersion": "apps/v1",
    "kind": "Deployment",
    "metadata": {
        "name": "my-deploy",
        "namespace": "gondor-system"
    },
    "spec": {
        "replicas": 3,
        "selector": {
            "matchLabels": {
                "app": "nginx"
            }
        },
        "template": {
            "metadata": {
                "labels": {
                    "app": "nginx"
                }
            },
            "spec": {
                "containers": [
                    {
                        "name": "nginx",
                        "image": "nginx",
                        "ports": [
                            {"containerPort": 80}
                        ]
                    }
                ]
            }
        }
    }
}
pykube.Deployment(api, obj).create()

Delete a Deployment:

obj = {
    "apiVersion": "apps/v1",
    "kind": "Deployment",
    "metadata": {
        "name": "my-deploy",
        "namespace": "gondor-system"
    }
}
pykube.Deployment(api, obj).delete()

Check server version:

api = pykube.HTTPClient(pykube.KubeConfig.from_file())
api.version

Requirements

• Python 3.6+

• requests (included in install_requires)

• PyYAML (included in install_requires)

Local Development

You can run pykube against your current kubeconfig context, e.g. local Minikube:

pipenv install --dev
pipenv run python3
>>> import pykube
>>> config = pykube.KubeConfig.from_file()
>>> api = pykube.HTTPClient(config)
>>> list(pykube.Deployment.objects(api))

To run PEP8 (flake8) checks and unit tests including coverage report:

make test

License

The code in this project is licensed under the Apache License, version 2.0 (included in this repository under LICENSE).

Contributing

Easiest way to contribute is to provide feedback! We would love to hear what you like and what you think is missing. Create an issue or ping try_except_ on Twitter.

PRs are welcome. Please also have a look at issues labeled with “help wanted”.

Code of Conduct

In order to foster a kind, inclusive, and harassment-free community, this project follows the Contributor Covenant Code of Conduct.