Kube Forwarder is a Kubernetes port forwarding manager designed to simplify the process of managing port forwarding configurations for Kubernetes clusters. The tool automates port forwarding tasks, allowing users to easily connect local machines to Kubernetes services without manual intervention.
Key Features:
Auto-reconnect functionality that monitors connection status and attempts reconnection on failure
Support for multiple clusters, enabling users to bookmark and forward services across different environments
Import/export capability to share or backup bookmarks with team members
Zero native dependencies, eliminating the need to install kubectl or explain complex commands to developers
Optional integrations with Google Cloud SDK and AWS CLI
Audience & Benefit:
Ideal for developers, DevOps engineers, and Kubernetes operators who need to manage port forwarding configurations. Kube Forwarder streamlines the process of connecting local development environments to Kubernetes services, reducing manual effort and improving team collaboration.
Available via winget package manager for easy installation.
Before you start forwarding internal resources to your local machine, you have to add cluster configuration.
To do this we have 3 different options in the app:
Auto-detection of ~/.kube/config file and parsing settings from it
Manual adding of Kubernetes config by selecting a file(s)
Manual adding of Kubernetes config by pasting a text
Import of the JSON file that could be generated via Kube Forwarder export functionality
When you add a new cluster via auto-detection (option 1) or manually using a file(a) selection (option 2), we could parse
configs and if there are multiple contexts inside we will suggest you to add multiple clusters to the app.
Few examples of yaml files we expect to have you could find
Also, you could add a cluster by filling a form manually (option 3). The form has the following fields:
Name - the name of a cluster withing Kube Forwarder app.
Storing method (Set destination to your kube config or paste it as a text) - the method of storing a config It has two options:
Set a path - storing a path to the config file. It will be read every time when you forwarding a port. It allows
a user to don't do any changes in Kube Forwarder's settings when a third-party app updates the config file.
For example, when azure-cli updates an access token (#13).
Paste as a text - storing a config just as a yml text.
Path (if storing method is Set a path) - the path to a config file.
Content (if storing method is Paste as a text) - Yml config as a text.
Current Context (if storing method is Set a path) - When you use Set a path, you must select a context from a file
which will be used to connect to a resource. Let's see an example of a problem that the field solves.
Let's say we don't have Current context field.
A user has a config file with two contexts: local-cluster and remote-cluster.
current-context in the yml file is local-cluster.
The user configured a cluster in Kube Forwarder with Set a path option.
The user created a resource postgres and successfully forwarded ports for some time.
Then the user executed kubectl config use-context remote-cluster
If the user tries to forward the resource in Kube Forwarder again, most likely there will be an error
since a connection will be established with remote-cluster, not local-cluster as the user expected,
and remote-cluster couldn't have postgres resource.
So, to avoid the error we should store the current context in a separate field.
Add a resource
Kube Forwarder supports forwarding of all types of resources that supported by kubectl – Pod, Deployment, Service.
We ask you to fill the form with the following fields:
Cluster Name - pick a cluster from one of the added clusters.
Namespace - the namespace of the resource you plan to forward.
Kind – pick one of the options Pod, Deployment or Service.
Name - name of the Pod, Deployment or Service.
Alias - alternative name of the resource that will be displayed on the homepage(optional)
Port Forwarding
Local port - port from your local machine where the resource will be forwarded. Note that ports <= 1024 are
restricted to user root
Resource port - port of the resource from the Kubernetes cluster
Use Custom Local Address - Check this and put an IP address or hostname into the text field to
use a different listen address. Putting each service on its own address avoids sharing/collisions between
services on cookies and port number. Specify a loopback address like 127.0.x.x or add entries to your
hosts file like 127.0.1.1 dashboard.production.kbf and put the assigned name in this column. If blank or
unchecked, localhost / 127.0.0.1 will be used.
Import/Export
Kube Forwarder allows you export cluster configuration in JSON that you could use to share with your team members or for the backup purpose. You could easily store it on Github. When you export cluster, you could export it with or without confidential information.
Install with Homebrew
brew cask install kube-forwarder
Contributing
We encourage you to contribute to Kube Forwarder!
We expect contributors to abide by our underlying code of conduct.
All conversations and discussions on GitHub (issues, pull requests)
must be respectful and harassment-free.
This project was generated with electron-vue@8fae476 using vue-cli. Documentation about the original structure can be found here.
How to contribute
Fork the project & clone locally. Follow the initial setup here.
Create a branch, naming it either a feature or bug: git checkout -b feature/that-new-feature or bug/fixing-that-bug
Code and commit your changes. Bonus points if you write a good commit message: git commit -m 'Add some feature'
Push to the branch: git push origin feature/that-new-feature
# Clone source code
git clone https://github.com//kube-forwarder
# install dependencies
npm install
# prepare .env files
cp .env.example .env
cp .env.example .env.production
# serve with hot reload in Electron Dev app
npm run dev
# serve WEB version with hot reload at localhost:9081
npm run web
Build
Build an application for production
# Build a target for current OS
npm run build
# Build a target for Windows
npm run build -- -- --win
# Build a target for Linux
npm run build -- -- --linux
# You can mix targets
npm run build -- -- --win --linux
# You can build static and target separately
npm run build:dist
npm run build:target -- --win
A built version will be appear in build directory.
Running the tests
We are using Cypress to run integration tests.
There are visual regression tests. It's important to run them inside docker
container to get same screenshots as in Drone CI.
npm run test:cypress
Or you can run it manually on a local machine.
# Run the web version to test it
npm run web
# Run this command in a separate terminal tab
npm run test:cypress:onhost
# Or you can open Cypress GUI
npm run test:cypress:open
Troubleshooting
Q) Node Sass could not find a binding for your current environment: OS X 64-bit with Node.js 12.x
A) npm rebuild node-sass
Q) Error: spawn .../kube-forwarder/node_modules/electron/dist/Electron.app/Contents/MacOS/Electron ENOENT
A) Reinstall node_modules: rm -rf node_modules && npm i
Release guide
Configure environment
Also, this steps could be used to configure CI environment.
Copy .env.example to .env.production and fill variables.
Release steps
Update the version in package.json and Push to release branch.
Run npm run release on a Mac computer to build packages. They will be automatically pushed to releases at Github.
Go to Releases in the repository.
Make sure that the created draft is OK and release it (Edit -> Release).
