feat(metrics): add s3 request error rate to alert rules (#898)

Signed-off-by: Shichao Nie <[email protected]>
AutoMQ · Mar 7, 2024 · a504f97 · a504f97
1 parent c5b6274
commit a504f97
Show file tree

Hide file tree

Showing 2 changed files with 48 additions and 0 deletions.
diff --git a/docker/telemetry/prometheus/rules/alert_rules_template.yml b/docker/telemetry/prometheus/rules/alert_rules_template.yml
@@ -120,3 +120,12 @@ groups:
         annotations:
           summary: "High group consumer lag {{ printf \"%0.f\" $value }} for consumer group {{ $labels.consumer_group }} in cluster {{ $labels.job }} on topic {{ $labels.topic }}."
           description: "The consumer lag of consumer group {{ $labels.consumer_group }} in cluster {{ $labels.job }} on topic {{ $labels.topic }} is exceeding threshold."
+
+      - alert: KafkaHighS3RequestErrorRate
+        expr: sum(rate(kafka_stream_operation_latency_count{operation_type="S3Request", status="failed"}[1m])) by (job, operation_name) > 0.1
+        for: 1m
+        labels:
+          severity: critical
+        annotations:
+          summary: "High Kafka S3 request {{ $labels.operation_name }} error rate {{ printf \"%0.2f\" $value }} req/s for cluster {{ $labels.job }}"
+          description: "The S3 request {{ $labels.operation_name }} error rate of Kafka cluster {{ $labels.job }} is exceeding threshold."
diff --git a/docker/telemetry/prometheus/rules/unit_tests.yml b/docker/telemetry/prometheus/rules/unit_tests.yml
@@ -517,3 +517,42 @@ tests:
             exp_annotations:
               summary: "High group consumer lag 12000 for consumer group test-group in cluster cluster_1 on topic test-topic."
               description: "The consumer lag of consumer group test-group in cluster cluster_1 on topic test-topic is exceeding threshold."
+  # Test KafkaHighS3RequestErrorRate
+  - interval: 1m
+    # Series data.
+    input_series:
+      - series: 'kafka_stream_operation_latency_count{job="cluster_1", instance="0", operation_name="get_object", operation_type="S3Request", status="failed"}'
+        values: '0 5 60 180 0 0'
+      - series: 'kafka_stream_operation_latency_count{job="cluster_1", instance="0", operation_name="get_object", operation_type="S3Request", status="success"}'
+        values: '0 100 200 300 400 500'
+      - series: 'kafka_stream_operation_latency_count{job="cluster_1", instance="0", operation_name="append", operation_type="S3Stream"}'
+        values: '0 100 200 300 400 500'
+
+    alert_rule_test:
+      # Test no alert.
+      - eval_time: 0m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+      - eval_time: 1m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+      - eval_time: 2m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+      - eval_time: 4m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+      - eval_time: 5m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+      # Test alert.
+      - eval_time: 3m
+        alertname: KafkaHighS3RequestErrorRate
+        exp_alerts:
+          - exp_labels:
+              severity: critical
+              job: cluster_1
+              operation_name: get_object
+            exp_annotations:
+              summary: "High Kafka S3 request get_object error rate 2.00 req/s for cluster cluster_1"
+              description: "The S3 request get_object error rate of Kafka cluster cluster_1 is exceeding threshold."