Prometheus Cloudwatch 导出器配置

Question

我有 Prometheus CloudWatch Exporter 设置并指向我们的 prometheus 实例。我正在构建一个配置，但我似乎无法让我的所有指标都显示在 Prometheus 中。

我目前有 aws_applicationelb_request_count_average 和 aws_ec2_network_packets_in_sum 工作，但没有别的..

这是我的配置：

  region: us-east-1

  metrics:
  - aws_namespace: AWS/ApplicationELB
    aws_metric_name: HealthyHostCount
    aws_dimensions: [AvailabilityZone, LoadBalancer]
    aws_statistics: [Average]

  - aws_namespace: AWS/ApplicationELB
    aws_metric_name: UnHealthyHostCount
    aws_dimensions: [AvailabilityZone, LoadBalancer]
    aws_statistics: [Average]

  - aws_namespace: AWS/ApplicationELB
    aws_metric_name: RequestCount
    aws_dimensions: [AvailabilityZone, LoadBalancer]
    aws_statistics: [Average]

  - aws_namespace: AWS/ApplicationELB
    aws_metric_name: HTTPCode_ELB_5XX_Count
    aws_dimensions: [LoadBalancer]
    aws_statistics: [Average]

  - aws_namespace: AWS/ApplicationELB
    aws_metric_name: RequestCountPerTarget
    aws_dimensions: [LoadBalancer, TargetGroup]
    aws_statistics: [Average]

  - aws_namespace: AWS/EC2
    aws_metric_name: NetworkPacketsIn
    aws_dimensions: [InstanceId]
    aws_statistics: [Maximum, Sum]

Answer 1

能否详细说明一下

是什么意思

but I can't seem to get all of my metrics to show up in Prometheus.

您是否检查过预期指标是否出现在 aws cloudwatch list-metrics --namespace AWS/ApplicationELB 的输出中以及 AWS/EC2 和 get-metric-statistics 命令的输出中是否类似？您也可以试试 this exporter。它使用不同的 CloudWatch API 来降低 AWS API 调用成本

Answer 2

我想通了这个问题。我试图从没有可用指标的 AWS 命名空间收集指标。添加指标后，可用数据开始显示..

Prometheus Cloudwatch 导出器配置

Prometheus Cloudwatch Exporter Config

monitoring

amazon-cloudwatch

prometheus