Only alert if the job processing rate is low for an extended period

[chef.git] / 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 0aa0b64cbea271981d34dfaec55bdef89f5118d4..a22914f2a8c4cfa909b5187a4f1f7a26de990be0 100644 (file)
--- 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: