Skip to main content

High-level Kubernetes resource configuration and management library.

Project description

PyPI version Documentation Status build status coverage report


kuber is Python library for the management of Kubernetes resources. It's ideal for for collectively managing groups of resources throughout their lifecycle. Resource definitions can be created and managed entirely in Python code (the pure-Python approach), but kuber is most effective when used in a hybrid fashion that combines configuration files and Python code. kuber also integrates and maintains compatibility with the lower-level official Kubernetes Python client, while abstracting basic CRUD operations into higher level constructs more inline with the behaviors of tools like kubectl and helm.

Key Functionality

Here are some key things that kuber does well:

  • A flexible workflow for managing Kubernetes resource configuration in Python code.
  • The ability to load resources directly from YAML or JSON configuration files, modify them in code and then use them or save them back to YAML/JSON files.
  • Resource bundling for managing groups of resource configurations collectively.
  • CRUD operations exposed directly on the resource objects to reduce the overhead in managing low-level clients.
  • Convenience functions that simplify common operations, e.g. managing containers within pods from the root resource.
  • Very thorough type-hinting and object structure to support creating accurate resource configurations and catch errors before runtime.
  • All resources and sub-resources support used in with blocks as context managers to simplify making multiple changes to a sub-resource.
  • Simultaneous support for multiple Kubernetes API versions. Manage multiple Kubernetes API versions (e.g. while promoting new versions from development to production) without having to pin and switch Python environments.


kuber available for installation with pip:

$ pip install kuber


kuber can be used to manage individual resources or a group of resources collectively. kuber is also very flexible about how resources are created - either directly from Python or by loading and parsing YAML/JSON configuration files. The first example shows the multi-resource management path:

import kuber
from kuber.latest import apps_v1

# Create a bundle and load all resource definitions from the
# `app_configs` directory as well as the `app-secret.yaml`
# configuration file from the local `secrets` directory.
resource_bundle = (

# Modify the metadata labels on all resources in the bundle.
for resource in resource_bundle.resources:

# Update the replica count of the loaded deployment named
# "my-app" to the desired initial count.
d: apps_v1.Deployment = resource_bundle.get(
d.spec.replicas = 20

# Load the current `kubeconfig` cluster configuration into
# kuber for interaction with the cluster.

# Turn this bundle script into a file that can be called from
# the command line to carry out CRUD operations on all the
# resources contained within it collectively. For example,
# to create the resources in this bundle, call this script
# file with a create argument.

Or managing resources individually:

from kuber.latest import batch_v1

job = batch_v1.Job()

# Populate metadata using context manager syntax for brevity.
with job.metadata as md: = 'my-job'
    md.namespace = 'jobs'

# Add a container to the job spec.
    env=[batch_v1.EnvVar('ENVIRONMENT', 'production')]

# Print the resulting YAML configuration for display. This
# could also be saved somewhere to use later as the
# configuration file to deploy to the cluster in cases
# like a multi-stage CI pipeline.

Check out the kuber documentation for more details and examples.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kuber-1.8.1.tar.gz (2.3 MB view hashes)

Uploaded Source

Built Distribution

kuber-1.8.1-py2.py3-none-any.whl (4.8 MB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page