This web application is responsible for allowing the user to manipulate the Model Servers (Endpoints) in his Kubeflow cluster. To achieve this it provides a user friendly way to handle the lifecycle of InferenceService CRs.
The web application currently works with v1beta1 versions of InferenceService objects.
The web application is installed alongside the other KServe components, either in the kserve or in the kubeflow namespace. There is a VirtualService that exposes the application via an Istio Ingress Gateway. Depending on the installation environment the following Ingress Gateway will be used.
| Installation mode | IngressGateway |
|---|---|
| Standalone KServe | knative-ingress-gateway.knative-serving |
| Kubeflow | kubeflow-gateway.kubeflow |
To access the application you will need to navigate with your browser to
${INGRESS_IP}/models/Alternatively you can access the application via kubectl port-forward. In that case you will need to configure the application to:
- Not perform any authorization checks, since there is no logged in user
- Work under the
/prefix - Disable Secure cookies, since the app will be exposed under plain http
You can apply the mentioned configurations by doing the following commands:
# edit the configmap
# CONFIG=config/overlays/kubeflow/kustomization.yaml
CONFIG=config/base/kustomization.yaml
vim ${CONFIG}
# Add the following env vars to the configMapGenerator's literals
# for kserve-models-web-app-config
- APP_PREFIX=/
- APP_DISABLE_AUTH="True"
- APP_SECURE_COOKIES="False"
# reapply the kustomization
# kustomize build config/overlays/kubeflow | kubectl apply -f -
kustomize build config/base | kubectl apply -f -The following is a list of environment variables that can be set for any web app that is using this base app.
| Environment variable | Default value | Description |
|---|---|---|
| APP_PREFIX | /models | Controls the application's prefix, by setting the base-url element |
| APP_DISABLE_AUTH | False | Controls whether the application should use SubjectAccessReviews to ensure the user is authorized to perform an action |
| APP_SECURE_COOKIES | True | Controls whether the app should use Secure CSRF cookies. By default the application expects to be exposed with https |
| CSRF_SAMESITE | Strict | Controls the SameSite value of the CSRF cookie |
| USERID_HEADER | kubeflow-userid | Header in each request that will contain the username of the logged in user |
| USERID_PREFIX | "" | Prefix to remove from the USERID_HEADER value to extract the logged in user name |
The frontend is build with Angular and the backend is written with the Python Flask framework.
This web application is utilizing common code from the kubeflow/kubeflow repository. We want to enforce the same user experience across our different Kubeflow web applications and also keep them in the same development state. In order to achieve this the applications will be using this shared common code.
This will require us to fetch this common code when we want to either build the application locally or in an OCI container image.
In order to run the application locally you will need to:
- Build the frontend, in watch mode
- Run the backend
The npm run build:watch command will build the frontend artifacts inside the backend's static folder for serving. So in order to see the application you will need to start the backend and connect to localhost:5000.
Requirements:
- node 22.0.0
- python 3.12
# build the common library
COMMIT=$(cat ./frontend/COMMIT)
cd $KUBEFLOW_REPOSITORY/components/crud-web-apps/common/frontend/kubeflow-common-lib
git checkout $COMMIT
npm i
npm run build
cd dist/kubeflow
npm link
# run the app frontend
cd $KSERVE_MODELS_WEB_APPLICATION_REPOSITORY/frontend
npm i
npm link kubeflow
npm run build:watch# create a virtual environment and install dependencies
# https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/
cd $KSERVE_MODELS_WEB_APPLICATION_REPOSITORY/backend
python3.12 -m pip install --user virtualenv
python3.12 -m venv web-application-development
source web-application-development/bin/activate
# install the dependencies on the activated virtual environment
KUBEFLOW_REPOSITORY="/path/to/kubeflow/kubeflow" make -C backend install-deps
# run the backend
make -C backend run-dev
