跳到主要内容

用夜莺监控 Azure Monitor

通过 Azure Monitor API 按资源类型拉取指标，用服务主体鉴权，只需 Monitoring Reader 角色。

最后校对于

Azure 插件调用 Azure Monitor 的指标 API，按订阅、资源组或资源类型（虚拟机、存储账户、SQL 数据库、应用服务等）拉取指标，配置里可以按资源类型指定要采的指标名和聚合方式。

需要注册一个服务主体，授予 Monitoring Reader 角色，把租户 ID、客户端 ID、密钥和订阅 ID 填进配置。Azure API 每小时限 12000 次读取，采集周期和指标数量要控制在这个额度内。

模板里有什么​

  • Categraf 采集配置：input.azure

怎么接入​

  1. 安装 Categraf，把下面的配置放到 conf/input.azure/azure.toml，按实际地址和凭据改好后重启 Categraf。启动日志里出现 input: local.azure started 就说明插件已加载。

    # Gather Azure resources metrics from Azure Monitor API
    # 每2m查询一次azure monitor, 查询时间范围也为2m
    interval = "2m"
    [[instances]]
    # can be found under Overview->Essentials in the Azure portal for your application/service
    subscription_id = ""
    # can be obtained by registering an application under Azure Active Directory
    client_id = ""
    # can be obtained by registering an application under Azure Active Directory.
    # If not specified Default Azure Credentials chain will be attempted:
    # - Environment credentials (AZURE_*)
    # - Workload Identity in Kubernetes cluster
    # - Managed Identity
    # - Azure CLI auth
    # - Developer Azure CLI auth
    client_secret = ""
    # can be found under Azure Active Directory->Properties
    tenant_id = ""
    # Define the optional Azure cloud option e.g. AzureChina, AzureGovernment or AzurePublic. The default is AzurePublic.
    # cloud_option = "AzurePublic"

    # 聚合粒度
    # 比如我采集周期是10m, 指标上报周期是30s, 那么在10m内会有20个原始点
    # 经过该参数的聚合处理后，比如聚合粒度是1m, 那么就会获取到10个点
    # 注意，最小值是1m
    # 支持的值为1m,5m,15m,30m,1h,6h,12h,1d
    # 且aggregation_interval不能大于interval
    aggregation_interval = "1m"

    ## 以下为采集目标配置，有三种方式，分别是指定资源、指定资源组、指定订阅
    ## 三种方法可同时使用
    # 方式一：指定具体的资源, 比如某台具体的主机
    # [[instances.resource_target]]
    # can be found under Overview->Essentials->JSON View in the Azure portal for your application/service
    # must start with 'resourceGroups/...' ('/subscriptions/xxxxxxxx-xxxx-xxxx-xxx-xxxxxxxxxxxx'
    # must be removed from the beginning of Resource ID property value)
    # resource_id = "resourceGroups/flashcat/providers/Microsoft.Compute/virtualMachines/lfn-test"

    # the metric names to collect
    # leave the array empty to use all metrics available to this resource
    # metrics = [ "<<METRIC>>", "<<METRIC>>" ]
    # metrics aggregation type value to collect
    # can be 'Total', 'Count', 'Average', 'Minimum', 'Maximum'
    # leave the array empty to collect all aggregation types values for each metric
    # aggregations = [ "<<AGGREGATION>>", "<<AGGREGATION>>" ]

    # 方式二：指定资源组
    # # resource group target #1 to collect metrics from resources under it with resource type
    # [[instances.resource_group_target]]
    # # the resource group name
    # resource_group = "flashcat"
    #
    # # defines the resources to collect metrics from
    # [[instances.resource_group_target.resource]]
    # # the resource type
    # resource_type = "Microsoft.Compute/virtualMachines"
    # metrics = [ "<<METRIC>>", "<<METRIC>>" ]
    # aggregations = [ "<<AGGREGATION>>", "<<AGGREGATION>>" ]

    # 方式三：指定订阅ID
    # # subscription target #1 to collect metrics from resources under it with resource type
    # [[instances.subscription_target]]
    # resource_type = "<<RESOURCE_TYPE>>"
    # metrics = [ "<<METRIC>>", "<<METRIC>>" ]
    # aggregations = [ "<<AGGREGATION>>", "<<AGGREGATION>>" ]

相关链接​

本页目录