Skip to main content

Data Model — golemsec_corp

The crown-jewel dataset. A single PostgreSQL database on db-pgmy (10.50.30.12:5432, database golemsec_corp) that models the whole finance business with realistic, interconnected data — the high-value target a real scanner/DLP engine should discover.

Sensitivity: this database intentionally holds rich PII and financial data (Israeli national IDs / teudat zehut, salaries in NIS, bank IBANs, health-fund membership, client holdings and trades). It is intentionally reachable at the data layer within the segmented data subnet.

Tables & volumes

TableRowsContents
employees200Full PII: national_id (teudat zehut), salary_nis, bank_iban, health_fund, contact, department
clients3,000Insurance & brokerage clients with PII
insurance_policies5,000Policies (life/health/auto/property) linked to clients
claims1,200Claims raised against policies
brokerage_accounts1,500Client brokerage accounts, managed by employees
holdings1,498Securities positions per account
trades20,000Executed trades
transactions10,000Cash movements

Total ≈ 42,000 rows of interlinked business data.

Generation

  • Seed script: D:\golemsec\provisioning\data\finance_seed.sql — idempotent, generate_series-based.
  • Load path: base64 → SSM → psql -d golemsec_corp.

Relationships

App-internal data vs. central data

The central golemsec_corp DB is fully seeded. The line-of-business apps (CRM, ERP, HR, helpdesk) are deployed and functional but several are lightly populated at the app layer — populating their internal records to match the central dataset is a visual-realism follow-up tracked in Build Status.

Other databases

EngineHostUse
PostgreSQL10.50.30.12:5432golemsec_corp (PII) + app DBs
MySQL10.50.30.12:3306App DBs (GLPI, SuiteCRM, OrangeHRM, Akaunting)
SQL Server 202210.50.30.38:1433MSSQL workload