【Azure 环境】【Azure Developer】使用Python代码获取Azure 中的资源的Metrics定义及数据

问题描述

使用Python SDK来获取Azure上的各种资源的Metrics的名称以及Metrics Data的示例

问题解答

通过 azure-monitor-query ，可以创建一个 metrics client，调用 client.list_metric_definitions 来获取Metrics 定义，然后通过 client.query_resource 获取Metrics data。

关键函数为：

#第一步：定义 client
client = MetricsQueryClient(credential=credential, endpoint='https://management.chinacloudapi.cn',
audience='https://management.chinacloudapi.cn')


#第二步：获取metrics name
response = client.list_metric_definitions(metric_uri)


#第三步：获取 metrcis data
response = client.query_resource(
        resource_uri=url,
        metric_names=[name],
        timespan=timedelta(hours=2),
        granularity=timedelta(minutes=5),
        aggregations=[MetricAggregationType.AVERAGE],
        )

需要注意：

endpoint 和 audience 需要根据代码的使用Azure环境不同而改变，以上为中国区Azure的Endpoint。与Global Azure 终结点对比文档见： Azure China developer guide | Microsoft Docs
metrics_url 可以在Azure 门户中的“属性”页面获取，当然也可以通过Python对于资源的SDK进行获取。示例代码见附录一.

全部示例代码：

# import required package
from ast import Try
from warnings import catch_warnings
from datetime import timedelta
from azure.monitor.query import MetricsQueryClient, MetricAggregationType
from azure.identity import AzureCliCredential   ## pip install azure-identity


# prepare credential
credential = AzureCliCredential()

#init metric query client, endpoint need to target China Azure
client = MetricsQueryClient(credential=credential, endpoint='https://management.chinacloudapi.cn',
audience='https://management.chinacloudapi.cn')


def printMetricsDataByName(url, name):
    ##metrics_uri =metric_uri; ### os.environ.get('METRICS_RESOURCE_URI')
    response = client.query_resource(
        resource_uri=url,
        metric_names=[name],
        timespan=timedelta(hours=2),
        granularity=timedelta(minutes=5),
        aggregations=[MetricAggregationType.AVERAGE],
        )

    for metric in response.metrics:
        print(metric.name + ' -- ' + metric.display_description)
        for time_series_element in metric.timeseries:
            for metric_value in time_series_element.data:
                print('\tThe {}  at {} is {}'.format(
                    name,
                    metric_value.timestamp,
                    metric_value.average
                ))




print("###  ..Special Reource URL.. ....")
# specific resource uri
metric_uri = '/subscriptions/<your-subscriptions-id>/resourceGroups/<your-resource-group>/providers/Microsoft.Cache/Redis/<your-resource-name>'

# do query...
response = client.list_metric_definitions(metric_uri)

for item in response:
    print(item.name + " ......  Metrics Data  ......")
    try:
        printMetricsDataByName(metric_uri,item.name)
    except Exception as e:
        print(e)

测试效果图：

附录一：例如在代码中获取Redis资源的Resource ID

from azure.mgmt.redis import RedisManagementClient  ## pip install azure-mgmt-redis
from azure.identity import AzureCliCredential   ## pip install azure-identity

# prepare credential
credential = AzureCliCredential()

redisClient = RedisManagementClient(credential, '<YOUR SUB>', 
base_url='https://management.chinacloudapi.cn', 
credential_scopes=[https://management.chinacloudapi.cn/.default])

for item in redisClient.redis.list_by_subscription():
    print(item.id)

以上代码执行结果：

附录二：credential = AzureCliCredential() 为访问Azure资源时提供认证授权的方法，如果出现权限不够，或者是无法访问的情况，会出现类似如下的提示，需要根据消息提示来解决权限问题。

Code: AuthorizationFailed
Message: The client '[email protected]' with object id 'xxxxxxxx-xxxx-xxxx-xxxxx-xxxxxxxxxxxx' 
does not have authorization to perform action 'Microsoft.Insights/metricDefinitions/read' 
over scope '/subscriptions/xxxxxxxx-xxxx-xxxx-xxxxx-xxxxxxxxxxxx/resourceGroups/xxxx-resource-group/providers/Microsoft.Cache/Redis/redis-xxxxxx/providers/Microsoft.Insights' 
or the scope is invalid. If access was recently granted, please refresh your credentials.