lab-orchestrator-lib 0.0.18

Manages Labs in your Kubernetes Cluster

Lab Orchestrator Lib

This package contains the lab orchestrator library.

Github
PyPi
Read The Docs

Installation

• pip3 install lab-orchestrator-lib

Documentation

Check out the developer documentation at laborchestratorlib.readthedocs.io.

Adapter

To use this library you need adapter classes. Adapter classes are used to connect the lab orchestrator lib to your database. How to write them is described in the adapter documentation

Adapter Libraries:

There is already one library that contains all adapters to use the lab orchestrator lib with django: LabOrchestratorLib-DjangoAdapter. This library also contains an example Django API.

The LabOrchestratorLib-FlaskSQLAlchemyAdapter project is not maintained and not working, but if you need an adapter for Flask-SQLAlchemy you can base them on this project.

Resources

The library contains two types of resources. The first is Kubernetes resources. This type contains NetworkPolicy, VirtualMachineInstance and Namespace objects. These are resources from Kubernetes that doesn't need to be saved.

The second type is database resources. They are saved in the database. For every database resource an adapter needs to be added. This type contains user, DockerImage, Lab and LabInstance objects.

A DockerImage is a link to a docker image that contains the VM image. A Lab contains one or multiple (currently not supported) VMs, each one is linked to a DockerImage. A lab can be started which results into a LabInstance. A LabInstance contains some VirtualMachineInstances running in a Namespace that can be accessed with VNC. The LabInstances are separated from other LabInstances with NetworkPolicys.

Controller

This library makes use of something called controllers. A controller is a class that controls one resource. The controllers can (and should) be used to create, get and update resources.

A controller collection is a collection of all controllers. You can create one with the lab_orchestrator_lib.controllers.controller_collection.create_controller_collection(...) function. This function takes all adapters, one api registry and a secret key for creating JWT tokens as parameter. The api registry is needed for the Kubernetes controllers and the adapters are injected into the database controllers.

Usage

To use this library create a APIRegistry and a controller collection. Than you can use the controllers in the controller collection to create new DockerImages, Labs and LabInstances.

For detailed information take a look at the developer documentation at laborchestratorlib.readthedocs.io.

Examples

An example of the implementation of the adapters and an example of how to used the controllers can be found in the LabOrchestratorLib-DjangoAdapter.

Contributing

Issues

Feel free to open issues.

Project Structure

The src folder contains the source code of the library. The tests folder contains the test cases. There is a makefile that contains some shortcuts for example to run the test cases and to make a release. Run make help to see all targets. The docs folder contains rst docs that are used in read the docs. Kubernetes yaml templates are placed in src/lab_orchestrator_lib/templates/.

Developer Dependencies

• Python 3.8
• Make
• pip install -r requirements.txt
• pip install -r requirements-dev.txt

Releases

Your part:

1. Create branch for your feature (issue/ISSUE_ID-SHORT_DESCRIPTION)
2. Code
3. Make sure test cases are running and add new ones for your feature
4. Create MR into master
5. Increase version number in src/lab_orchestrator_lib/__init__.py (semantic versioning)

Admin part:

1. Check and accept MR
2. Merge MR
3. Run make release

Docs

To generate the docs run: cd docs && make html.