使用见解监视和配置 Azure 负载均衡器Using Insights to monitor and configure your Azure Load Balancer

通过用于网络的 Azure Monitor,可以获得适用于负载均衡器的函数依赖关系可视化和预配置的指标仪表板。Through Azure Monitor for networks, you're provided functional dependency visualizations and pre-configured metrics dashboard for your Load Balancers. 这些视觉对象有助于你制定明智的设计决策,并快速定位、诊断和解决任何故障。These visuals help empower you to make informed design decisions and rapidly localize, diagnose, and resolve any faults.

备注

请注意,此功能为预览版,函数依赖关系视图和预配置的仪表板以后可能会改变以优化体验Please note this feature is in Preview and the functional dependency view and preconfigured dashboard may change to improve this experience

重要

需要标准负载均衡器才能在预配置的指标仪表板中查看来自负载均衡器命名空间的指标。The Standard Load Balancer is required to see metrics from the Load Balancer namespace in the pre-configured metrics dashboard. 你仍然可以从 VM、虚拟机规模集和连接监视器命名空间中看到指标,但是,对于任何生产工作负载,我们建议升级到标准,以充分利用可靠的负载均衡器指标集。You will still be able to see metrics from the VM, virtual machine scale set, and Connection Monitor namespaces however, we recommend upgrading to Standard for any production workloads to take advantage of the robust set of Load Balancer metrics.

函数依赖关系视图Functional dependency view

借助函数依赖关系视图,甚至可以直观显示最复杂的负载均衡器设置。The functional dependency view will enable you to picture even the most complex load balancer setups. 通过负载均衡器最新配置的视觉反馈,你可以在更新的同时保留对配置的记忆。With visual feedback on your latest Load Balancer configuration, you can make updates while keeping your configuration in mind.

你可以通过访问 Azure 中的负载均衡器资源的“见解”边栏选项卡来访问此视图。You can access this view by visiting the Insights blade of your Load Balancer resource in Azure.

函数依赖关系视图描述。可以看到,负载均衡器的前端通过配置的规则连接到后端池成员。对于标准负载均衡器,连接负载均衡规则与后端池实例的直线根据运行状况探测状态进行了颜色编码。

对于标准负载平衡器,后端池资源基于运行状况探测状态进行了颜色编码,指示用于提供流量的后端池的当前可用性。For Standard Load Balancers, your backend pool resources are color-coded with Health Probe status indicating the current availability of your backend pool to serve traffic. 除了上述拓扑外,还为你提供了运行状况的时间图,并提供了应用程序运行状况的快照视图。Alongside the above topology you're presented with a time-wise graph of health status, giving a snapshot view of the health of your application.

指标仪表板Metrics dashboard

在负载均衡器的“见解”边栏选项卡中,可以选择“更多详细指标”,查看预配置的 Azure Monitor 工作簿,其中包含与负载均衡器特定方面相关的指标视觉对象。From the Insights blade of your Load Balancer, you can select More Detailed Metrics to view a pre-configured Azure Monitor Workbook containing metrics visuals relevant to specific aspects of your Load Balancer. 该仪表板显示负载均衡器状态,并在页面顶部显示相关文档的链接。This dashboard will show the Load Balancer status and links to relevant documentation at the top of the page.

首先会显示“概述”选项卡。可以在可用选项卡中导航,每个选项卡均包含与负载均衡器特定方面相关的视觉对象。At first you'll be presented with the Overview tab. You can navigate through the available tabs each of which contain visuals relevant to a specific aspect of your Load Balancer. 每个选项卡底部的仪表板中都提供针对每个选项的明确指导。Explicit guidance for each is available in the dashboard at the bottom of each tab.

