Makhzon 360 · RB-DBLAYER

Supabase PostgreSQL Data Layer — incident runbook

Runbook hubMakhzon console

The Supabase PostgreSQL data layer backs every Makhzon engine. Latency here amplifies into POS, inventory, marketplace and reporting at the same time, so treat it as a platform-wide incident from the first warning.

Warning at p95 query latency ≥ 350 ms, critical at ≥ 900 ms. Critical means user-visible impact: page the escalation owner.

dashboard: sim://grafana/db-layerlogs: sim://logs/db-layerowner: Data Platformrunbook: RB-DBLAYER

Signals & thresholds

  • Primary metric: p95 query latency (ms)
  • Warning threshold: ≥ 350 ms
  • Critical threshold: ≥ 900 ms
  • Dashboard: sim://grafana/db-layer
  • Log source: sim://logs/db-layer
  • SLA: 99.95% monthly availability
  • SLO: p95 query < 350ms
  • External dependencies: Supabase PostgreSQL / APIs / Realtime

1. Detection

  1. Database health monitor reports p95 query latency at or above the warning threshold for two consecutive evaluation windows.
  2. Operators report slow screens or timeouts for the affected area.

2. Triage

  1. Open the database dashboard and compare p95 query latency, active connections and CPU against the previous hour.
  2. Check for long-running or blocking queries and identify the top statements by total time.
  3. Confirm whether connection pool saturation or a lock wait chain is present.
  4. Check whether a migration, index build, bulk import or report job started just before the spike.
  5. Correlate with Performance / RPC Monitoring to see which RPCs degraded first.

3. Mitigation

  1. Cancel or terminate identified runaway queries (heaviest first) and record the statement for follow-up.
  2. Pause non-critical background jobs: reporting rebuilds, bulk sync, reconciliation runs.
  3. Reduce report concurrency and stop ad-hoc exports until latency recovers.
  4. If pool saturation is the cause, lower per-worker pool size or scale the compute tier.
  5. If a specific query pattern is the cause, apply the prepared index or query fix from the owning team.

4. Verification

  1. Confirm p95 query latency has been below the warning threshold for at least 10 minutes.
  2. Confirm the affected business services are healthy on the dashboard.
  3. Confirm no alerts remain open for this component and its downstream consumers.
  4. Spot-check one real end-to-end operation through the affected flow.

5. Rollback / restore

  1. If a deploy caused the incident, roll back to the previous release and re-verify.
  2. If a configuration or threshold change caused it, restore the previous value and record why.
  3. Re-enable any job, feature flag or channel that was paused during mitigation, one at a time.

Escalation path

  • First responder: NOC on duty — acknowledge within 5 minutes.
  • Component owner: Data Platform.
  • Escalation owner (page on critical): Data Platform On-Call.
  • Escalate if the metric is still critical 15 minutes after mitigation, or if data integrity is in doubt.
  • Declare a major incident if two or more critical components are alerting at the same time.

Blast radius — business services affected (60)

Identity & LoginStore / Tenant ManagementMulti-Store / Group ManagementBranch / Location ManagementStaff ManagementRoles & PermissionsProduct CatalogUnits & PacksVariantsBatch ManagementSerial Number ManagementBarcode OperationsInventory LedgerWarehouse InventoryStock TransferStocktake & AdjustmentSales / POSSales InvoicingCustomer ManagementCustomer CreditCollectionsDebt AgingCollection Follow-UpCRMLoyaltyPromotionsSupplier ManagementPurchase OrdersPurchasingSupplier AccountSupplier PaymentsSupplier GovernanceWholesale CatalogB2B Order ManagementB2B NegotiationB2B ReceivingTrade Relationship ManagementRental ManagementExpense ManagementGeneral Ledger AccountingCost CentersProfitabilityFinancial ReportingVAT ManagementZATCA E-InvoicingE-Commerce IntegrationE-Commerce Order ImportSubscription ManagementEmail NotificationPush NotificationSupport / HelpdeskSupport SLA & QualityAudit & ComplianceMonitoring & DiagnosticsSearchStore AlertsData ImportsSMS MessagingReporting & AnalyticsPlatform Administration

External dependencies

Supabase PostgreSQL / APIs / Realtime — Primary data plane, RPC and realtime channels

Simulated incidents that exercise this runbook

Supabase DB degradation — Primary Postgres data plane latency spike; RPC timeouts across engines.

Reporting notes

  • Record start time, actions taken and recovery time in the incident record.
  • Attach the dashboard screenshot and the dominant error message to the postmortem.
  • Content drafted for the monitoring demo — confirm the exact commands with the owning team before production use.