From: Tom Hughes Date: Mon, 10 Apr 2023 15:04:05 +0000 (+0100) Subject: Base site power alerts on a one hour rolling average X-Git-Url: https://git.openstreetmap.org./chef.git/commitdiff_plain/b2dbf6bffb0f59ef340942fcdec0a9e36b33730a Base site power alerts on a one hour rolling average --- diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb index 440d2ac8b..8cf177669 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -11,7 +11,7 @@ groups: annotations: current: "{{ $value | humanize }}A" - alert: site power - expr: sum(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"} / 100) > 3 + expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="amsterdam",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 3 for: 6m labels: alertgroup: "amsterdam" @@ -124,7 +124,7 @@ groups: annotations: current: "{{ $value | humanize }}A" - alert: site power - expr: sum(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"} / 100) > 4 + expr: sum(avg_over_time(rPDU2PhaseStatusApparentPower{site="dublin",rPDU2PhaseStatusIndex="1"}[1h]) / 100) > 4 for: 6m labels: alertgroup: "dublin"