Add alerts for site power usage in Amsterdam and Dublin

[chef.git] / cookbooks / prometheus / templates / default / alert_rules.yml.erb

diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
index b0094a83b24869ba8e5555751b45e5f1418135c3..8f2986e6ab85340d339be8e23f9adc219f0911ac 100644 (file)
--- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb
+++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
@@ -17,6 +17,13 @@ groups:
           alertgroup: "amsterdam"
         annotations:
           current: "{{ $value | humanize }}A"
           alertgroup: "amsterdam"
         annotations:
           current: "{{ $value | humanize }}A"

+      - alert: site power
+        expr: sum(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 100) > 3
+        for: 0m
+        labels:
+          alertgroup: "amsterdam"
+        annotations:
+          current: "{{ $value | humanize }}kVA"

       - alert: site temperature
         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 26
         for: 6m
       - alert: site temperature
         expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 26
         for: 6m


@@ -90,8 +97,8 @@ groups:
   - name: cpu
     rules:
       - alert: cpu pressure
   - name: cpu
     rules:
       - alert: cpu pressure

-        expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.6
-        for: 15m
+        expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.75
+        for: 60m

         labels:
           alertgroup: "{{ $labels.instance }}"
         annotations:
         labels:
           alertgroup: "{{ $labels.instance }}"
         annotations:


@@ -130,6 +137,13 @@ groups:
           alertgroup: "dublin"
         annotations:
           current: "{{ $value | humanize }}A"
           alertgroup: "dublin"
         annotations:
           current: "{{ $value | humanize }}A"

+      - alert: site power
+        expr: sum(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"} / 100) > 4
+        for: 0m
+        labels:
+          alertgroup: "dublin"
+        annotations:
+          current: "{{ $value | humanize }}kVA"

       - alert: site temperature
         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
         for: 6m
       - alert: site temperature
         expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 26
         for: 6m


@@ -578,7 +592,7 @@ groups:
         labels:
           alertgroup: "{{ $labels.instance }}"
       - alert: smart ssd wearout approaching
         labels:
           alertgroup: "{{ $labels.instance }}"
       - alert: smart ssd wearout approaching

-        expr: smart_percentage_used >= 80
+        expr: smart_percentage_used / 100 >= 0.8

         for: 60m
         labels:
           alertgroup: "{{ $labels.instance }}"
         for: 60m
         labels:
           alertgroup: "{{ $labels.instance }}"


@@ -587,8 +601,8 @@ groups:
   - name: smokeping
     rules:
       - alert: packet loss
   - name: smokeping
     rules:
       - alert: packet loss

-        expr: 100 - (rate(smokeping_response_duration_seconds_count[5m]) * 100 / rate(smokeping_requests_total[5m])) > 0
-        for: 5m
+        expr: 1 - (rate(smokeping_response_duration_seconds_count[5m]) / rate(smokeping_requests_total[5m])) > 0.02
+        for: 10m

         labels:
           alertgroup: smokeping
         annotations:
         labels:
           alertgroup: smokeping
         annotations: