The project dsc-traek is an Open Source implementation of a service to host the Powershell Desired State Configuration (DSC) Pull mode protocol.
This is a community supported project with no official support from Microsoft. Although the maintainers/contributors for the project can be Microsoft employees (and from the PowerShell DSC engineering team) that does not implicitly create a support relationship. There is no SLA for resolving issues or on call support.
If a supported and scalable solution is required, the recommended implementation of the Pull mode protocol is Azure Automation DSC.
For a fully supported but limited implementation of Pull service protocol for isolated network scenarios, consider Windows DSC Pull Service.
For information on how to use PowerShell DSC and how to configure clients to register with a Pull service, see the official PowerShell DSC documentation.
The journey for this project is to explore the possibility of using innovative approaches to cloud scale applications as a platform for delivering configuration management functionality.
The implementation is written in Node.js following best practices for micro-services with an expectation that the only release vehicle would be a container platform. The most likely hosting environment would be Kubernetes.
Please note that contributors to the project are learning Node.js and Kubernetes during the process of development, which makes this an exciting learning opportunity but likely to introduce breaking changes during early phases of development.
This solution is intended to provide three primary functions:
- a repository for node configurations (mof) and PowerShell modules containing DSC resources
- assignment of node configurations to target nodes
- reporting
The repository stores all configurations and resources required by target nodes. This streamlines the process of deploying configurations and associated DSC resources to all target nodes within your organization. You do not need to make sure that target nodes have the required resources before they can enact the configuration because a node will just get any missing resources from the pull server before applying the configuration.
Once a target node is configured to pull node configurations from a pull server, an Administrator no longer needs to directly access any target nodes in order to assign a node configuration to it. They simply assign the desired node configuration to the desired target node(s) and next time the target node communicates with the server, it will get the assigned configuration and apply it.
In order for any configuration management solution to be viable, it must provide visibility into configuration state across all managed nodes. The traek project provides data to support reporting with high level configuration status as well as details about each node's configuration so that you can quickly and easily understand the state of your environment and what if any steps need to be taken.
DSC on Stack Exchange DSC on Twitter
The DSC Pull Server is licensed under the MIT license. License
<Build status??> To learn how you can help make the DSC Pull Server a success visit here.