Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

add rule group related alerts #111

Merged
merged 5 commits into from
Sep 3, 2024
Merged

add rule group related alerts #111

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
24f4ea5
add rule group related alerts
asiyani Sep 2, 2024
75b2486
use increase for missed alerts
asiyani Sep 2, 2024
47bfccd
ignore spikes on re-start
asiyani Sep 2, 2024
0e62bd0
added alert rules for evaluation failures
asiyani Sep 2, 2024
2c67995
added extra alerts and converted all alerts to per instance
asiyani Sep 3, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 77 additions & 2 deletions common/metrics.yaml.tmpl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,8 @@
groups:
- name: metrics
rules:
# thanos-compact is a slow crasher, so it needs a more sensitive
# "ContainerRestartingOften" alert than the stock one
# thanos-compact is a slow crasher, so it needs a more sensitive
# "ContainerRestartingOften" alert than the stock one
- alert: ThanosCompactRestartingOften
expr: increase(kube_pod_container_status_restarts_total{container="thanos-compact"}[2h]) > 3
labels:
Expand All @@ -16,3 +16,78 @@ groups:
action: "Check pod status and container logs to figure out if there's a problem"
command: "`kubectl --context $ENVIRONMENT-$PROVIDER --namespace {{ $labels.namespace }} describe pod {{ $labels.pod }}`"
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.namespace}}\",app_kubernetes_io_name=\"{{$labels.label_app_kubernetes_io_name}}\"}"}]|link>
# https://github.com/thanos-io/thanos/blob/main/examples/alerts/alerts.md
- alert: ThanosRuleQueueIsDroppingAlerts
expr: sum by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name) (rate(thanos_alert_queue_alerts_dropped_total{app="thanos-rule"}[5m])) > 0
for: 5m
labels:
team: infra
annotations:
summary: "Thanos Rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} is failing to queue alerts"
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>
- alert: ThanosRuleSenderIsFailingAlerts
expr: sum by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name) (rate(thanos_alert_sender_alerts_dropped_total{app="thanos-rule"}[5m])) > 0
for: 5m
labels:
team: infra
annotations:
summary: "Thanos Rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} is failing to send alerts to alertmanager."
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>
- alert: ThanosNoRuleEvaluations
expr: |
sum by (kubernetes_cluster,kubernetes_namespace,kubernetes_pod_name) (rate(prometheus_rule_evaluations_total{app="thanos-rule"}[5m])) <= 0
and
sum by (kubernetes_cluster,kubernetes_namespace,kubernetes_pod_name) (thanos_rule_loaded_rules{app="thanos-rule"}) > 0
for: 5m
labels:
team: infra
annotations:
summary: "Thanos Rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} did not perform any rule evaluations in the past 10 minutes."
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>
- alert: ThanosRuleEvaluationLatencyHigh
expr: |
count by (kubernetes_cluster, kubernetes_namespace, kubernetes_pod_name) (
sum by(kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name, rule_group) (prometheus_rule_group_last_duration_seconds{app="thanos-rule"})
>
sum by(kubernetes_cluster, kubernetes_namespace, kubernetes_pod_name, rule_group) (prometheus_rule_group_interval_seconds{app="thanos-rule"})
) > 5
for: 5m
labels:
team: infra
annotations:
summary: "Thanos rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} has higher evaluation latency than interval for more then 5 group rules"
impact: "Slow evaluation can result in missed evaluations"
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>
- alert: ThanosRuleHighRuleEvaluationFailures
expr: |
count by (kubernetes_cluster, kubernetes_namespace, kubernetes_pod_name) (
sum by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name, rule_group) (rate(prometheus_rule_evaluation_failures_total{app="thanos-rule"}[5m]))
/
sum by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name, rule_group) (rate(prometheus_rule_evaluations_total{app="thanos-rule"}[5m]))
* 100 > 5
) > 5
for: 5m
labels:
team: infra
annotations:
summary: "Thanos Rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} is failing to evaluate rules."
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>
- alert: ThanosRuleNoEvaluationFor10Intervals
expr: |
time() - max by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name, rule_group) (prometheus_rule_group_last_evaluation_timestamp_seconds{app="thanos-rule"})
>
10 * max by (kubernetes_cluster,kubernetes_namespace, kubernetes_pod_name, rule_group) (prometheus_rule_group_interval_seconds{app="thanos-rule"})
for: 5m
labels:
team: infra
annotations:
summary: Thanos Rule {{$labels.namespace}}/{{$labels.kubernetes_pod_name}} has rule groups that did not evaluate for 10 intervals.
description: The rule group {{$labels.rule_group}} did not evaluate for at least 10x of their expected interval.
impact: "Alerts are not evaluated hence they wont be fired even if conditions are met"
dashboard: <https://grafana.$ENVIRONMENT.$PROVIDER.uw.systems/d/35da848f5f92b2dc612e0c3a0577b8a1/thanos-rule?refresh=5sv"|link>
logs: <https://grafana.$ENVIRONMENT.aws.uw.systems/explore?left=["now-1h","now","Loki",{"expr":"{kubernetes_cluster=\"{{$labels.kubernetes_cluster}}\",kubernetes_namespace=\"{{$labels.kubernetes_namespace}}\",kubernetes_pod_name=~\"{{$labels.kubernetes_pod_name}}\"}"}]|link>