In Kubernetes with Helm

Helm charts allows you to configure, install, and upgrade Kaskada within a Kubernetes cluster.

Prerequisites

Helm 3 or above. See Installing Helm.
A running Kubernetes cluster

Kaskada-Canary Helm Chart

The Kaskada Canary Helm Chart installs a persistant version of the kaskada service in the simplest configuration possible.

In this setup, the kaskada service runs as two containers in a single pod, and persists its database as a file on attached storage. It can store data locally, but ideally it should have access to an object store.

This chart is primarily intended for initial testing and should not be used in production scenarios.

Install the Kaskada Canary Helm Chart

Add Kaskada’s Helm repository to your local Helm installation:

helm repo add kaskada https://kaskada.io/helm-charts

Update the chart repository:

helm repo update

If desired, override settings from the values.yaml file. For example, the following config configures s3 as the object store using AWS access-key/secret as the authentication mechanism, and adds a load balancer to the service for ingress:

auth:
  aws:
    accessKeyId: <your_access_key_id>
    secretAccessKey: <your_secret_access_key>
    region: <s3_bucket_region>

storage:
  objectStore:
    bucket: <s3_bucket_name>
    type: s3
  dbData:
    storageClassName: gp2
  tmpData:
    storageClassName: gp2

service:
  type: LoadBalancer

Deploy the chart:
- Deploy with the defined configuration:

helm install kaskada-canary kaskada/kaskada-canary -f values.yaml