From: Tom Hughes Date: Sun, 29 May 2022 17:40:41 +0000 (+0100) Subject: Only alert if the job processing rate is low for an extended period X-Git-Url: https://git.openstreetmap.org./chef.git/commitdiff_plain/be877e437a5beb2e44a48b63c48cabe8e481608f Only alert if the job processing rate is low for an extended period --- diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb index 0aa0b64cb..a22914f2a 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -426,7 +426,7 @@ groups: error_rate: "{{ $value | humanizePercentage }}" - alert: job processing rate expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[5m]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[5m]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1 - for: 5m + for: 15m labels: alertgroup: web annotations: