Feat/monitoring dashboard

[Mrchinedum]

fixed #380

feat: implement comprehensive monitoring dashboard

No comprehensive monitoring dashboard existed. The platform had
Prometheus + Grafana infrastructure but no sustainability-aware
panels, no platform KPI gauges, and no alert rules for contract-level
health. This change closes that gap end-to-end.

Changes

indexer/src/performance/metrics.service.ts

• Declare 8 new Prometheus Gauge fields for sustainability KPIs
• Register all 8 gauges in the constructor with descriptive help text:
  teachlink_contract_sustainability_invocations_total
  teachlink_contract_sustainability_storage_writes_total
  teachlink_contract_sustainability_events_emitted_total
  teachlink_contract_sustainability_rewards_distributed_total
  teachlink_contract_sustainability_content_minted_total
  teachlink_contract_sustainability_active_users_total
  teachlink_contract_sustainability_efficiency_score
  teachlink_contract_sustainability_health_score
• Add updateSustainabilityMetrics(m) method to set all 8 gauges
  in a single call, keeping Prometheus state in sync with each query

indexer/src/reporting/dashboard.service.ts

• Add getSustainabilitySnapshot(): computes real-time KPIs from the
  existing getCurrentAnalytics() result:
  efficiencyScore — bridge success rate as basis-point proxy
  healthScore — weighted composite (50% efficiency,
  25% low dispute rate, 25% reward activity)
  escrowDisputeRateBps — disputes / total escrows * 10 000
  rewardClaimRate — claimed / total rewards * 10 000
  Calls metricsService.updateSustainabilityMetrics() on every
  invocation so Prometheus gauges stay current without a separate
  scrape job

indexer/src/reporting/reporting.controller.ts

• Add GET /analytics/sustainability endpoint that delegates to
  getSustainabilitySnapshot() and returns the KPI object

indexer/observability/prometheus/alerts.yml

• Add teachlink-sustainability alert group with 5 rules:
  TeachLinkLowEfficiencyScore efficiency < 7000 bps, 15m, warning
  TeachLinkCriticalEfficiencyScore efficiency < 5000 bps, 5m, critical
  TeachLinkLowHealthScore health < 60/100, 15m, warning
  TeachLinkHighEscrowDisputeRate API 5xx rate > 10%, 10m, warning
  TeachLinkNoNewTransactions invocations flat 30m, 30m, warning

indexer/observability/grafana/dashboards/teachlink-monitoring-dashboard.json (new)

• 691-line Grafana dashboard (uid: teachlink-monitoring-dashboard)

• Auto-refresh every 30s, default window now-6h

• 4 collapsible row sections, 20 panels total:

  Row 1 — Real-Time Platform Health (6 stat panels)
  Sustainability Health Score (0-100, color thresholds 60/80)
  Contract Efficiency Score (0-10000 bps, thresholds 7000/9000)
  Indexer Availability (up + probe_success, 0-2)
  Ledger Lag (seconds, thresholds 300/900)
  HTTP Error Rate (%, thresholds 5%/10%)
  API Avg Latency (seconds, thresholds 0.5/1.0)

  Row 2 — Historical Trends (4 time-series panels)
  Sustainability Scores Over Time (health score, efficiency %)
  Contract Resource Usage Over Time (invocations, writes, events)
  HTTP Throughput (req/s by route + status)
  Platform Growth Over Time (content minted, active users,
  rewards distributed in XLM)

  Row 3 — Alert Management (3 panels)
  Firing Alerts time-series (ALERTS{alertstate="firing"})
  Active Alert Count stat (color thresholds 1/3)
  Critical Alerts stat (color thresholds 1/2)

  Row 4 — Platform Insights (5 panels)
  Dashboard Cache Hit Ratio (percentunit time-series)
  API Latency Percentiles (avg, p95, p99 time-series)
  Dependency Health (database, horizon, indexer_state)
  Indexer Progress & Errors (events processed, errors, ledger)
  HTTP Status Code Breakdown (2xx / 4xx / 5xx by route)

Acceptance criteria met

✓ Real-time metrics — 6 live stat panels + Prometheus gauges updated
on every /analytics/sustainability call
✓ Historical trends — 4 time-series panels covering scores, resource
usage, throughput, and platform growth
✓ Alert management — 5 new Prometheus alert rules + 3 alert panels
in the dashboard
✓ Platform insights — cache efficiency, latency percentiles,
dependency health, indexer progress, HTTP
status breakdown

[drips-wave]

@Mrchinedum Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits