🔧 SDLC Fase 6: Maintenance & Operations

[ProgrammingDevelopment]

🔧 Maintenance & Operations

PRISMA — Platform Informasi & Sistem Manajemen RT 04

SDLC Fase 6: Maintenance

Versi Dokumen: 1.0
Tanggal: 10 April 2026

---

1. Monitoring

1.1 Health Check Schedule

Service	Metode	Interval	Alert
Frontend (SSG)	Telegram Bot ping	5 menit	Admin Telegram
API Gateway	/health endpoint	1 menit	Docker healthcheck
Warga Service	/health endpoint	1 menit	Docker healthcheck
Keuangan Service	/health endpoint	1 menit	Docker healthcheck
Keamanan Service	/health endpoint	1 menit	Docker healthcheck
Surat Service	/health endpoint	1 menit	Docker healthcheck
AI Service	/health endpoint	1 menit	Docker healthcheck
Ollama LLM	Port check 11434	5 menit	Bot alert

1.2 Log Management

Source	Location	Retention
Frontend	Browser console	Session
Gateway	docker logs gateway	7 hari
Microservices	docker logs <service>	7 hari
Security audit	localStorage security_audit_log	Max 1000 entries
Bot	Console output / PM2 logs	30 hari

---

2. Backup Strategy

Data	Metode	Frekuensi	Lokasi
SQLite DB (client)	SqliteDB.exportDB() → download .db	Manual / on-demand	User download
SQLite DB (services)	Docker volume backup	Harian	Host filesystem
JSON data files	Git version control	Per commit	GitHub
Environment vars	Separate secure storage	Per change	Password manager
Docker images	Docker registry	Per release	Docker Hub / GHCR

---

3. Incident Response

3.1 Severity Levels

Level	Deskripsi	Response Time	Contoh
P1 Critical	Sistem down total	< 1 jam	All services down
P2 High	Fitur utama tidak berfungsi	< 4 jam	Login gagal, data loss
P3 Medium	Fitur minor terganggu	< 24 jam	Chatbot error, UI bug
P4 Low	Cosmetic / enhancement	< 1 minggu	Typo, styling issue

3.2 Response Procedure

1. DETECT  → Health check alert / user report
2. ASSESS  → Determine severity level
3. CONTAIN → Isolate affected service (docker stop)
4. FIX     → Apply fix (code change or config)
5. TEST    → Verify fix in staging
6. DEPLOY  → Roll out fix (docker restart / redeploy)
7. REVIEW  → Post-mortem documentation

---

4. Regular Maintenance Tasks

4.1 Weekly

• Review audit log entries
• Check database sizes
• Verify all health checks green
• Review bot usage stats (/status)

4.2 Monthly

• Run npm audit for dependency vulnerabilities
• Update dependencies (npm outdated)
• Review and rotate API keys/tokens
• Backup all databases

4.3 Quarterly

• Full security review
• Performance benchmark (Lighthouse)
• Review and update documentation
• User feedback review

---

5. Scaling Procedure

5.1 Horizontal Scaling (Microservices)

# Scale specific service
docker-compose up -d --scale warga-service=3

# With load balancer (nginx)
# Update nginx.conf upstream block

5.2 Vertical Scaling

# docker-compose.yml — resource limits
services:
  warga-service:
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 256M

---

Dokumen ini adalah bagian dari SDLC Waterfall Phase 6 — Maintenance