iankelling.org Git - distro-setup/blob - filesystem/etc/prometheus/rules/iank.yml

   1 # other rules to consider:
   2 # filesystem, network, ntp rules:
   3 # https://github.com/cloudalchemy/ansible-prometheus defaults/main.yml
   4 # on my system, the interpolated values are in /a/opt/ansible-prometheus/rules.yml
   5 #
   6
   7
   8 groups:
   9 - name: standard
  10   rules:
  11   - alert: mailtest-check
  12     expr: |-
  13       time() - mailtest_check_last_usec > 60 * 12
  14     labels:
  15       severity: day
  16     annotations:
  17       description: '{{ $labels.instance }} mailtest-check'
  18       summary: '{{ $labels.instance }} mailtest-check'
  19
  20   # 42 mins: enough for a 30 min queue run plus 12
  21   - alert: mailtest-check
  22     expr: |-
  23       time() - mailtest_check_last_usec > 60 * 42
  24     labels:
  25       severity: prod
  26     annotations:
  27       description: '{{ $labels.instance }} mailtest-check'
  28       summary: '{{ $labels.instance }} mailtest-check'
  29
  30   - alert: 1pmtest
  31     expr: hour() == 18 and minute() < 5
  32     for: 0m
  33     labels:
  34       severity: daytest
  35     annotations:
  36       summary: Prometheus daily test alert (instance {{ $labels.instance }})
  37       description: "Prometheus daily test alert if no other alerts. It
  38     is an end to end test.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
  39
  40
  41 # https://awesome-prometheus-alerts.grep.to/rules
  42
  43
  44 # todo, we should probably group the prometheus alerts that indicate a
  45 # host-local problem.
  46 # eg, set a label  alert-group: local-prom, then make a receiver that
  47 # groups by it when the alert-group is local-prom.
  48
  49 - name: awesome prometheus alerts
  50   rules:
  51
  52   - alert: PrometheusJobMissing
  53     expr: absent(up{job="prometheus"})
  54     for: 30m
  55     labels:
  56       severity: day
  57     annotations:
  58       summary: Prometheus job missing (instance {{ $labels.instance }})
  59       description: "A Prometheus job has disappeared\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
  60
  61   - alert: PrometheusTargetMissing
  62     expr: up == 0
  63     for: 30m
  64     labels:
  65       severity: warn
  66     annotations:
  67       summary: Prometheus target missing (instance {{ $labels.instance }})
  68       description: "A Prometheus target has disappeared. An exporter might be crashed.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
  69
  70       # todo: this should supress the above alert
  71   # - alert: PrometheusAllTargetsMissing
  72   #   expr: count by (job) (up) == 0
  73   #   for: 30m
  74   #   labels:
  75   #     severity: day
  76   #     alert-group: local-prom
  77   #   annotations:
  78   #     summary: Prometheus all targets missing (instance {{ $labels.instance }})
  79   #     description: "A Prometheus job does not have living target anymore.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
  80
  81   - alert: PrometheusConfigurationReloadFailure
  82     expr: prometheus_config_last_reload_successful != 1
  83     for: 30m
  84     labels:
  85       severity: day
  86     annotations:
  87       summary: Prometheus configuration reload failure (instance {{ $labels.instance }})
  88       description: "Prometheus configuration reload error\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
  89
  90       # I have an out of band alert to make sure prometheus is up. this
  91       # looks like it would generate false positives. todo: think
  92       # through what a valid crash loop detection would look like.
