[Prometheus] Add new alerts

Signed-off-by: Yohann D'ANELLO <ynerant@crans.org>

host_vars/fyre.adm.crans.org.yml

@@ -219,7 +219,7 @@ loc_prometheus:
         metrics_path: '/snmp'
         params:
           module:
-            - eatonups
+            - apc
         relabel_configs:
           - source_labels:
               - __address__

roles/prometheus/templates/prometheus/alert.rules.yml.j2

@@ -1,9 +1,9 @@
 {{ ansible_header | comment }}
 {# As this is also using brackets it will conflict without a raw block #}
 {% raw %}
-# Synced with https://awesome-prometheus-alerts.grep.to/rules.html on 2021-06-07
+# Synced with https://awesome-prometheus-alerts.grep.to/rules.html on 2022-08-09
 # We remove descriptions as we only send summary on IRC.
-# UPS, APT and RADIUS configuration is made by Crans.
+# UPS, APT and printer configuration are made by Crans.
 
 groups:
 - name: alert.rules
@@ -151,7 +151,7 @@ groups:
       summary: Le RAID sur {{ $labels.instance }} a perdu {{ $value }} disque(s)
 
   - alert: HostOomKillDetected
-    expr: increase(node_vmstat_oom_kill[1m]) > 0
+    expr: increase(node_vmstat_oom_kill[5m]) > 0
     for: 0m
     labels:
       severity: warning
@@ -174,6 +174,14 @@ groups:
     annotations:
       summary: Host EDAC Uncorrectable Errors detected (instance {{ $labels.instance }})
 
+  - alert: HostNetworkBondDegraded
+    expr: (node_bonding_active - node_bonding_slaves) != 0
+    for: 2m
+    labels:
+      severity: warning
+    annotations:
+      summary: Host Network Bond Degraded (instance {{ $labels.instance }})
+
   # This happend in June 2021 at Crans
   - alert: HostConntrackLimit
     expr: node_nf_conntrack_entries / node_nf_conntrack_entries_limit > 0.8
@@ -183,6 +191,30 @@ groups:
     annotations:
       summary: Host conntrack limit (instance {{ $labels.instance }})
 
+  - alert: HostClockSkew
+    expr: (node_timex_offset_seconds > 0.05 and deriv(node_timex_offset_seconds[5m]) >= 0) or (node_timex_offset_seconds < -0.05 and deriv(node_timex_offset_seconds[5m]) <= 0)
+    for: 2m
+    labels:
+      severity: warning
+    annotations:
+      summary: Host clock skew (instance {{ $labels.instance }})
+
+  - alert: HostClockNotSynchronising
+    expr: min_over_time(node_timex_sync_status[1m]) == 0 and node_timex_maxerror_seconds >= 16
+    for: 2m
+    labels:
+      severity: warning
+    annotations:
+      summary: Host clock not synchronising (instance {{ $labels.instance }})
+
+  - alert: HostRequiresReboot
+    expr: node_reboot_required > 0
+    for: 4h
+    labels:
+      severity: info
+    annotations:
+      summary: Host requires reboot (instance {{ $labels.instance }})
+
   ############
   # Blackbox #
   ############
@@ -203,82 +235,92 @@ groups:
     annotations:
       summary: Blackbox SSL certificate will expire soon (instance {{ $labels.instance }})
 
