Some Kubectl and Helm Examples for Superfacility Projects

Note: These are beta projects, use at your own discretion

Related Work

1. Andrew Naylor's user-space prometheus-based Workflow Monitoring Toolchain

Getting Started

Before installing Helm Charts, you will need:

1. Have a valid install of kubectl and helm -- we provide helper scripts that will download these to opt/bin;
2. Obtain the kubeconfig from the Spin cluster where your project runs (either the "development" or "production" cluster). Remeber to use the entrypoint script when running the scripts in this project.

Use the Entrypoint

We want to ensure that the scripts have have predictable behavior, regardless of the execution context (e.g. where scripts are being run from). Bash kinda sucks at this (especially when handling relative paths) -- an imperfect (but acceptable) solution is to use a launcher script: entrypoint.sh which takes the command to run as its arguments. This can be run either as a standalone script, or in interactive mode.

entrypoint.sh sets two environment variables __PREFIX__, and __DIR__:

• __PREFIX__ contains the location of the project root (i.e. where entrypoint.sh is saved)
• __DIR__ contains the location of he script being run

Standalone Mode

Standalone mode is invoked by passing the path to the script you want to run (relative the entrypoint.sh) as its arguments

./entrypoint.sh opt/util/get_kubectl.sh -n linux -a amd64 

Interactive Mode

Interactive mode is invoked whenever the command passed to entrypoint.sh is not a script in this project (i.e.the absolute path does not start with $__PREFIX__). Most often this is used to launch a shell:

./entrypoint.sh bash

This would drop you into a shell containing __PREFIX__.

Note

In interactive mode $__PREFIX__/opt/bin:$__PREFIX__/opt/util are appended to PATH. Furthemore __DIR__ is unset (which might cause some scripts -- which are not meant to operate in interactive mode -- to fail).

Download KUBECONFIG

KUBECONFIG is a YAML file containing the deteails of the k8s cluster, such as its address, and your own authentication credentials. It can be downloaded from the Spin.

1. Login to rancher2.spin.nersc.gov;
2. click the cluster name (as of May 12, 2024, either "development" or "production");
3. hover the mouse pointer over the "page" icon on the top right of the page, it should say "Download KubeConfig", click it to download.
4. create $HOME/.kube directory if not existing, and save the downloaded file to $HOME/.kube/config; alternatively, you can set the KUBECONFIG environment variable to the path to the downloaded YAML file.

(Optional) Create KUBECONFIGsecret

Using kubectl, create a kubeconfig secret in the targeted namespace (replace <targeted_namespace> and <path to kubeconfig> accordingly):

kubectl -n <targeted_namespace> create secret generic kubeconfig --from-file=kubeconfig=<path to kubeconfig>

Frequently used kubectl and helm Commands

List all namespaces in the cluster:

kubectl get namespaces

Get all developments in a given namespace:

kubectl get deployments -n <name of namespace>

Get logs of a specific development:

kubectl logs deployments/<deployment name> -n <name of namespace>

Get all pods in a namespace:

kubectl get pods -n <name of namespace>

Run command on pod:

kubectl exec <name of pod> -n <name of namespace> -- <command>

If you want to "drop into" interactive shell on pod

kubectl exec <name of pod> -n <name of namespace> -it -- /bin/bash

Install a Helm chart:

helm install <app name> ./<chart path> -n <name of namespace> [--debug] [--dry-run]

Delete an app installed with Helm:

helm uninstall <app name> -n <name of namespace>

Scale a deployment:

kubectl scale -n <name of namespace> --replicas <number of replicas> deployment/<name of deployment>

Create a new namespace in Spin

This guide describes how to create namespaces on Spin.

Create a database deployment on spin

Take a look at our examples for mysql and postgresql

Comunity Resources

Postgress

Setup

For a basic setup, see Digitical Ocean's tutorial on How to Deploy Postgres on K8s. An enterprise-scale deployment is described in Cloudraft's tutorial on How to Build Scalable and Reliable PostgreSQL Systems on Kubernetes.

Autoscaling

Neon has a Git repo on Postgres Autoscaling. A detailed description of setting up a scalabe Postgres cluster on K8s is given in this Medium article. Also note that K8s operators can be used for autscaling, for example Crunchy data's K8s operator