From 25744dd67b70140bcfff2b6457c698a97705ef0b Mon Sep 17 00:00:00 2001 From: machine424 Date: Tue, 6 Apr 2021 17:59:32 +0200 Subject: [PATCH] [prometheus-rabbitmq-exporter] Polish up Prometheus rules (#724) Signed-off-by: machine424 --- .../prometheus-rabbitmq-exporter/Chart.yaml | 2 +- .../templates/prometheusrule.yaml | 86 ++++++++++++------- 2 files changed, 55 insertions(+), 33 deletions(-) diff --git a/charts/prometheus-rabbitmq-exporter/Chart.yaml b/charts/prometheus-rabbitmq-exporter/Chart.yaml index 2713d640..7d7f01d7 100644 --- a/charts/prometheus-rabbitmq-exporter/Chart.yaml +++ b/charts/prometheus-rabbitmq-exporter/Chart.yaml @@ -1,7 +1,7 @@ apiVersion: v1 description: Rabbitmq metrics exporter for prometheus name: prometheus-rabbitmq-exporter -version: 0.7.0 +version: 1.0.0 appVersion: v0.29.0 home: https://github.com/kbudde/rabbitmq_exporter sources: diff --git a/charts/prometheus-rabbitmq-exporter/templates/prometheusrule.yaml b/charts/prometheus-rabbitmq-exporter/templates/prometheusrule.yaml index 1f9b3f44..92543a4a 100644 --- a/charts/prometheus-rabbitmq-exporter/templates/prometheusrule.yaml +++ b/charts/prometheus-rabbitmq-exporter/templates/prometheusrule.yaml @@ -12,36 +12,58 @@ metadata: {{- end }} spec: groups: - - name: rabbitmq - rules: - - alert: RabbitmqDown - expr: | - absent(rabbitmq_up == 1) - for: 1m - labels: - severity: critical - - alert: RabbitmqNotRunning - expr: | - rabbitmq_running{self="1"} != 1 - for: 1m - labels: - severity: critical - - alert: RabbitmqMessages - expr: | - rabbitmq_queue_messages > 250 - for: 10m - labels: - severity: warning - - alert: RabbitmqMessages - expr: | - rabbitmq_queue_messages > 500 - for: 10m - labels: - severity: critical - - alert: RabbitmqPartition - expr: | - rabbitmq_partitions{self="1"} != 0 - for: 1m - labels: - severity: critical + - name: {{ template "prometheus-rabbitmq-exporter.fullname" . }} + rules: + - alert: RabbitmqNodeDown + expr: rabbitmq_running{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} == 0 + for: 5m + labels: + severity: warning + annotations: + summary: A Rabbitmq node is down + description: | + The Rabbitmq node {{ "{{ $labels.node }}" }} of + the cluster tracked by {{ "{{ $labels.service }}" }} was not running during the last 5m. + - alert: RabbitmqClusterDown + expr: | + rabbitmq_up{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} == 0 + for: 5m + labels: + severity: critical + annotations: + summary: The Rabbitmq cluster {{ "{{ $labels.service }}" }} is maybe down. + description: | + The Rabbitmq exporter couldn't scrape any Rabbitmq node of the + cluster tracked by {{ "{{ $labels.service }}" }} during the last 5m, the cluster is maybe down. + - alert: RabbitMQClusterPartition + expr: rabbitmq_partitions{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} > 0 + for: 5m + labels: + severity: critical + annotations: + summary: A cluster partition was detected + description: | + Cluster partition in Rabbitmq cluster tracked by {{ "{{ $labels.service }}" }} was detected + by the node {{ "{{ $labels.node }}" }} + - alert: RabbitmqOutOfMemory + expr: | + rabbitmq_node_mem_used{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} + / rabbitmq_node_mem_limit{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} + * 100 > 90 + for: 5m + labels: + severity: warning + annotations: + summary: The Rabbitmq node {{ "{{ $labels.node }}" }} is Out of memory + description: | + Memory available for Rabbmitmq node {{ "{{ $labels.node }}" }} is lower than 10% + - alert: RabbitmqTooManyConnections + expr: rabbitmq_connectionsTotal{service="{{ template "prometheus-rabbitmq-exporter.fullname" . }}"} > 1000 + for: 5m + labels: + severity: warning + annotations: + summary: Too many connections to the Rabbitmq cluster + description: | + The Rabbitmq cluster tracked by {{ "{{ $labels.service }}" }} has too many connections {{ "{{ $value }}" }} (> 1000) {{- end }} \ No newline at end of file