X-Git-Url: https://git.openstreetmap.org./chef.git/blobdiff_plain/2e92afd15fdc5dd832bd975aa0cdd9caa0bff8a1..f8a576aca1270f703f38c3257cf533390692de06:/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 d8671699c..a0cea5792 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -25,14 +25,14 @@ 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" @@ -105,13 +105,20 @@ groups: pressure: "{{ $value | humanizePercentage }}" - name: database rules: - - alert: postgres replication delay - expr: pg_replication_lag_seconds > 30 - for: 15m + - alert: active rails queries + 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: - delay: "{{ $value | humanizeDuration }}" + queries: "{{ $value }}" + - alert: active cgimap queries + 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 }}" - name: discourse rules: - alert: discourse job failure rate @@ -414,6 +421,13 @@ 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 @@ -579,13 +593,6 @@ groups: alertgroup: "{{ $labels.instance }}" annotations: new_deadlocks: "{{ $value }}" - - alert: postgresql slow queries - expr: pg_slow_queries > 0 - for: 5m - labels: - alertgroup: "{{ $labels.instance }}" - annotations: - queries: "{{ $value }}" - alert: postgresql idle transactions expr: sum(pg_process_idle_seconds_count{state="idle in transaction"}) by (instance, server) > sum(pg_process_idle_seconds_bucket{state="idle in transaction",le="300"}) by (instance, server) for: 5m @@ -605,6 +612,11 @@ 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