Add notifications

Author: Pablo Mendez

notifications.yaml

@@ -0,0 +1,114 @@
+endpoints:
+  - name: "High CPU Usage Check"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=avg%20by%20(instance)%20(rate(node_cpu_seconds_total%7Bmode%21%3D%22idle%22%7D%5B2m%5D))"
+    method: "GET"
+    interval: "30s"  
+    conditions:
+      - "[BODY].data.result[0].value[1] > 0.80"  # Extract the second element of 'value' array
+    definition:
+      title: "High CPU Usage Alert"
+      description: "Triggers if CPU usage exceeds 80% for 1 minute."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "CPU usage above 80%"
+        failure-threshold: 2  
+        success-threshold: 1  
+        send-on-resolved: true
+
+  - name: "Low Available Memory Check"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=node_memory_MemAvailable_bytes%20%2F%20node_memory_MemTotal_bytes"
+    method: "GET"
+    interval: "30s"  # Check every 30 seconds
+    conditions:
+      - "[BODY].data.result[0].value[1] < 0.10"  # Less than 10% memory available
+    definition:
+      title: "Host Out of Memory Alert"
+      description: "Triggers if available memory drops below 10% for 2 minutes."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "Available memory below 10% for 2 minutes"
+        failure-threshold: 4  # (Every 30s * 4 = 2 minutes)
+        success-threshold: 1  
+        send-on-resolved: true
+
+  - name: "Low Disk Space Check"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=(node_filesystem_avail_bytes%7Bfstype!~%22^(fuse.*|tmpfs|cifs|nfs)%22%7D%20%2F%20node_filesystem_size_bytes%20%3C%200.10%20and%20on%20(instance%2C%20device%2C%20mountpoint)%20node_filesystem_readonly%20%3D%3D%200)"
+    method: "GET"
+    interval: "30s"  # Check every 30 seconds
+    conditions:
+      - "[BODY].data.result[0].value[1] < 0.10"  # Less than 10% disk space remaining
+    definition:
+      title: "Host Out of Disk Space Alert"
+      description: "Triggers if disk space falls below 10% for 2 minutes."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "Disk space below 10% for 2 minutes"
+        failure-threshold: 4  # (30s check interval * 4 = 2 minutes)
+        success-threshold: 1  
+        send-on-resolved: true
+
+  - name: "Disk Filling Up Prediction"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=predict_linear(node_filesystem_avail_bytes%7Bfstype!~%22^(fuse.*|tmpfs|cifs|nfs)%22%7D%5B1h%5D%2C%2086400)%20%3C%3D%200%20and%20node_filesystem_avail_bytes%20%3E%200"
+    method: "GET"
+    interval: "30s"  # Check every 30 seconds
+    conditions:
+      - "[BODY].data.result[0].value[1] <= 0"  # Predicts disk will be full within 24 hours
+    definition:
+      title: "Host Disk May Fill in 24 Hours Alert"
+      description: "Triggers if the system predicts disk space will be exhausted within 24 hours."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "Filesystem predicted to run out of space in 24 hours"
+        failure-threshold: 4  # (30s * 4 = 2 minutes)
+        success-threshold: 1  
+        send-on-resolved: true
+
+  - name: "Physical Component Overheating"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=node_hwmon_temp_celsius%20%3E%20node_hwmon_temp_max_celsius"
+    method: "GET"
+    interval: "30s"  # Check every 30 seconds
+    conditions:
+      - "[BODY].data.result[0].value[1] > 0"  # Ensures overheating condition is detected
+    definition:
+      title: "Host Physical Component Overheating Alert"
+      description: "Triggers if any hardware component exceeds its maximum temperature threshold."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "Hardware component temperature too high for 5 minutes"
+        failure-threshold: 10  # (30s * 10 = 5 minutes)
+        success-threshold: 1  
+        send-on-resolved: true
+
+  - name: "Node Overtemperature Alarm"
+    enabled: true
+    group: "host"
+    url: "http://prometheus.dms.dappnode:9090/api/v1/query?query=%28node_hwmon_temp_crit_alarm_celsius%20%3D%3D%201%29%20or%20%28node_hwmon_temp_alarm%20%3D%3D%201%29"
+    method: "GET"
+    interval: "30s"  # Check every 30 seconds
+    conditions:
+      - "[BODY].data.result[0].value[1] == 1"  # Overtemperature alarm triggered
+    definition:
+      title: "Host Node Overtemperature Alarm"
+      description: "Triggers immediately if a critical temperature alarm is raised."
+    alerts:
+      - type: custom
+        enabled: true
+        description: "Node temperature alarm triggered"
+        failure-threshold: 1  # Immediate alert
+        success-threshold: 1  
+        send-on-resolved: true