Host CPU Usage
--%
Baseline Threshold: 85%
Host Memory
--%
Baseline Threshold: 85%
Response Latency
-- ms
SLA Target: < 800 ms
Failure Risk Score
--%
AI Model Classification
Real-Time CPU & Memory UtilizationLive 3s Polling
Network I/O Throughput (Mbps)psutil delta
Monitored Microservice Target: demo-microservice
Simulated Order & Payment Gateway workload integrated with automated self-healing.
Healthy
Disk Storage
--%
Inbound Network
-- Mbps
Outbound Network
-- Mbps
Active Connections
--
Application Response Latency (ms)
Monitored Cloud Services & Health Endpoints
| Service Name | Type | Port / URL | Status | Health Check Probe |
|---|---|---|---|---|
| Cloud Sentinel API Core | FastAPI Backend | :8000/api/health | RUNNING | HTTP 200 (OK) |
| Prometheus Metrics Exporter | Prometheus Client | :8000/metrics | SCRAPING | 200 OK text/plain |
| Order & Payment Microservice | Monitored Demo Workload | :8001/health | MONITORED | Active Liveness Probe |
Model Accuracy
100%
Scikit-learn Random Forest
Precision (Macro)
100%
Healthy / Warning / Critical
Recall (Macro)
100%
True Positive Detection Rate
Training Samples
6,000
Stratified 80/20 Train-Test Split
Model Feature Importances
Confusion Matrix Visualizer
Dynamic Anomaly Threshold Settings
Active Infrastructure Anomaly Alerts
| Alert ID | Timestamp | Metric | Current / Threshold | Severity | Description |
|---|
Controlled Failure Simulation Lab
Demonstrate actual failure detection, ML prediction, and automated recovery actions.
Approved Safe Remediation Actions
Self-Healing Execution Audit Trail & Probes
| Action ID | Timestamp | Service | Action Taken | Verification Status | Duration | Remediation Details |
|---|
Historical Failure & Anomaly Incidents
| Incident ID | Timestamp | Severity | Title | Status | Action |
|---|
Generative AI Provider Configuration
Note: When no API key is set, the system automatically uses the offline intelligent heuristic engine clearly labeled as fallback mode.
Prometheus & Grafana Endpoints
• Prometheus Metrics Exporter:
http://localhost:8000/metrics• Prometheus Server (Docker/K8s):
http://localhost:9090• Grafana Dashboard (Docker/K8s):
http://localhost:3000• Monitored Microservice Target:
http://localhost:8001