Enterprise Monitor Dashboard
Unified command center for system health, security incidents, and data recovery.
On this page
The Enterprise Monitor is the unified command center for SveltyCMS administrators. It provides a real-time, “Single Pane of Glass” view into the system’s security posture, operational health, and recovery options.
🚀 Overview
The dashboard is accessible via Config > Enterprise Monitor and uses a high-performance polling engine to keep metrics current without excessive database load.
The 4 Core Quadrants
| Quadrant | Metrics Tracked | Admin Action |
|---|---|---|
| 🛡️ Security Status | Active incidents, blocked IPs, CSP violations. | Investigate threat feed / Unblock IPs. |
| 🗑️ Trash Recovery | Deleted item count, storage impact. | Restore content (purge is not implemented) |
| 🪝 Webhook Health | Active endpoints, failure rates, delivery latency. | Review logs / Trigger retries. |
| ⚙️ System Vitals | RAM/CPU usage, DB pool health, Watchdog status. | Monitor for resource exhaustion. |
🔬 Enterprise Performance Auditing (Audit)
While the Enterprise Monitor provides real-time health data, SveltyCMS also maintains a High-Fidelity Performance Matrix (Audit) for baseline verification.
- Deterministic Reliability: The monitor surfaces health, but the Performance Matrix ensures that the core system meets the 10ms persistence budget across all supported databases.
- CI/CD Integration: Every release is audited against the Performance Benchmarks to prevent regressions in latency or bundle size.
🛡️ Security Incident Feed
The monitor features a real-time security feed that logs more than just errors. It captures:
- Unauthorized Access Attempts: Triggered by the FLAC engine when unauthorized field writes are blocked.
- Firewall Events: Brute-force detection and pattern-based blocking.
- CSP Reports: Browser-reported violations of the Content Security Policy.
🗑️ Fast-Path Recovery
Integrated directly into the monitor is the Quick Restore utility.
- Mangle-on-Delete: SveltyCMS mangles unique fields (like slugs) upon deletion. The monitor allows you to see exactly which items are recoverable and restores the original unique identifiers.
- One-Click Restore:
POST /api/trash/restorereverses the “trashed” state. The list rendersdeletedAt/deletedBywhen present and falls back toupdatedAt/updatedByfor rows trashed before those columns existed. - No scheduled purge: restore is implemented; a scheduled or automatic 30-day purge is not shipped. Nothing removes a trashed row on a timer — an operator has to delete it permanently through the collection.
⚙️ Autonomous Maintenance Feedback
The Enterprise Monitor surfaces the activity of the Autonomous System Watchdog:
- Cache Compaction: Displays when the system last performed an automated cache cleanup (triggered at 30% fragmentation).
- Index Health: Reports on database query performance and suggests manual index optimizations if slow queries are detected.
- Uptime:
metricsService.getSystemMetrics().uptimeis milliseconds (Date.now() - startTime). The monitor renders it withformatUptimeMs()(@utils/format-uptime) →29m,1d 2h 3m,< 1m, orJust startedfor a zero/invalid value. Both functions are unit-explicit:formatUptime()in@utils/datetakes seconds and is a different formatter.
📊 Performance & Polling
- Glassmorphism UI: Uses modern blur-heavy aesthetics optimized for readability in low-light environments.
- Smart Polling: The dashboard polls the server every 30 seconds.
- L1/L2 Cache Integration: Metrics are retrieved primarily from the MetricsService in-memory buffer, resulting in < 5ms response time for the monitoring API.