Merge pull request #40 from se4ai2526-uniba/fix-grafana-prom

docker/nginx.conf

@@ -81,6 +81,15 @@ http {
             proxy_set_header Host $host;
         }
 
+        # FastAPI Metrics endpoint for Prometheus
+        location /metrics {
+            proxy_pass http://fastapi/metrics;
+            proxy_set_header Host $host;
+            proxy_set_header X-Real-IP $remote_addr;
+            proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+            proxy_set_header X-Forwarded-Proto $scheme;
+        }
+
         # Grafana
         location = /grafana {
             return 301 /grafana/;

docs/design_choices.md

@@ -465,11 +465,12 @@ async def monitor_requests(request, call_next):
 ### Grafana Visualization
 
 **Dashboard Panels:**
-1. API Request Rate (time series)
-2. API Latency Percentiles (heatmap)
-3. Drift Detection Status (stat panel)
-4. Drift P-Value Trend (time series)
-5. Error Rate (gauge)
+1. Request Rate (gauge)
+2. Request Latency p50/p95 (time series)
+3. In-Progress Requests (stat panel)
+4. Error Rate 5xx (stat panel)
+5. Model Prediction Time (time series)
+6. Requests by Endpoint (bar chart)
 
 **Data Sources:**
 - Prometheus: Real-time metrics

docs/milestone_summaries.md

@@ -242,10 +242,12 @@ Jobs:
 
 ### Grafana Dashboards
 
-- **API Request Rate**: Real-time requests per second
-- **API Latency**: P50, P90, P99 percentiles
-- **Drift Detection Status**: Binary indicator (0/1)
-- **Drift P-Value**: Statistical significance metric
+- **Request Rate**: Real-time requests per second
+- **Request Latency (p50, p95)**: Response time percentiles
+- **In-Progress Requests**: Currently processing requests
+- **Error Rate (5xx)**: Failed request percentage
+- **Model Prediction Time**: Inference latency
+- **Requests by Endpoint**: Traffic distribution
 
 ### Data Drift Detection
 

docs/user_guide.md

@@ -406,11 +406,12 @@ The pre-configured dashboard includes:
 
 | Panel | Description |
 |-------|-------------|
-| API Request Rate | Real-time requests per endpoint |
-| API Latency | Response time distribution |
-| Drift Detection Status | Binary indicator (0=No Drift, 1=Drift) |
-| Drift P-Value | Statistical significance |
-| Drift Distance | KS test distance metric |
+| Request Rate | Real-time requests per second |
+| Request Latency (p50, p95) | Response time percentiles |
+| In-Progress Requests | Currently processing requests |
+| Error Rate (5xx) | Percentage of failed requests |
+| Model Prediction Time | Average model inference latency |
+| Requests by Endpoint | Traffic distribution per endpoint |
 
