About this project

This repository provides a Helm chart for deploying Qualytics, a closed-source, container-native platform for enterprise data quality assessment, monitoring, and facilitation. The chart deploys a single-tenant instance to a CNCF-compliant Kubernetes cluster. Architecture: The deployment is split into a Controlplane (API, CMD services, and Frontend UI), a Dataplane (a Spark application with driver and executor pods), and the external Datastores it monitors. PostgreSQL holds platform state (in-cluster StatefulSet by default or external), and RabbitMQ serves as the message broker. Datastores are configured after deployment via the UI, supporting JDBC (Snowflake, BigQuery, Redshift, Databricks, PostgreSQL, Oracle, SQL Server) and DFS (S3, GCS, ADLS) connectors. Prerequisites: Kubernetes 1.30+, kubectl, Helm 3.12+, a Qualytics-issued image registry token, and a unique deployment identifier. Authentication can be database-backed (recommended) or Auth0. Usage: Work with a Qualytics account manager to obtain installation materials. Steps include creating a CNCF-compliant cluster (Terraform templates for AWS EKS, GCP GKE, Azure AKS are provided), creating a configuration file from template.values.yaml, setting required values (deployment identifier, DNS record, authentication, security secrets), deploying via Helm, configuring DNS and TLS, and activating a license. Air-gapped deployments are supported, with the only ongoing internet dependency being Auth0 for federated auth (optional). Container images, datastore access, and JDBC driver resolution (via Maven Central or internal repository) must be planned. Documentation covers authentication, Docker image mirroring, custom Maven repositories, AWS identity, observability, license management, and cluster sizing.