Skip to content

Latest commit

 

History

History
153 lines (112 loc) · 5.92 KB

README.md

File metadata and controls

153 lines (112 loc) · 5.92 KB
Raw

Building Kind cluster with everything installed

Pre-requirements

Supported platforms

A Kind cluster is not intended for production purposes; it is only meant as a local execution example. However, running a Kind Kubernetes cluster with KubeFlow pipelines (KFP) and local data storage (Minio) requires significant memory. Therefore, we recommend deploying it on machines with at least 32 GB of RAM and 8-9 CPU cores. RHEL OS requires more resources, e.g. 64 GB RAM and 32 CPU cores.

Note: for MacOS users, see the following comments

Preinstalled software

The following programs should be manually installed:

  • Helm 3.10.0 or greater must be installed and configured on your machine.
  • Kind tool for running local Kubernetes clusters 0.14.0 or newer must be installed on your machine.
  • Kubectl 1.26 or newer must be installed on your machine.
  • MinIO Client (mc) must be installed on your machine. Please choose your OS system, and process according to "(Optional) Install the MinIO Client". You have to install the mc client only.
  • git client, we use git client to clone installation repository
  • lsof usually it is part of Linux or MacOS distribution.
  • Container agent such as Docker or Podman

NOTE: Additionally, ensure that nothing is running on port 8080, which is used by the Kind cluster ingress.

If you do not want to upload the testing data into the local Minio, and reduce memory footprint, please set:

export POPULATE_TEST_DATA ?= 0

Preparing Kind cluster for testing

This is a manual build instruction. As an alternative, you can execute the make setup makefile rule in the project kind directory instead. make setup performs complete installation, including validation that port 8080 is available, creation of the cluster, installing required software (NGNIX, KubeRay and KFP), creating ingresses and secrets and loading local data to Minio.

Create cluster

Run the following command to create the cluster:

cd /tmp
git clone git@github.com:IBM/data-prep-kit.git
cd data-prep-kit
ROOT_DIR=$PWD/kind/
kind create cluster --name dataprep --config ${ROOT_DIR}/hack/kind-cluster-config.yaml

Note that by default this will create a kind cluster with 2 worker nodes. If you would like a different amount of node, modify cluster configuration

Install KFP

Install Kubeflow Pipelines and wait for it to be ready:

PIPELINE_VERSION=1.8.5
cd /tmp
git clone https://github.com/kubeflow/pipelines.git
cd pipelines
git checkout tags/${PIPELINE_VERSION}
kubectl apply -k manifests/kustomize/cluster-scoped-resources
kubectl wait --for condition=established --timeout=60s crd/applications.app.k8s.io
sed -i.back '/inverse-proxy$/d' manifests/kustomize/env/dev/kustomization.yaml
kubectl apply -k manifests/kustomize/env/dev
kubectl wait --for=condition=ready --all pod -n kubeflow --timeout=300s

Add permissions with RBAC role:

kubectl create clusterrolebinding pipeline-runner-extend --clusterrole cluster-admin --serviceaccount=kubeflow:pipeline-runner

Install KubeRay

Install Kuberay:

helm repo add kuberay https://ray-project.github.io/kuberay-helm/
helm repo update
helm install kuberay-operator kuberay/kuberay-operator -n kuberay --version 1.0.0 --set image.pullPolicy=IfNotPresent --create-namespace

Next install API server and wait for it to be ready:

helm install -f ${ROOT_DIR}/hack/ray_api_server_values.yaml kuberay-apiserver kuberay/kuberay-apiserver --version 1.1.0 -n kuberay
kubectl wait --for=condition=ready --all pod -n kuberay --timeout=120s

Install NGNIX

To access the API server and Kubeflow pipeline UI externally, we make use NGINX ingress.

Install Ingress NGNIX and wait for it to be ready:

kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/kind/deploy.yaml

kubectl wait --namespace ingress-nginx \
          --for=condition=ready pod \
          --selector=app.kubernetes.io/component=controller \
          --timeout=90s

To deploy the ingress for ray apiserver, kfp and Minio execute the following:

kubectl apply -f $ROOT_DIR/hack/ray_api_server_ingress.yaml
kubectl apply -f $ROOT_DIR/hack/kfp_ingress.yaml
kubectl apply -f $ROOT_DIR/hack/minio_ingress.yaml

Open the Kubeflow Pipelines UI at http://localhost:8080/kfp/

Finally a secret need to be created for accessing Minio using the following command:

kubectl apply -f $ROOT_DIR/hack/s3_secret.yaml

Working with a Minio server instead of S3 storage

You can work with a real S3 storage, but for testing you can use the Mino server which is deployed as part of the KFP installation.

Copy test data

Populating Minio server with test data can be done using mc. Use the following command:

./hack/populate_minio.sh

This file create an mc alias, create the test bucket and copy local data to Minio. If you need to load additional data, please load it using additional mc commands, similar to the ones being used by populate_minio.sh

Finally, note, that kfp version of Minio, is using a username (minio) as an access key and password ('minio123') as the secret. The secret to use for Minio access is located in kubeflow ns with the name s3-secret

Cleanup

delete the cluster:

kind delete cluster --name dataprep

alternatively you can execute

make clean