X-Git-Url: https://git.openstreetmap.org./chef.git/blobdiff_plain/cf4bd571513a1f63e898d1f17c29d472bffe41a6..6ca4c5b1adbeff9556a2f63e7716d21d02495901:/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 20136a771..305afbd90 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -3,13 +3,20 @@ groups: - name: amsterdam rules: - - alert: uplink + - alert: he uplink expr: junos_interface_up{site="amsterdam",name=~"ge-[01]/2/2"} != 1 for: 6m labels: alertgroup: "amsterdam" annotations: status: "{{ $value }}" + - alert: equinix uplink + expr: junos_interface_up{site="amsterdam",name=~"xe-[01]/2/0"} != 1 + for: 6m + labels: + alertgroup: "amsterdam" + annotations: + status: "{{ $value }}" - alert: pdu current draw expr: rPDU2PhaseStatusCurrent{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 10 > 28 for: 6m @@ -25,14 +32,14 @@ groups: annotations: current: "{{ $value | humanize }}kVA" - alert: site temperature - expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 18 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 26 + expr: min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 < 15 or min(rPDU2SensorTempHumidityStatusTempC{site="amsterdam"}) / 10 > 32 for: 6m labels: alertgroup: "amsterdam" annotations: temperature: "{{ $value | humanize }}C" - alert: site humidity - expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.25 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.65 + expr: max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 < 0.08 or max(rPDU2SensorTempHumidityStatusRelativeHumidity{site="amsterdam"}) / 100 > 0.8 for: 6m labels: alertgroup: "amsterdam" @@ -106,19 +113,19 @@ groups: - name: database rules: - alert: active rails queries - expr: sum(pg_stat_activity_count{instance="snap-01",datname="openstreetmap",usename="rails",state="active"}) > 50 and on (instance) chef_role{name="db-master"} + expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="rails",state="active"}) by (instance) > 50 and on (instance) chef_role{name="db-master"} for: 5m labels: alertgroup: database annotations: queries: "{{ $value }}" - alert: active cgimap queries - expr: sum(pg_stat_activity_count{instance="snap-01",datname="openstreetmap",usename="cgimap",state="active"}) > 30 and on (instance) chef_role{name="db-master"} + expr: sum(pg_stat_activity_count{datname="openstreetmap",usename="cgimap",state="active"}) by (instance) > 30 and on (instance) chef_role{name="db-master"} for: 5m labels: alertgroup: database annotations: - delay: "{{ $value }}" + queries: "{{ $value }}" - name: discourse rules: - alert: discourse job failure rate @@ -130,13 +137,20 @@ groups: failure_rate: "{{ $value }} jobs/s" - name: dublin rules: - - alert: uplink + - alert: he uplink expr: junos_interface_up{site="dublin",name=~"ge-[01]/2/2"} != 1 for: 6m labels: alertgroup: "dublin" annotations: status: "{{ $value }}" + - alert: equinix uplink + expr: junos_interface_up{site="dublin",name=~"xe-[01]/2/0"} != 1 + for: 6m + labels: + alertgroup: "dublin" + annotations: + status: "{{ $value }}" - alert: pdu current draw expr: rPDU2PhaseStatusCurrent{site="dublin",rPDU2PhaseStatusIndex="1"} / 10 > 28 for: 6m @@ -331,6 +345,15 @@ groups: alertgroup: "{{ $labels.site }}" annotations: power: "{{ $value }} dBm" + - name: load + rules: + - alert: load average + expr: sum(node_load5) by (instance) / count(node_cpu_frequency_max_hertz) by (instance) > 2 + for: 5m + labels: + alertgroup: "{{ $labels.instance }}" + annotations: + load: "{{ $value | humanizePercentage }}" - name: mail rules: - alert: exim down @@ -421,6 +444,13 @@ groups: alertgroup: "{{ $labels.instance }}" annotations: connections_used: "{{ $value | humanizePercentage }}" + - alert: mysql connection errors + expr: increase(mysql_global_status_connection_errors_total[1m]) > 0 + for: 0m + labels: + alertgroup: "{{ $labels.instance }}" + annotations: + error_count: "{{ $value }}" - name: network rules: - alert: interface redundancy lost @@ -481,6 +511,11 @@ groups: alertgroup: nominatim annotations: delay: "{{ $value | humanizeDuration }}" + - alert: nominatim connections + expr: sum(nginx_connections_writing and on (instance) chef_role{name="nominatim"}) > 2500 + for: 15m + labels: + alertgroup: nominatim - name: overpass rules: - alert: overpass osm database age @@ -605,6 +640,11 @@ groups: for: 10m labels: alertgroup: "prometheus" + - alert: node exporter text file scrape error + expr: node_textfile_scrape_error > 0 + for: 10m + labels: + alertgroup: "prometheus" - name: raid rules: - alert: raid controller battery failed @@ -643,6 +683,13 @@ groups: alertgroup: "{{ $labels.instance }}" annotations: new_errors: "{{ $value }}" + - name: resolved + rules: + - alert: dnssec validation failures + expr: rate(resolved_dnssec_verdicts_total{result="bogus"}[1m]) > 1 + for: 5m + labels: + alertgroup: "{{ $labels.instance }}" - name: smart rules: - alert: smart failure @@ -781,7 +828,7 @@ groups: - name: web rules: - alert: web error rate - expr: sum(rate(api_call_count_total{status=~"50[0-8]|5[1-9][0-9]"}[5m])) by (instance) / sum(rate(api_call_count_total[5m])) by (instance) > 0.002 + expr: sum(rate(api_call_count_total{status=~"50[0-8]|5[1-9][0-9]"}[5m])) by (instance) / sum(rate(api_call_count_total[5m])) by (instance) > 0.002 and sum(rate(api_call_count_total{status=~"50[0-8]|5[1-9][0-9]"}[5m])) by (instance) > 0.01 for: 5m labels: alertgroup: web