Resilience: hbf-client-integrations

Error handling and retry patterns for this service. Platform-wide patterns: docs/architecture/resilience.md

HTTP Retry

• Library: got with retry enabled for GET, POST, PUT, PATCH, DELETE
• Attempts: 3 (got default)
• Backoff: Exponential with jitter (got default, not explicitly configured)
• On failure: HttpException thrown with status code; 409 CONFLICT treated as non-fatal in post()

Queue Retry (if applicable)

N/A

Timeouts

Call	Timeout	Configured in
General requests	DISTRIBUTER_SERVICE_REQUEST_TIMEOUT env var or 5000ms fallback	Environment config
Client module requests	DISTRIBUTER_SERVICE_REQUEST_TIMEOUT env var or 15000ms fallback	Environment config
kelly-hbf-interaction	60000ms (60s)	Hardcoded
getBuffer()	No timeout	N/A

Circuit Breakers

None.

Fallback Strategy

Failure scenario	Behaviour	User impact
HTTP error	HttpException thrown with upstream status code	Error propagated to caller
409 CONFLICT on POST	Treated as non-fatal, request continues	No user impact

Known Gaps

• No timeout on getBuffer() calls (may hang indefinitely on large payloads).
• Backoff configuration relies on got defaults without explicit tuning.
• No circuit breaker across 15+ client services in /distributor/clients/.
• No DLQ for failed requests.
• No health endpoint.