当前可用的仪表板选项卡是:The dashboard tabs currently available are:

  • 概述Overview
  • 前端和后端可用性Frontend and Backend Availability
  • 数据吞吐量Data Throughput
  • 流分发Flow Distribution
  • 连接监视器Connection Monitors
  • 指标定义Metric Definitions

概述选项卡Overview tab

“概述”选项卡包含一个可搜索的网格,其中包含连接到负载均衡器的每个前端 IP 的总体数据路径可用性和运行状况探测状态。The Overview tab contains a searchable grid with the overall Data Path Availability and Health Probe Status for each of the Frontend IPs attached to your Load Balancer. 这些指标指示前端 IP 是否响应,并且后端池中的计算实例是否分别响应入站连接。These metrics indicate whether the Frontend IP is responsive and the compute instances in your Backend Pool are individually responsive to inbound connections.

还可以在此页面上查看每个前端 IP 的总体数据吞吐量,了解是否正在产生和接收预期水平的流量。You can also view the overall data throughput for each Frontend IP on this page to get a sense of whether you are producing and receive expected traffic levels. 如果看到任何不正常的值,可通过页面底部的指导进入正确的标签。The guidance at the bottom of the page will direct you to the appropriate tab should you see any irregular values.

“前端和后端可用性”选项卡Frontend and Backend Availability tab

“前端和后端可用性”选项卡以一些有用的视图显示了数据路径吞吐量和运行状况探测状态指标。The Frontend and Backend Availability tabs show the Data Path Throughput and Health Probe Status metrics presented in a few useful views. 第一张图显示了聚合值,因此你可以确定是否存在问题。The first graph shows the aggregate value so you can determine whether there's an issue. 其余图显示这些指标按不同维度进行划分,以便你可以排除故障并确定任何入站可用性问题的根源。The rest of the graphs show these metrics split by various dimensions so that you can troubleshoot and identify the sources of any inbound availability issues.

页面底部提供了用于查看这些图的工作流,其中包括各种症状的常见原因。A workflow for viewing these graphs is provided at the bottom of the page with common causes for various symptoms.

“数据吞吐量”选项卡Data Throughput tab

“数据吞吐量”选项卡使你可以查看入站和出站吞吐量,确定流量模式是否符合预期。The Data Throughput tab allows you to review your inbound and outbound throughput to identify if your traffic patterns are as expected. 它将显示按前端 IP 和前端端口划分的入站和出站数据吞吐量,以便你可以确定所运行的服务是否单独执行。It will show the inbound and outbound data throughput split by Frontend IP and Frontend Port so that you can identify if how the services you have running are performing individually.

流分发Flow Distribution

“流量分配”选项卡可帮助你直观了解和管理后端实例正在接收和产生的流量数量。The Flow Distribution Tab will help you visualize and manage the number of flows your backend instances are receiving and producing. 它显示入站和出站流量的流创建速率和流计数,以及每个 VM 和虚拟机规模集实例正在接收的网络流量。It shows the Flow Creation Rate and Flow Count for inbound and outbound traffic as well as the Network Traffic each VM and virtual machine scale set instance is receiving.

这些视图可为你提供有关负载均衡器配置或流量模式是否导致流量不平衡的反馈。These views can give you feedback on whether your Load Balancer configuration or traffic patterns are leading to imbalanced traffic. 例如,是否已配置会话关联,以及某个客户端发出的请求数量是否不成比例。For example, if you have session affinity configured and a single client is making a disproportionate number of requests. 它还会告知你的计算机大小是否已达到每个 VM 流量限制It will also let you know if you are approaching the per VM flow limit for your machine size.

连接监视器Connection Monitors

“连接监视器”选项卡会在全局地图上显示已配置的所有连接监视器的往返延迟。The Connection Monitors tab will show you the round-trip latency on a global map for all of the Connection Monitors you've configured. 这些视觉对象为具有严格延迟要求的服务提供有用的信息。These visuals provide useful information for services with strict latency requirements. 为了满足自己的要求,你可能需要添加其他区域部署。To meet your requirements you may need to add additional regional deployments.

指标定义Metric Definitions

“指标定义”选项卡包含“多维指标标准”文章中显示的所有信息。The Metric Definitions tab contains all the information shown in the Multi-dimensional Metrics article.

后续步骤Next steps