X-Git-Url: https://git.openstreetmap.org./chef.git/blobdiff_plain/6e0c9995ffe09af3f2fe836d95a612709179ebca..fa2ffc4b6ec0b9572dac9470bcfb474894b8a722:/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 7d7fa4abf..92c38d5c5 100644 --- a/cookbooks/prometheus/templates/default/alert_rules.yml.erb +++ b/cookbooks/prometheus/templates/default/alert_rules.yml.erb @@ -25,12 +25,12 @@ groups: - name: cpu rules: - alert: cpu pressure - expr: rate(node_pressure_memory_waiting_seconds_total[5m]) > 0.3 + expr: rate(node_pressure_cpu_waiting_seconds_total[5m]) > 0.3 for: 15m labels: alertgroup: "{{ $labels.instance }}" annotations: - major_page_faults: "{{ $value | humanizePercentage }}" + pressure: "{{ $value | humanizePercentage }}" - name: database rules: - alert: postgres replication delay @@ -40,6 +40,15 @@ groups: alertgroup: database annotations: delay: "{{ $value | humanizeDuration }}" + - name: fastly + rules: + - alert: error rate + expr: sum(rate(fastly_rt_status_group_total{status_group="5xx"}[5m])) by (service_name, datacenter) / sum(rate(fastly_rt_status_group_total[5m])) by (service_name, datacenter) > 0.005 + for: 15m + labels: + alertgroup: fastly + annotations: + error_rate: "{{ $value | humanizePercentage }}" - name: filesystem rules: - alert: readonly filesystem @@ -95,13 +104,13 @@ groups: in_max_volts: "{{ with printf \"node_hwmon_in_max_volts{instance='%s',chip='%s',sensor='%s'}\" $labels.instance $labels.chip $labels.sensor | query }}{{ . | first | value | humanize }}V{{end}}" - name: io rules: - - alert: cpu pressure + - alert: io pressure expr: rate(node_pressure_io_waiting_seconds_total[5m]) > 0.6 for: 60m labels: alertgroup: "{{ $labels.instance }}" annotations: - major_page_faults: "{{ $value | humanizePercentage }}" + pressure: "{{ $value | humanizePercentage }}" - name: ipmi rules: - alert: ipmi fan alarm @@ -130,6 +139,22 @@ groups: for: 5m labels: alertgroup: "{{ $labels.instance }}" + - name: mail + rules: + - alert: exim queue length + expr: exim_queue > exim_queue_limit + for: 60m + labels: + alertgroup: mail + annotations: + queue_length: "{{ $value }}" + - alert: mailman queue length + expr: mailman_queue_length > 200 + for: 60m + labels: + alertgroup: mail + annotations: + queue_length: "{{ $value }}" - name: mdadm rules: - alert: mdadm array inactive @@ -177,7 +202,7 @@ groups: labels: alertgroup: "{{ $labels.instance }}" annotations: - major_page_faults: "{{ $value | humanizePercentage }}" + pressure: "{{ $value | humanizePercentage }}" - alert: oom kill detected expr: increase(node_vmstat_oom_kill[1m]) > 0 for: 0m