- E2E Testing for Main Connectors
- Helm Deployment Documentation
- K8s Operator for Multi-Transfer Deployments
Set up comprehensive end-to-end tests in the CI pipeline for the following main connectors:
- Postgres
- MySQL
- Clickhouse
- Yandex Database (YDB)
- YTsaurus (YT)
- Configure test environments in CI for each connector.
- Design E2E test scenarios covering various transfer modes (snapshot, replication, etc.).
- Automate test execution for all supported connectors.
- Set up reporting and logs for test failures.
Achieve fully automated E2E testing across all major connectors to ensure continuous integration stability.
Provide detailed documentation on deploying the transfer engine using Helm on Kubernetes clusters.
- Create Helm chart for easy deployment of the transfer engine.
- Write comprehensive Helm deployment guide.
- Define key parameters for customization (replicas, resources, etc.).
- Instructions for various environments (local, cloud).
- Test Helm deployment process on common platforms (GKE, EKS, etc.).
Enable seamless deployment of the transfer engine via Helm with clear and accessible documentation.
Develop a Kubernetes operator to manage multiple data transfers, simplifying the process for large-scale environments.
- Define CRD (Custom Resource Definitions) for transfer configurations.
- Implement operator logic for scaling and managing multi-transfer deployments.
- Add support for monitoring, scaling, and error recovery.
- Write user documentation for deploying and managing transfers via the operator.
Provide a scalable solution for managing multiple data transfers in Kubernetes environments with an operator.
- Q2-Q3: Focus on E2E testing for core connectors.
- Q3: Publish Helm deployment documentation and final testing.
- Q3-Q4: Develop and release the Kubernetes operator for multi-transfer management.
This roadmap aims to enhance testing, simplify deployment, and provide advanced scalability options for the transfer engine.