Setup

Teaching: 0 min · Exercises: 0 min · Total: 0 min

Prerequisites for NRP Training

Before attending the training session, please ensure you have completed the following setup steps.

Two ways to follow along

There are two ways to run the hands-on exercises in this training:

1. NRP USCMS Analysis Hub (recommended). A JupyterHub-based environment with the training materials, kubectl, and kubelogin already installed — nothing to set up on your laptop ahead of time. It still needs one login step done at the start of the session (see below). This is also the only path that works for the whole training: the later CMS Data Access lesson uses hub-only tools (grid-cert-import, grid-proxy-init) with no local equivalent, so anyone on their own machine ends up switching over for that lesson anyway.
2. Your own machine (alternative). You run kubectl locally against the Nautilus cluster. This only works if you've completed the setup steps below before the session — kubectl, kubelogin, and your kubeconfig all need to be installed and verified in advance. Useful if you prefer working in your own local environment, but see the caveats in that section below before committing to it.

Everything past this setup page — namespaces, kubectl commands, YAML manifests — is identical either way.

Jump to: NRP USCMS Analysis Hub (recommended) · Your own machine (alternative)

1. NRP Access Requirements

Institutional Account Access

• You must have an institutional account with NRP access via Authentik
  • You can use CERN account or institutional account, but choose one and stick with it
• NRP access is available to users from US academic institutions or those collaborating with US institutions
• If you don't have access, see Getting Started with NRP

Namespace Membership

• You must be part of at least one namespace to participate in the training
• Check your namespaces at: https://nrp.ai/namespaces/
• Students: Contact your research supervisor to be added to their namespace
• Faculty/Researchers: Request namespace admin status in Matrix

Method 1: NRP USCMS Analysis Hub (recommended)

You still need the account and namespace access from NRP Access Requirements above — this method just skips installing anything on your laptop. kubectl and kubelogin are already installed on the hub image; you only need to point kubectl at the cluster and log in once per session.

Get kubectl working in the hub terminal

Open a terminal in JupyterLab (File → New → Terminal) and run:

grid-kube-setup

This fetches a fresh kubeconfig from https://nrp.ai/config and installs it at ~/.kube/config (backing up anything already there). It doesn't log you in by itself — the next kubectl command you run triggers a device-code login:

kubectl get pods -n us-cms

kubelogin prints a URL and a short code. Open the URL on your laptop, enter the code, and the hub terminal picks up the token automatically — nothing more to click on the hub side.

jovyan@jupyter-...:~$ grid-kube-setup
Fetching   https://nrp.ai/config
user 'oidc':
    + --grant-type=device-code
    + --skip-open-browser

Backed up  /home/jovyan/.kube/config.xxxxxxxxxxx.bak
Updated    /home/jovyan/.kube/config

Next: run any kubectl command to trigger the login, e.g.

  kubectl get pods

kubelogin will print a URL and a code. Open the URL on your laptop, enter the
code, and this terminal will pick up the token.

jovyan@jupyter-...:~$ kubectl get pods -n us-cms
Please visit the following URL in your browser: https://authentik.nrp-nautilus.io/device?code=XXXXXXXXX
NAME                      READY   STATUS    RESTARTS   AGE
mlflow-667f8c984c-thzsr   2/2     Running   0          2d22h
mlflow-postgres-0         1/1     Running   0          10d

The login lasts for the rest of your hub session — you don't need to repeat this for every new terminal, only after grid-kube-setup re-installs the kubeconfig or your token expires.

Kubernetes Basics, the first hands-on lesson, starts with a one-time callout covering the rest of the hub setup (grid certificate, proxy, username) — do that once and the rest of the training just works.

Method 2: Your own machine (alternative)

Complete these steps before the session — they can't be done live.

curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl

brew install kubectl

curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/darwin/amd64/kubectl"
chmod +x kubectl
sudo mv kubectl /usr/local/bin/

brew install kubelogin

mkdir ~/.kube

args:
  - oidc-login
  - get-token
  - --browser-command="/mnt/c/Program Files/Google/Chrome/Application/chrome.exe"
  - --listen-port=18000
  - --token-cache-storage=disk

kubectl config get-contexts

kubectl config use-context nautilus

kubectl get nodes

kubectl get pods -n <YOUR_NAMESPACE>

kubectl config set-context nautilus --namespace <YOUR_NAMESPACE>

git clone --branch materials/cms-hats --single-branch https://github.com/nrp-nautilus/nrp-training.git ~/cms-hats
cd ~/cms-hats/workspace

cd ~/cms-hats
git pull
cd workspace

CLI Tools

Below are two tools which reduce the amount of kubernetes commands you need to type out.

• k9s: A console-like CLI tool
• Lens: More of a GUI CLI tool

These are not required. All parts of the tutorial and using Kubernetes in general can be done via the command line, however these tools make things easier. If you wish to use these PLEASE INSTALL PRIOR TO THE EXERCISE.

Getting Help

If you encounter issues during setup:

• Support Chat: Join NRP's Support Chat for community support
• Email: usersupport@nrp-nautilus.io
• Documentation: NRP Getting Started Guide

Additional Resources

• NRP Portal
• NRP Documentation
• Namespace Management