Best Practices

This section describes the best practices and ESW recommendations that you can follow to ensure the smooth flow of data between your servers and ESW servers.

Transport security

• Expose your receiving endpoint over HTTPS.
• Use a valid, publicly trusted certificate.
• Ensure your server presents the full certificate chain.

ESW verifies SSL certificates when delivering payloads.

Delivery semantics (distributed systems reality)

• Expect retries on network or endpoint failures.
• Do not assume strict ordering of events.
• Expect duplicate deliveries in some failure modes.
• Make your handlers idempotent.
• Treat each delivery as independent and safe to process more than once.

ESW uses a queuing model and delivers events on a best-effort basis. Ordering is not guaranteed.

Support and troubleshooting

When you contact ESW about delivery or integration issues, include:

• Your environment (Sandbox or Production).
• Your endpoint URL and the approximate time window.
• The HTTP status codes you returned.
• The response body you returned (if any).
• Any request identifiers or correlation IDs (if provided).

Rate limiting and traffic spikes

Rate limits can turn a short spike into a long replay backlog. Plan for bursts.

• Align provider-level throttles with your expected peak traffic.
• Avoid blanket throttles that convert bursts into sustained failures (for example 429 / 5xx).
• Ensure your endpoint can absorb replay traffic after an outage.
• Use internal buffering (queue/worker pattern) so you can accept deliveries quickly.

Example failure loop (what to avoid)

You launch a new product. Traffic spikes. Your provider starts throttling requests.

ESW continues generating events. Deliveries fail (429/5xx). ESW queues and retries later.

Retries add more traffic. Throttling continues. The queue grows for hours.