From: Tom Hughes Date: Wed, 17 May 2023 19:41:52 +0000 (+0100) Subject: Reduce sensitivity of job processing rate alert X-Git-Url: https://git.openstreetmap.org./chef.git/commitdiff_plain/97751762159405edfcd074565c79c67c4f09b726 Reduce sensitivity of job processing rate alert --- diff --git a/cookbooks/prometheus/templates/default/alert_rules.yml.erb b/cookbooks/prometheus/templates/default/alert_rules.yml.erb index 4e00763e6..02c41ce52 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -727,8 +727,8 @@ groups: annotations: 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: 15m + expr: rate(pg_stat_user_tables_n_tup_del{datname="openstreetmap",relname="delayed_jobs"}[1h]) / rate(pg_stat_user_tables_n_tup_ins{datname="openstreetmap",relname="delayed_jobs"}[1h]) < 0.9 and ignoring(job, name, datname, relname, schemaname, server) chef_role{name="db-master"} == 1 + for: 1h labels: alertgroup: web annotations: