Why VIES Is Unreliable - and How We Built Around It

Building and maintaining a reliable service like EuroValidate that interfaces with unreliable or inconsistent third-party services such as VIES (EU VAT Validation Service) requires careful design and implementation of several strategies to ensure robustness, availability, and usability for end-users. Here are the key aspects and considerations based on your description:

1. Graceful Degradation

EuroValidate ensures that it never returns a bare HTTP 500 error by providing cached data with reduced confidence levels when the upstream service (VIES) is down or unavailable. This approach allows clients to make informed decisions about how to proceed, rather than crashing or presenting an uninformative error message.

2. Retry Mechanisms

Implementing retry logic with exponential backoff and jitter helps manage transient errors effectively:

• Exponential Backoff: Increases the wait time between retries exponentially.
• Jitter: Adds randomness to the delay intervals to prevent a synchronized surge of retries when the service becomes available again, which could otherwise cause another overload.

3. Caching

Using both in-memory and persistent caching (e.g., Redis) ensures that data is readily available even if the upstream service is temporarily unavailable:

• In-Memory Cache: For quick

Read the full article at DEV Community

---

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.