ɳSentry Architecture
هذا المحتوى غير متوفر بلغتك بعد.
ɳSentry Architecture
Section titled “ɳSentry Architecture”ɳSentry is a 13-plugin observability stack for production nSelf deployments. It separates concerns into monitoring (uptime), incident response, and error tracking.
Plugin Layers
Section titled “Plugin Layers”Layer 1: Monitoring (Baseline 7)
Section titled “Layer 1: Monitoring (Baseline 7)”nself-uptime-monitor (port 3831)
- HTTP/HTTPS endpoint checks every 1/5/15/60 min
- Latency tracking, SSL cert expiry alerts
- Geography-aware probes (multiple regions)
nself-synthetic-monitor (port 3836)
- Multi-step user flows (login → action → logout)
- JavaScript execution, form filling
- Screenshot + video recordings on failure
nself-rum (port 3837)
- Real User Monitoring from browsers
- PageView, error, performance beacons
- Error source mapping + session replay
Layer 2: Incident Response (Baseline 4)
Section titled “Layer 2: Incident Response (Baseline 4)”nself-status-page (port 3832)
- Public status page (white-labeled, custom domain)
- Incident timeline, maintenance windows
- Component status matrix
nself-incident-mgmt (port 3833)
- Open/acknowledge/resolve incidents
- Runbook links, post-mortems
- Integration with chat/email
nself-alert-router (port 3834)
- Alert aggregation + deduplication
- Routing rules (team, severity, service)
- Multi-channel delivery (Slack, email, PagerDuty)
nself-oncall (port 3840)
- On-call schedule management
- Escalation policies
- Alert auto-assignment to oncall
Layer 3: SLO + Errors (Baseline 2 + Expansion 4)
Section titled “Layer 3: SLO + Errors (Baseline 2 + Expansion 4)”nself-slo-tracker (port 3835)
- Service-level objective definitions
- Error budget tracking
- Burn-rate alerts
nself-errors (port 3838)
- Error ingestion from SDKs (web, iOS, Android, Go, Python)
- Source mapping (JS, TS)
- Error grouping + deduplication
nself-cron-monitor (port 3839)
- Cron job heartbeat + failure tracking
- Timeout alerts
- Job duration trends
nself-crash (port 3841)
- Native crash reporting (iOS, Android)
- Stack trace symbolication
- Crash trend analysis
nself-anomaly (port 3842)
- Anomaly detection on metrics
- Baseline establishment (1-week learn period)
- Alert on deviation
nself-audit (port 3843, free)
- Audit log storage (who changed what, when)
- Compliance reporting (SOC2, GDPR)
- Immutable append-only log
Data Flow
Section titled “Data Flow”App SDKs (errors, crashes, logs) ↓ nself-errors + nself-crash ↓ Postgres (np_errors table) ↓ Alert Router + Incident Mgmt ↓ Slack / Email / PagerDuty ↓ Status PagePorts & Services
Section titled “Ports & Services”| Plugin | Port | Role |
|---|---|---|
| uptime-monitor | 3831 | HTTP checks |
| status-page | 3832 | Public status |
| incident-mgmt | 3833 | Incident lifecycle |
| alert-router | 3834 | Alert routing |
| slo-tracker | 3835 | SLO tracking |
| synthetic-monitor | 3836 | User flow testing |
| rum | 3837 | Browser monitoring |
| errors | 3838 | Error ingestion |
| cron-monitor | 3839 | Job tracking |
| oncall | 3840 | On-call management |
| crash | 3841 | Crash reporting |
| anomaly | 3842 | Anomaly detection |
| audit | 3843 | Audit logs (free) |
All plugins are proxied through Nginx — access via sentry.yourdomain.com/<plugin-name>.