fix(hardening): close HAIP production readiness gaps - #360
raedaltawil19-cmyk wants to merge 3 commits into
Conversation
- make deposit authorization failures explicit and staff-visible - alert on channel sync failures without notification storms - add Prometheus metrics, Grafana dashboard, and alert rules - enforce production Keycloak and secret configuration - add owner isolation probes and guest property checks - route Connect modifications through locked canonical inventory checks
|
Hey — thanks for putting this together. Really appreciate the attempt to close those production-readiness gaps; the direction on inventory locking, deposit-auth visibility, channel sync alerts, tenant isolation, and Keycloak/prod config is solid. We’re going to pass on merging this as one PR though. Bundling concurrency, deposit alerts, channel alerts, metrics, and Keycloak/prod changes together makes subtle breaks hard to catch in review. Could you reopen (or open fresh PRs) as smaller, focused changes — ideally one issue per PR (#306, #308, #309, #312, #314, plus Keycloak separately)? That way we can land and verify each piece cleanly. One note for the metrics PR specifically: please keep /metrics from being world-readable by default (network restriction and/or app-level protection), not only a reverse-proxy footnote. Thanks again — happy to review the split versions. |
Hardens HAIP booking concurrency, deposit failure visibility, channel sync alerts, tenant isolation, production Keycloak configuration, and observability.
Verification:
Addresses #306, #308, #309, #312, and #314.