Skip to content

Advanced Topics

This section dives deeper into resilience, failure handling, and direct access to YDB services. Use it when you need precise control over execution behavior, recovery, and performance.

Retries & Idempotency

  • Explains which error codes are retryable and when retries are safe.
  • Shows how .idempotent(true) affects single‑call retries and how to design idempotent business logic in transactions.
  • Covers retry strategy considerations for Topic streaming.
  • See: /advanced/retries.

Error Handling

  • Describes SDK error types (e.g., YDBError, CommitError) and how to distinguish them.
  • Demonstrates structured error logging and extracting issues for diagnostics.
  • Provides patterns for handling aborts and timeouts consistently.
  • See: /advanced/errors.

Low‑level Clients (driver)

  • How to create gRPC clients via driver.createClient(ServiceDefinition).
  • When and why to use low‑level services (Discovery, Scheme, etc.).
  • Auth, TLS/mTLS considerations for service clients.
  • See: /advanced/driver-low-level.