+  ##############
+  # PostgreSQL #
+  ##############
+
+  - alert: PostgresqlDown
+    expr: pg_up == 0
+    for: 0m
+    labels:
+      severity: critical
+    annotations:
+      summary: Postgresql down (instance {{ $labels.instance }})
+
+  - alert: PostgresqlTableNotAutoVacuumed
+    expr: (pg_stat_user_tables_last_autovacuum > 0) and (time() - pg_stat_user_tables_last_autovacuum) > 60 * 60 * 24 * 10
+    for: 0m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql table not auto vacuumed (instance {{ $labels.instance }})
+
+  - alert: PostgresqlTableNotAutoAnalyzed
+    expr: (pg_stat_user_tables_last_autoanalyze > 0) and (time() - pg_stat_user_tables_last_autoanalyze) > 24 * 60 * 60 * 10
+    for: 0m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql table not auto analyzed (instance {{ $labels.instance }})
+
+  - alert: PostgresqlTooManyConnections
+    expr: sum by (datname) (pg_stat_activity_count{datname!~"template.*|postgres"}) > pg_settings_max_connections * 0.8
+    for: 2m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql too many connections (instance {{ $labels.instance }})
+
+  - alert: PostgresqlNotEnoughConnections
+    expr: sum by (datname) (pg_stat_activity_count{datname!~"template.*|postgres"}) < 5
+    for: 2m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql not enough connections (instance {{ $labels.instance }})
+
+  - alert: PostgresqlDeadLocks
+    expr: increase(pg_stat_database_deadlocks{datname!~"template.*|postgres"}[1m]) > 5
+    for: 0m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql dead locks (instance {{ $labels.instance }})
+
+  - alert: PostgresqlHighRollbackRate
+    expr: rate(pg_stat_database_xact_rollback{datname!~"template.*"}[3m]) / rate(pg_stat_database_xact_commit{datname!~"template.*"}[3m]) > 0.02
+    for: 0m
+    labels:
+      severity: warning
+    annotations:
+      summary: Postgresql high rollback rate (instance {{ $labels.instance }})
+
+
+  ########
+  # Bird #
+  ########
+
+  # Check BGP routes
+  - alert: BGPRoutesMissing
+    expr: bird_protocol_prefix_import_count{proto="BGP", import_filter="ACCEPT"} < 5
+    for: 5m
+    labels:
+      severity: warning
+    annotations:
+      summary: Pas de route BGP importée depuis {{ $labels.name }}
+
   #######
   # UPS #
   #######
 
   # Check UPS
-  - alert: UpsOutputSourceChanged
-    expr: upsOutputSource != 3
-    for: 5m
+  - alert: UpsTooHighPower
+    expr: sum(rPDUIdentDevicePowerWatts) > 2000
+    for: 3m
     labels:
       severity: warning
     annotations:
-      summary: La source d'alimentation de {{ $labels.instance }} a changé !
-
-  - alert: UpsBatteryStatusChanged
-    expr: upsBatteryStatus != 2
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: L'état de la batterie de {{ $labels.instance }} a changé !
-
-  - alert: UpsTemperatureWarning
-    expr: (xupsEnvRemoteTemp < 10) or (xupsEnvRemoteTemp > 26)
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: La température autour de {{ $labels.instance }} est de {{ $value }}°C
-
-  - alert: UpsTemperatureCritical
-    expr: (xupsEnvRemoteTemp < 0) or (xupsEnvRemoteTemp > 30)
-    for: 5m
-    labels:
-      severity: critical
-    annotations:
-      summary: La température autour de {{ $labels.instance }} est de {{ $value }}°C
-
-  - alert: UpsHighHumidity
-    expr: xupsEnvRemoteHumidity > 65
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: L'humidité autour de {{ $labels.instance }} est de {{ $value }}%
-
-  - alert: UpsVeryHighHumidity
-    expr: xupsEnvRemoteHumidity > 85
-    for: 5m
-    labels:
-      severity: critical
-    annotations:
-      summary: L'humidité autour de {{ $labels.instance }} est de {{ $value }}%
-
-  - alert: UpsHighLoad
-    expr: upsOutputPercentLoad > 70
-    for: 5m
-    labels:
-      severity: critical
-    annotations:
-      summary: La charge de {{ $labels.instance }} est de {{ $value }}%
-
-  - alert: UpsWrongInputVoltage
-    expr: (upsInputVoltage < 210) or (upsInputVoltage > 250)
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: La tension d'entrée de {{ $labels.instance }} est de {{ $value }}V
-
-  - alert: UpsWrongOutputVoltage
-    expr: (upsOutputVoltage < 215) or (upsOutputVoltage > 245)
-    for: 5m
-    labels:
-      severity: warning
-    annotations:
-      summary: La tension de sortie de {{ $labels.instance }} est de {{ $value }}V
+      summary: La puissance totale tirée est trop grande ({{ $labels.rPDUIdentDevicePowerWatts }} W)
 
   #######
   # iLO #