 ### Data Drift Detection
 

monitoring/grafana/dashboards/hopcroft_dashboard.json

@@ -62,7 +62,7 @@
       "pluginVersion": "9.0.0",
       "targets": [
         {
-          "expr": "rate(fastapi_requests_total[1m])",
+          "expr": "sum(rate(hopcroft_requests_total[1m]))",
           "refId": "A"
         }
       ],
@@ -131,12 +131,12 @@
       "pluginVersion": "9.0.0",
       "targets": [
         {
-          "expr": "histogram_quantile(0.95, rate(fastapi_request_duration_seconds_bucket[5m])) * 1000",
+          "expr": "histogram_quantile(0.95, sum(rate(hopcroft_request_duration_seconds_bucket[5m])) by (le)) * 1000",
           "legendFormat": "p95",
           "refId": "A"
         },
         {
-          "expr": "histogram_quantile(0.50, rate(fastapi_request_duration_seconds_bucket[5m])) * 1000",
+          "expr": "histogram_quantile(0.50, sum(rate(hopcroft_request_duration_seconds_bucket[5m])) by (le)) * 1000",
           "legendFormat": "p50 (median)",
           "refId": "B"
         }
@@ -152,32 +152,25 @@
           "color": {
             "mode": "thresholds"
           },
-          "mappings": [
-            {
-              "options": {
-                "0": {
-                  "color": "red",
-                  "index": 1,
-                  "text": "No Drift"
-                },
-                "1": {
-                  "color": "green",
-                  "index": 0,
-                  "text": "Drift Detected"
-                }
-              },
-              "type": "value"
-            }
-          ],
+          "mappings": [],
           "thresholds": {
             "mode": "absolute",
             "steps": [
               {
                 "color": "green",
                 "value": null
+              },
+              {
+                "color": "yellow",
+                "value": 1
+              },
+              {
+                "color": "red",
+                "value": 5
               }
             ]
-          }
+          },
+          "unit": "short"
         }
       },
       "gridPos": {
@@ -201,13 +194,13 @@
       "pluginVersion": "9.0.0",
       "targets": [
         {
-          "expr": "drift_detected",
+          "expr": "sum(hopcroft_in_progress_requests)",
           "refId": "A"
         }
       ],
-      "title": "Data Drift Status",
+      "title": "In-Progress Requests",
       "type": "stat",
-      "description": "Current data drift detection status (1 = drift detected, 0 = no drift)"
+      "description": "Number of requests currently being processed"
     },
     {
       "datasource": "Prometheus",
@@ -216,7 +209,7 @@
           "color": {
             "mode": "thresholds"
           },
-          "decimals": 4,
+          "decimals": 2,
           "mappings": [],
           "thresholds": {
             "mode": "absolute",
@@ -235,7 +228,7 @@
               }
             ]
           },
-          "unit": "short"
+          "unit": "percentunit"
         }
       },
       "gridPos": {
@@ -259,13 +252,13 @@
       "pluginVersion": "9.0.0",
       "targets": [
         {
-          "expr": "drift_p_value",
+          "expr": "sum(rate(hopcroft_requests_total{http_status=~\"5..\"}[5m])) / sum(rate(hopcroft_requests_total[5m]))",
           "refId": "A"
         }
       ],
-      "title": "Drift P-Value",
+      "title": "Error Rate (5xx)",
       "type": "stat",
-      "description": "Statistical significance of detected drift (lower = more significant)"
+      "description": "Percentage of requests resulting in 5xx errors"
     },
     {
       "datasource": "Prometheus",
@@ -305,7 +298,7 @@
               }
             ]
           },
-          "unit": "short"
+          "unit": "s"
         }
       },
       "gridPos": {
@@ -328,14 +321,84 @@
       "pluginVersion": "9.0.0",
       "targets": [
         {
-          "expr": "drift_distance",
-          "legendFormat": "Distance",
+          "expr": "rate(hopcroft_prediction_processing_seconds_sum[5m]) / rate(hopcroft_prediction_processing_seconds_count[5m])",
+          "legendFormat": "Avg Prediction Time",
+          "refId": "A"
+        }
+      ],
+      "title": "Model Prediction Time",
+      "type": "timeseries",
+      "description": "Average time spent processing model predictions"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "drawStyle": "bars",
+            "fillOpacity": 80,
+            "gradientMode": "none",
+            "hideFrom": {
+              "tooltip": false,
+              "viz": false,
+              "legend": false
+            },
+            "lineInterpolation": "linear",
+            "lineWidth": 1,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": false
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 24,
+        "x": 0,
+        "y": 14
+      },
+      "id": 6,
+      "options": {
+        "legend": {
+          "calcs": ["sum"],
+          "displayMode": "table",
+          "placement": "right"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum by (endpoint) (increase(hopcroft_requests_total[5m]))",
+          "legendFormat": "{{endpoint}}",
           "refId": "A"
         }
       ],
-      "title": "Drift Distance Over Time",
+      "title": "Requests by Endpoint",
       "type": "timeseries",
-      "description": "Statistical distance between baseline and current data distribution"
+      "description": "Number of requests per endpoint over time"
     }
   ],
   "refresh": "10s",
@@ -353,6 +416,6 @@
   "timezone": "",
   "title": "Hopcroft ML Model Monitoring",
   "uid": "hopcroft-ml-dashboard",
-  "version": 1,
+  "version": 2,
   "weekStart": ""
 }

monitoring/grafana/provisioning/datasources/prometheus.yml

@@ -6,7 +6,7 @@ datasources:
     access: proxy
     uid: prometheus
     orgId: 1
-    url: http://127.0.0.1:9090
+    url: http://prometheus:9090
     isDefault: true
     editable: true
     jsonData: