Resilience: hbf-notifications

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

HTTP Retry

• Library: got (v11.8.5) + hbf-core-api (inherited)
• Attempts: 3 (hbf-core-api only); direct got calls have no retry
• Backoff: Exponential with jitter (hbf-core-api only), network errors only
• On failure: hbf-core-api returns HBFCoreApiResponse with 503; AllExceptionsFilter returns structured error

Queue Retry (if applicable)

N/A

Timeouts

Call	Timeout	Configured in
All HTTP calls	Not set	N/A

Circuit Breakers

None.

Fallback Strategy

Failure scenario	Behaviour	User impact
hbf-core unreachable	Returns HBFCoreApiResponse with 503 status	Notification delivery fails gracefully
Unhandled exception	AllExceptionsFilter returns structured error response	Client receives error details

Known Gaps

• hbf-core-api retry triggers only on network errors, not on 5xx responses.
• No timeout configured on any HTTP request (requests may hang indefinitely).
• No health endpoint.
• Direct got calls (outside hbf-core-api) have no retry at all.