  93   # - alert: PrometheusTooManyRestarts
  94   #   expr: changes(process_start_time_seconds{job=~"prometheus|pushgateway|alertmanager"}[15m]) > 10
  95   #   for: 0m
  96   #   labels:
  97   #     severity: warning
  98   #   annotations:
  99   #     summary: Prometheus too many restarts (instance {{ $labels.instance }})
 100   #     description: "Prometheus has restarted more than twice in the last 15 minutes. It might be crashlooping.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 101
 102   - alert: PrometheusAlertmanagerJobMissing
 103     expr: absent(up{job="alertmanager"})
 104     for: 30m
 105     labels:
 106       severity: warn
 107     annotations:
 108       summary: Prometheus AlertManager job missing (instance {{ $labels.instance }})
 109       description: "A Prometheus AlertManager job has disappeared\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 110
 111   - alert: PrometheusAlertmanagerConfigurationReloadFailure
 112     expr: alertmanager_config_last_reload_successful != 1
 113     for: 30m
 114     labels:
 115       severity: day
 116     annotations:
 117       summary: Prometheus AlertManager configuration reload failure (instance {{ $labels.instance }})
 118       description: "AlertManager configuration reload error\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 119
 120   - alert: PrometheusNotConnectedToAlertmanager
 121     expr: prometheus_notifications_alertmanagers_discovered < 1
 122     for: 30m
 123     labels:
 124       severity: day
 125     annotations:
 126       summary: Prometheus not connected to alertmanager (instance {{ $labels.instance }})
 127       description: "Prometheus cannot connect the alertmanager\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 128
 129   - alert: PrometheusRuleEvaluationFailures
 130     expr: increase(prometheus_rule_evaluation_failures_total[3m]) > 0
 131     for: 30m
 132     labels:
 133       severity: warn
 134     annotations:
 135       summary: Prometheus rule evaluation failures (instance {{ $labels.instance }})
 136       description: "Prometheus encountered {{ $value }} rule evaluation failures, leading to potentially ignored alerts.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 137
 138   - alert: PrometheusTemplateTextExpansionFailures
 139     expr: increase(prometheus_template_text_expansion_failures_total[3m]) > 0
 140     for: 30m
 141     labels:
 142       severity: warn
 143     annotations:
 144       summary: Prometheus template text expansion failures (instance {{ $labels.instance }})
 145       description: "Prometheus encountered {{ $value }} template text expansion failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 146
 147   - alert: PrometheusRuleEvaluationSlow
 148     expr: prometheus_rule_group_last_duration_seconds > prometheus_rule_group_interval_seconds
 149     for: 5m
 150     labels:
 151       severity: warn
 152     annotations:
 153       summary: Prometheus rule evaluation slow (instance {{ $labels.instance }})
 154       description: "Prometheus rule evaluation took more time than the scheduled interval. It indicates a slower storage backend access or too complex query.\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 155
 156   - alert: PrometheusNotificationsBacklog
 157     expr: min_over_time(prometheus_notifications_queue_length[30m]) > 0
 158     for: 0m
 159     labels:
 160       severity: warn
 161     annotations:
 162       summary: Prometheus notifications backlog (instance {{ $labels.instance }})
 163       description: "The Prometheus notification queue has not been empty for 10 minutes\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 164
 165   - alert: PrometheusAlertmanagerNotificationFailing
 166     expr: rate(alertmanager_notifications_failed_total[1m]) > 0
 167     for: 30m
 168     labels:
 169       severity: warn
 170     annotations:
 171       summary: Prometheus AlertManager notification failing (instance {{ $labels.instance }})
 172       description: "Alertmanager is failing sending notifications\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 173
 174   # file_sd doesnt count as service discovery, so 0 is expected.
