Logging & Monitoring
Good logs make support faster. Capture enough detail to reproduce the failed request and confirm what happened to the order.
Mandatory Log Fields
| Field |
Why it matters |
trace_id / request_id |
Correlates frontend, backend, SDK, and support logs. |
| Client code or user ID |
Identifies impacted account. |
| Environment |
Separates UAT and PROD failures. |
| Endpoint and method |
Confirms API surface. |
| Order ID / basket ID |
Required for lifecycle debugging. |
| Strategy tag |
Connects related order legs and retries. |
| Request and response body |
Reproduces validation/backend failures. |
| Latency and retry count |
Shows performance and retry behavior. |
| WebSocket event timeline |
Confirms whether state updates arrived. |
Monitoring Metrics
- Error rate by source and endpoint.
- Order timeout percentage.
- Retry exhaustion rate.
- WebSocket reconnect frequency.
- RMS rejection ratio.
- API latency p95 and p99.
- Authentication failure rate.
Alert Examples
| Alert |
Suggested trigger |
| Order timeout spike |
More than 5% order timeout rate in 5 minutes. |
| Reconnect storm |
Repeated reconnects across many clients. |
| Trading API degradation |
Repeated 500 or 503 on trading endpoints. |
| Unknown order status |
Order status is still unclear after repeated checks. |
| Auth failure spike |
Sudden increase in 401/403/440 responses. |
Support Bundle
Environment:
Endpoint:
HTTP method:
Timestamp with timezone:
Trace ID / request ID:
Client code:
Order ID / basket ID:
Strategy tag:
Request body:
Response status:
Response body:
Retry count:
WebSocket events:
Latest orderbook result:
