X-Git-Url: https://git.openstreetmap.org./chef.git/blobdiff_plain/ab63149db6f2c9739f9dc39e0f0b18891f865690..28326173066466bb20f54bad702b2885b64f925c:/cookbooks/prometheus/templates/default/alert_rules.yml.erb?ds=inline

diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
index 289b33f57..8c1aeee48 100644
--- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb
+++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb
@@ -1,13 +1,6 @@
 # DO NOT EDIT - This file is being maintained by Chef
 
 groups:
-  - name: alertmanager
-    rules:
-      - alert: prometheus target missing
-        expr: up == 0
-        for: 5m
-        labels:
-          alertgroup: "prometheus"
   - name: amsterdam
     rules:
       - alert: pdu current draw
@@ -75,11 +68,15 @@ groups:
         for: 5m
         labels:
           alertgroup: "{{ $labels.site }}"
+        annotations:
+          fan_rpm: "{{ with printf \"rlPhdUnitEnvParamFan1Speed{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}rpm{{end}}"
       - alert: cisco temperature alarm
         expr: rlPhdUnitEnvParamTempSensorStatus{rlPhdUnitEnvParamTempSensorStatus!="ok"} > 0
         for: 5m
         labels:
           alertgroup: "{{ $labels.site }}"
+        annotations:
+          temp_celsius: "{{ with printf \"rlPhdUnitEnvParamTempSensorValue{site='%s',instance='%s',rlPhdUnitEnvParamStackUnit='%s'}\" $labels.site $labels.instance $labels.rlPhdUnitEnvParamStackUnit | query }}{{ . | first | value | humanize }}C{{end}}"
       - alert: cisco main power alarm
         expr: rlPhdUnitEnvParamMainPSStatus{rlPhdUnitEnvParamMainPSStatus!="normal"} > 0
         for: 5m
@@ -108,15 +105,55 @@ groups:
           alertgroup: database
         annotations:
           delay: "{{ $value | humanizeDuration }}"
+  - name: dublin
+    rules:
+      - alert: pdu current draw
+        expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 13
+        for: 5m
+        labels:
+          alertgroup: "dublin"
+        annotations:
+          current: "{{ $value | humanize }}A"
+      - alert: site current draw
+        expr: sum(rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10) > 17
+        for: 5m
+        labels:
+          alertgroup: "dublin"
+        annotations:
+          current: "{{ $value | humanize }}A"
+      - alert: site temperature
+        expr: min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="dublin"}) / 10 > 25
+        for: 5m
+        labels:
+          alertgroup: "dublin"
+        annotations:
+          temperature: "{{ $value | humanize }}C"
+      - alert: site humidity
+        expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="dublin"}) / 100 > 0.65
+        for: 5m
+        labels:
+          alertgroup: "dublin"
+        annotations:
+          humidity: "{{ $value | humanizePercentage }}"
   - name: fastly
     rules:
-      - alert: error rate
+      - alert: fastly error rate
         expr: sum(rate(fastly_rt_status_group_total{status_group="5xx"}[5m])) by (service_name, datacenter) / sum(rate(fastly_rt_status_group_total[5m])) by (service_name, datacenter) > 0.005
         for: 15m
         labels:
           alertgroup: fastly
         annotations:
           error_rate: "{{ $value | humanizePercentage }}"
+      - alert: fastly healthcheck failing
+        expr: count(fastly_healthcheck_status == 0) by (service) > 0
+        for: 15m
+        labels:
+          alertgroup: fastly
+      - alert: multiple fastly healthchecks failing
+        expr: count(fastly_healthcheck_status == 0) by (service) > 4
+        for: 5m
+        labels:
+          alertgroup: fastly
   - name: filesystem
     rules:
       - alert: readonly filesystem
@@ -207,6 +244,18 @@ groups:
         for: 5m
         labels:
           alertgroup: "{{ $labels.instance }}"
+  - name: juniper
+    rules:
+      - alert: juniper fan alarm
+        expr: sum_over_time(jnxOperatingState{jnxOperatingContentsIndex="4",jnxOperatingState!~"running.*"}[6m]) > 0
+        for: 5m
+        labels:
+          alertgroup: "{{ $labels.site }}"
+      - alert: juniper power alarm
+        expr: sum_over_time(jnxOperatingState{jnxOperatingContentsIndex="2",jnxOperatingState!~"running.*"}[6m]) > 0
+        for: 5m
+        labels:
+          alertgroup: "{{ $labels.site }}"
   - name: mail
     rules:
       - alert: exim queue length
@@ -295,12 +344,19 @@ groups:
         annotations:
           bandwidth_used: "{{ $value | humanizePercentage }}"
       - alert: interface transmit errors
-        expr: rate(node_network_transmit_errs_total[1m]) / rate(node_network_transmit_packets_total[1m]) > 0.01