 175   # - alert: PrometheusTargetEmpty
 176   #   expr: prometheus_sd_discovered_targets == 0
 177   #   for: 30m
 178   #   labels:
 179   #     severity: day
 180   #   annotations:
 181   #     summary: Prometheus target empty (instance {{ $labels.instance }})
 182   #     description: "Prometheus has no target in service discovery\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 183
 184   - alert: PrometheusTargetScrapingSlow
 185     expr: prometheus_target_interval_length_seconds{quantile="0.9"} > 90
 186     for: 30m
 187     labels:
 188       severity: warn
 189     annotations:
 190       summary: Prometheus target scraping slow (instance {{ $labels.instance }})
 191       description: "Prometheus is scraping exporters slowly\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 192
 193   - alert: PrometheusLargeScrape
 194     expr: increase(prometheus_target_scrapes_exceeded_sample_limit_total[10m]) > 10
 195     for: 30m
 196     labels:
 197       severity: warn
 198     annotations:
 199       summary: Prometheus large scrape (instance {{ $labels.instance }})
 200       description: "Prometheus has many scrapes that exceed the sample limit\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 201
 202   - alert: PrometheusTargetScrapeDuplicate
 203     expr: increase(prometheus_target_scrapes_sample_duplicate_timestamp_total[5m]) > 0
 204     for: 30m
 205     labels:
 206       severity: warn
 207     annotations:
 208       summary: Prometheus target scrape duplicate (instance {{ $labels.instance }})
 209       description: "Prometheus has many samples rejected due to duplicate timestamps but different values\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 210
 211   - alert: PrometheusTsdbCheckpointCreationFailures
 212     expr: increase(prometheus_tsdb_checkpoint_creations_failed_total[1m]) > 0
 213     for: 30m
 214     labels:
 215       severity: warn
 216     annotations:
 217       summary: Prometheus TSDB checkpoint creation failures (instance {{ $labels.instance }})
 218       description: "Prometheus encountered {{ $value }} checkpoint creation failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 219
 220   - alert: PrometheusTsdbCheckpointDeletionFailures
 221     expr: increase(prometheus_tsdb_checkpoint_deletions_failed_total[1m]) > 0
 222     for: 30m
 223     labels:
 224       severity: warn
 225     annotations:
 226       summary: Prometheus TSDB checkpoint deletion failures (instance {{ $labels.instance }})
 227       description: "Prometheus encountered {{ $value }} checkpoint deletion failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 228
 229   - alert: PrometheusTsdbCompactionsFailed
 230     expr: increase(prometheus_tsdb_compactions_failed_total[1m]) > 0
 231     for: 30m
 232     labels:
 233       severity: warn
 234     annotations:
 235       summary: Prometheus TSDB compactions failed (instance {{ $labels.instance }})
 236       description: "Prometheus encountered {{ $value }} TSDB compactions failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 237
 238   - alert: PrometheusTsdbHeadTruncationsFailed
 239     expr: increase(prometheus_tsdb_head_truncations_failed_total[1m]) > 0
 240     for: 30m
 241     labels:
 242       severity: warn
 243     annotations:
 244       summary: Prometheus TSDB head truncations failed (instance {{ $labels.instance }})
 245       description: "Prometheus encountered {{ $value }} TSDB head truncation failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 246
 247   - alert: PrometheusTsdbReloadFailures
 248     expr: increase(prometheus_tsdb_reloads_failures_total[1m]) > 0
 249     for: 30m
 250     labels:
 251       severity: warn
 252     annotations:
 253       summary: Prometheus TSDB reload failures (instance {{ $labels.instance }})
 254       description: "Prometheus encountered {{ $value }} TSDB reload failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 255
 256   - alert: PrometheusTsdbWalCorruptions
 257     expr: increase(prometheus_tsdb_wal_corruptions_total[1m]) > 0
 258     for: 30m
 259     labels:
 260       severity: warn
 261     annotations:
 262       summary: Prometheus TSDB WAL corruptions (instance {{ $labels.instance }})
 263       description: "Prometheus encountered {{ $value }} TSDB WAL corruptions\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"
 264
 265   - alert: PrometheusTsdbWalTruncationsFailed
 266     expr: increase(prometheus_tsdb_wal_truncations_failed_total[1m]) > 0
 267     for: 30m
 268     labels:
 269       severity: warn
 270     annotations:
 271       summary: Prometheus TSDB WAL truncations failed (instance {{ $labels.instance }})
 272       description: "Prometheus encountered {{ $value }} TSDB WAL truncation failures\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"