Deploy and Host Hasura on Sealos

Hasura is a high-performance GraphQL API engine for your data. This template deploys Hasura GraphQL Engine with PostgreSQL metadata storage and a Data Connector Agent on Sealos Cloud.

About Hosting Hasura

Hasura provides instant GraphQL APIs, metadata-driven schema management, role-based authorization, and event/webhook integrations on top of your data services. In this template, Hasura runs as the primary API and console service and connects to a managed PostgreSQL cluster for metadata.

The deployment includes two application workloads: hasura/graphql-engine and hasura/graphql-data-connector. The Data Connector Agent provides connector endpoints for federated and external data backends, while Hasura keeps metadata and runtime state in PostgreSQL provisioned by Kubeblocks.

Sealos also provisions HTTPS ingress, a public domain, persistent database storage, and Kubernetes-native lifecycle management in Canvas.

Common Use Cases

• Internal API Platform: Build a unified GraphQL API for internal tools and dashboards.
• SaaS Backend Acceleration: Expose PostgreSQL-backed GraphQL endpoints quickly for product development.
• Composable Data Gateway: Connect multiple data systems through Hasura metadata and data connectors.
• Event-Driven Workflows: Trigger webhooks and async business processes from data changes.
• Prototype to Production: Start with one-click deployment and scale through Canvas resource controls.

Dependencies for Hasura Hosting

The Sealos template includes all required dependencies: Hasura GraphQL Engine, Hasura Data Connector Agent, and a managed PostgreSQL 16.4 cluster with persistent storage.

Deployment Dependencies

• Hasura Documentation - Official product and operation guides
• Hasura GraphQL Engine Repository - Source code and release history
• Hasura Data Connectors - Connector and integration concepts
• Hasura Discord Community - Community support and discussions

Implementation Details

Architecture Components:

This template deploys the following services:

• Hasura GraphQL Engine: Main API and Console service exposed via HTTPS ingress.
• Data Connector Agent: Companion service for connector endpoints on port 8081.
• PostgreSQL (Kubeblocks): Metadata database cluster (postgresql-16.4.0) with persistent storage.

Configuration:

• Hasura reads database connection fields from Kubeblocks-managed secret keys (host, port, username, password).
• The startup command composes a full PostgreSQL DSN before launching graphql-engine.
• Ingress is provisioned with TLS and a public domain, and the default deployment is single replica for both workloads.

License Information:

Hasura GraphQL Engine is licensed under the Apache License 2.0.

Why Deploy Hasura on Sealos?

Sealos is an AI-assisted Cloud Operating System built on Kubernetes that unifies development, deployment, and operations. By deploying Hasura on Sealos, you get:

• One-Click Deployment: Launch Hasura, PostgreSQL, and connector components without manual YAML orchestration.
• Managed Kubernetes Runtime: Keep Kubernetes reliability with less operational complexity.
• Easy Customization: Update environment variables and resources through Canvas dialogs and resource cards.
• Instant HTTPS Access: Get a public URL with TLS for Hasura Console and API access.
• Persistent Database Storage: Store metadata safely with built-in persistent volumes.
• Scale on Demand: Adjust CPU, memory, and replicas from Canvas as traffic grows.
• AI-Assisted Ops: Describe desired changes in the AI dialog and apply updates quickly.

Deploy Hasura on Sealos to focus on API delivery instead of infrastructure management.

Deployment Guide

1. Open the Hasura template and click Deploy Now.
2. Configure deployment parameters in the popup dialog.
3. Wait for deployment to complete (typically 2-3 minutes). After deployment, you will be redirected to Canvas.
4. Access the generated URL and open Hasura Console:
   • Hasura Console: https://<your-domain>/console
   • Version/Health Check: https://<your-domain>/v1/version and https://<your-domain>/healthz

Configuration

After deployment, configure Hasura through:

• AI Dialog: Describe required changes (for example enabling admin secret or changing runtime flags).
• Resource Cards: Edit Deployment environment variables, probes, and resource limits.
• Hasura Console: Manage metadata, permissions, and tracked data sources.

Recommended production baseline:

• Set HASURA_GRAPHQL_ADMIN_SECRET.
• Set CORS and authentication-related settings based on your environment.
• Disable development mode (HASURA_GRAPHQL_DEV_MODE=false) for production workloads.

Scaling

To scale this deployment:

1. Open the application in Canvas.
2. Select the Hasura and Data Connector Deployments.
3. Increase CPU/Memory and replica count as needed.
4. Apply changes and monitor pod health and response latency.

Troubleshooting

Common Issues

Issue: Hasura pod restarts during initial deployment

• Cause: PostgreSQL cluster initialization may still be in progress.
• Solution: Wait until PostgreSQL status is Running and check Hasura pod logs again.

Issue: Cannot access Console

• Cause: Ingress propagation or TLS issuance is not completed yet.
• Solution: Wait briefly, then verify ingress host and certificate status in Canvas.

Issue: Metadata DB connection errors after custom changes

• Cause: Database-related env vars were overridden with incomplete values.
• Solution: Keep DB fields sourced from Kubeblocks secret keys and preserve DSN composition logic.

Getting Help

• Hasura Docs
• Hasura GitHub Issues
• Sealos Discord

Additional Resources

• Hasura Deployment Guide
• Hasura Metadata API
• Hasura Permissions Guide

License

This Sealos template is provided under the template repository license. Hasura GraphQL Engine itself is licensed under Apache License 2.0.