Files
DashBoard/docs/migration/portal-shell-route-view-integration/migration-observability-report.md

1.7 KiB

Migration Observability Dashboard/Report

Last updated: 2026-02-11

Monitoring Scope

  • Route errors: shell route 4xx/5xx, unknown-route fallback count, dynamic module load errors.
  • Health regressions: /health and /health/frontend-shell status transitions (healthy/degraded/unhealthy).
  • Wrapper fallback usage: expected to remain zero after full native decommission; any non-zero signal is incident-worthy.

Key Metrics

  1. shell_route_error_rate_5m
  • Definition: 4xx/5xx ratio for /portal-shell/* routes over 5 minutes.
  • Threshold: warning at 0.5%, critical at 1.0%.
  1. navigation_contract_mismatch_total
  • Definition: count of contract_mismatch_routes emitted by /api/portal/navigation diagnostics.
  • Threshold: must be 0.
  1. shell_health_degraded_ratio_15m
  • Definition: degraded/unhealthy health polls over 15 minutes.
  • Threshold: warning at 5%, critical at 10%.
  1. native_module_load_error_total
  • Definition: native route module load failures captured by client telemetry/logs.
  • Threshold: must be 0 for stable rollout.
  1. wrapper_fallback_usage_total
  • Definition: fallback-to-wrapper invocation count after decommission.
  • Threshold: must be 0.

Dashboard Panels

  • Panel A: Route errors by route id and render mode.
  • Panel B: Health summary state timeline with error/warning counts.
  • Panel C: Route contract mismatch and unknown-route fallback trend.
  • Panel D: Wave A/Wave B smoke pass trend and gate pass/fail timeline.
  • Panel E: Wrapper fallback usage (target line at zero).

Operational Notes

  • During canary/partial rollout, all panels must stay within threshold before progressing.
  • Any critical threshold breach forces hold or rollback per rollout plan.