+        expr: rate(node_network_transmit_errs_total{device!~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device!~"wg.*"}[1m]) > 0.01
         for: 5m
         labels:
           alertgroup: "{{ $labels.instance }}"
         annotations:
           error_rate: "{{ $value | humanizePercentage }}"
+      - alert: wireguard interface transmit errors
+        expr: rate(node_network_transmit_errs_total{device=~"wg.*"}[1m]) / rate(node_network_transmit_packets_total{device=~"wg.*"}[1m]) > 0.05
+        for: 1h
+        labels:
+          alertgroup: "{{ $labels.instance }}"
+        annotations:
+          error_rate: "{{ $value | humanizePercentage }}"
       - alert: interface receive errors
         expr: rate(node_network_receive_errs_total[1m]) / rate(node_network_receive_packets_total[1m]) > 0.01
         for: 5m
@@ -315,6 +371,31 @@ groups:
           alertgroup: "{{ $labels.instance }}"
         annotations:
           entries_used: "{{ $value | humanizePercentage }}"
+  - name: nominatim
+    rules:
+      - alert: nominatim replication delay
+        expr: nominatim_replication_delay > 10800
+        for: 1h
+        labels:
+          alertgroup: nominatim
+        annotations:
+          delay: "{{ $value | humanizeDuration }}"
+  - name: overpass
+    rules:
+      - alert: overpass osm database age
+        expr: overpass_database_age_seconds{database="osm"} > 3600
+        for: 1h
+        labels:
+          alertgroup: overpass
+        annotations:
+          age: "{{ $value | humanizeDuration }}"
+      - alert: overpass area database age
+        expr: overpass_database_age_seconds{database="area"} > 86400
+        for: 1h
+        labels:
+          alertgroup: overpass
+        annotations:
+          age: "{{ $value | humanizeDuration }}"
   - name: planet
     rules:
       - alert: planet dump overdue
@@ -381,7 +462,7 @@ groups:
         annotations:
           connections_used: "{{ $value | humanizePercentage }}"
       - alert: postgresql deadlocks
-        expr: increase(pg_stat_database_deadlocks[1m]) > 5
+        expr: increase(pg_stat_database_deadlocks{datname!="nominatim"}[1m]) > 5
         for: 0m
         labels:
           alertgroup: "{{ $labels.instance }}"
@@ -394,6 +475,46 @@ groups:
           alertgroup: "{{ $labels.instance }}"
         annotations:
           queries: "{{ $value }}"
+  - name: prometheus
+    rules:
+      - alert: prometheus configuration error
+        expr: prometheus_config_last_reload_successful == 0
+        for: 10m
+        labels:
+          alertgroup: "prometheus"
+      - alert: prometheus target missing
+        expr: up == 0
+        for: 10m
+        labels:
+          alertgroup: "prometheus"
+  - name: raid
+    rules:
+      - alert: raid array degraded
+        expr: ohai_array_info{status="degraded"} > 0
+        for: 5m
+        labels:
+          alertgroup: "{{ $labels.instance }}"
+      - alert: raid disk failed
+        expr: ohai_disk_info{status="failed"} > 0
+        for: 5m
+        labels:
+          alertgroup: "{{ $labels.instance }}"
+  - name: rasdaemon
+    rules:
+      - alert: memory controller errors
+        expr: increase(rasdaemon_mc_events_total[1m]) > 0
+        for: 0m
+        labels:
+          alertgroup: "{{ $labels.instance }}"
+        annotations:
+          new_errors: "{{ $value }}"
+      - alert: pcie aer errors
+        expr: increase(rasdaemon_aer_events_total[1m]) > 0
+        for: 0m
+        labels:
+          alertgroup: "{{ $labels.instance }}"
+        annotations:
+          new_ercrors: "{{ $value }}"
   - name: smart
     rules:
       - alert: smart failure
@@ -402,7 +523,7 @@ groups:
         labels:
           alertgroup: "{{ $labels.instance }}"
       - alert: smart ssd wearout approaching
-        expr: smart_percentage_used >= 90
+        expr: smart_percentage_used >= 80
         for: 60m
         labels:
           alertgroup: "{{ $labels.instance }}"
@@ -436,7 +557,7 @@ groups:
     rules:
       - alert: statuscake uptime check failing
         expr: statuscake_uptime{status="down",paused="false"} > 0
-        for: 0m
+        for: 10m
         labels:
           alertgroup: statuscake
   - name: systemd
@@ -446,7 +567,7 @@ groups:
         for: 5m
         labels:
           alertgroup: "{{ $labels.instance }}"
-      - alert: systemd failed service
+      - alert: systemd failed chef client service
         expr: node_systemd_unit_state{state="failed",name="chef-client.service"} == 1
         for: 6h
         labels: