Readme

---

This is a personal fork for my own use case (LAN) so stuff may be changed to my own liking, or fixed up if I can figure out something.

You're not supposed to use this software in an official, company setting as that would be violation of Microsoft license agreement. However, feel free to use this if you need something for personal use, maybe in your home, maybe for your friends.

Fork changes

• Templates lightly translated to Polish and default theme changed to Dark theme
• Updated bulma.min.css to v1.0.2
• Remove hyperlink to License from main page
• Runs as rootless user by default with UID and GID of 1001
• Removed entrypoint.py
• Dockerfile updated to use alpine:edge and python 3.12, removed bash and shadow from base image, purge /var/cache/apt to free up space after updates and installation.
• Update package versions in requirements.txt
• Use /app instead of /home/py-kms
• Activates following versions of Windows Server:
  • [Fork] Windows Server 2025 [Azure Core, Datacenter Azure Edition, Datacenter, Standard)]
  • [Fork] Windows Server 2022 [Azure Core, Datacenter Azure Edition, Datacenter, Standard, Datacenter (Semi-Annual Channel), Standard (Semi-Annual Channel)]
  • Windows Server 2019 [Azure Core, Essentials, Datacenter, Standard, ARM64, Datacenter (Semi-Annual Channel), Standard (Semi-Annual Channel)]
  • Windows Server 2016 [Azure Core, Essentials, Datacenter, Standard, ARM64, Datacenter (Semi-Annual Channel), Standard (Semi-Annual Channel), Cloud Storage]
  • Windows Server 2012 R2 [Essentials, Datacenter, Standard, Cloud Storage]
  • Windows Server 2012 [Essentials, Datacenter, Standard, MultiPoint Premium, MultiPoint Standard]
  • Windows Server 2008 R2 A [MultiPoint Server 2010, Web, HPC Edition]
  • Windows Server 2008 R2 B [Standard, Enterprise]
  • Windows Server 2008 R2 C [Datacenter, Enterprise for Itanium]
  • Windows Server 2008 A [Web, Computer Cluster]
  • Windows Server 2008 B [Standard, Standard without Hyper-V, Enterprise, Enterprise without Hyper-V]
  • Windows Server 2008 C [Datacenter, Datacenter without Hyper-V, Enterprise for Itanium]
  • Windows Server Next [Preview Datacenter, Preview Standard, Preview Web, Preview ServerHI]
  • Windows 10 ServerRdsh VL [Enterprise multi-session]
• Activates following Volume Licensing versions of Windows:
  • Windows Vista
  • Windows 7
  • Windows 8
  • Windows 8.1
  • Windows 10
  • Windows 10 Insider Preview
  • Windows 11
  • Windows 11 Insider Preview
• Activates following Volume Licensing versions of Office:
  • Office 2013
  • Office 2013 (Pre-Release)
  • Office 2016 (+ Preview)
  • Office 2019 (+ Preview)
  • [Fork] Office LTSC 2021 (+ Preview)
  • [Fork] Office LTSC 2024 (+ Preview)

Network Ports

Important

Use IP environmental variable to enable IPv6 support if you require it. To do so, you set the environmental variable to following value IP=::. Otherwise, it will listen only on IPv4.

Following ports are used inside of the container:

• 1688/tcp for KMS server
• 8080/tcp for Web Server (Overview)

If you need different web ports, use --publish in podman run command, PublishPort= in Quadlet, or Ports: in Docker Compose. Relevant docs for Podman: podman run, systemd-unit, Docker Compose

podman run -d localhost/py-kms:latest -p 127.0.0.1:9012:8080/tcp

Quadlet configurations can be seen on my personal GitHub repository here

Volumes

You should mount the database file for data persistence. It's read by the Web Server component. Make sure that directory permissions match!

The directory you need to mount is /app/db and you can do so in following manner. You will need to create a directory inside of your users' home folder called AppData/2_PERSIST/PYKMS that is owned by following UID and GID: 1001:1001

mkdir -p ~/AppData/2_PERSIST/PYKMS

Quadlet

Warning

Quadlet functionality is only available since Podman version 5.2+. This feature is not present on Docker.

In case you want to let the container process fix up the permissions by itself, you can pass U flag to the Volume= argument. This is easiest way to solve permission problems with rootless containers.

Volume=${HOME}/AppData/2_PERSIST/PYKMS:/app/db:rw,Z,U

Here's a full config, based on my home one. Fully ready for production! :) Put it inside ~/.config/systemd/containers/PYKMS.container, reload systemd - systemctl --user daemon-reload and then start it - systemctl --user start PYKMS

[Unit]
Description=KMS - Key Managment Service, used for Windows & Office activations

[Service]
Restart=on-failure

[Install]
WantedBy=default.target

[Container]
# Base
Image=localhost/py-kms:latest
ContainerName=kms
Volume=${HOME}/AppData/2_PERSIST/PYKMS:/app/db:rw,Z
# Needed for private IPC
Tmpfs=/dev/shm
# Networks
Network=pasta:--ipv4-only
# KMS port
PublishPort=0.0.0.0:1688:1688/tcp
# Dashboard port
PublishPort=127.0.0.1:9012:8080/tcp

Docker-Compose

To bind mount inside the directory, you can do like so:

volumes:
    - ${HOME}/AppData/2_PERSIST/PYKMS:/app/db:rw,Z

Not providing full compose.yaml as I haven't touched it in well over a year. I'm a heavy Quadlet user right now.

Building the image locally

Clone this fork, it will be inside of py-kms directory

git clone https://github.com/Rubberverse/py-kms.git -b next

Change your directory

cd py-kms

Move Dockerfile from py-kms/docker/docker-py3-kms/Dockerfile to root of your current directory

mv py-kms/docker/docker-py3-kms/Dockerfile .

Start build with following command, replace podman with docker if you're using docker

podman build -f Dockerfile -t localhost/py-kms:latest

If all goes well, it should build successfully. Now you can reference it in your Quadlet or Docker Compose and deploy it! Just use localhost/py-kms:latest as your Image registry entry. Image=localhost/py-kms:latest

---

Original fork information

Here you can see the project information.

History

py-kms is a port of node-kms created by cyrozap, which is a port of either the C#, C++, or .NET implementations of KMS Emulator. The original version was written by CODYQX4 and is derived from the reverse-engineered code of Microsoft's official KMS. This version of py-kms is for itself a fork of the original implementation by SystemRage, which was abandoned early 2021.

Features

• Responds to v4, v5, and v6 KMS requests.
• Supports activating:
  • Windows Vista
  • Windows 7
  • Windows 8
  • Windows 8.1
  • Windows 10 ( 1511 / 1607 / 1703 / 1709 / 1803 / 1809 )
  • Windows 10 ( 1903 / 1909 / 20H1, 20H2, 21H1, 21H2 )
  • Windows 11 ( 21H2 )
  • Windows Server 2008
  • Windows Server 2008 R2
  • Windows Server 2012
  • Windows Server 2012 R2
  • Windows Server 2016
  • Windows Server 2019
  • Windows Server 2022
  • Microsoft Office 2010 ( Volume License )
  • Microsoft Office 2013 ( Volume License )
  • Microsoft Office 2016 ( Volume License )
  • Microsoft Office 2019 ( Volume License )
  • Microsoft Office 2021 ( Volume License )
  • It's written in Python (tested with Python 3.10.1).
  • Supports execution by Docker, systemd and many more...
  • Uses sqlite for persistent data storage (with a simple web-based explorer).

Documentation

The wiki has been completly reworked and is now available on readthedocs.io. It should provide you all the necessary information about how to setup and to use py-kms , all without clumping this readme. The documentation also houses more details about activation with py-kms and how to get GVLK keys.

Quick start

• To start the server, execute python3 pykms_Server.py [IPADDRESS] [PORT], the default IPADDRESS is :: ( all interfaces ) and the default PORT is 1688. Note that both the address and port are optional. It's allowed to use IPv4 and IPv6 addresses. If you have a IPv6-capable dual-stack OS, a dual-stack socket is created when using a IPv6 address. In case your OS does not support IPv6, make sure to explicitly specify the legacy IPv4 of 0.0.0.0!
• To start the server automatically using Docker, execute docker run -d --name py-kms --restart always -p 1688:1688 ghcr.io/py-kms-organization/py-kms.
• To show the help pages type: python3 pykms_Server.py -h and python3 pykms_Client.py -h.

License

• py-kms is