问题描述

在微软云环境中，使用python SDK连接存储账号（Storage Account）需要计算Blob大小？虽然Azure提供了一个专用工具Azure Storage Explorer可以统计出Blob的大小：

 

但是它也是只能一个Blob Container一个的统计，如果Container数量巨大，这将是一个繁琐的工作。而作为开发者，应该让代码来帮助完成。下文使用最快上手的Python代码来计算Blob中容量的大小。

 

完整代码

 
import os, uuid, datetime, threading
import logging
from azure.storage.blob import BlobServiceClient, BlobClient, ContainerClient, __version__


def calculateBlob(connect_string, count):
    try:
        blob_service_client = BlobServiceClient.from_connection_string(connect_string)
    except Exception as e:
        messages = str(count) + "Connect_String Error, Messages:" + e.args.__str__()
        print(messages)
        logging.info(messages)
    else:
        all_containers = blob_service_client.list_containers()
        for c in all_containers:
            count_name = c.name
            print(count_name)
            if count_name not in blobSize_Total:
                blobSize_Total[count_name] = 0
            if count_name not in blobSize_Daily:
                blobSize_Daily[count_name] = 0
            container_client = blob_service_client.get_container_client(count_name)
            generator = container_client.list_blobs()

            total_size_container = 0
            daily_size_container = 0

            for blob in generator:
                total_size_container += blob.size
                blob_create_time = blob.creation_time.strftime("%Y%m%d")
                if blob_create_time != now_date:
                    continue
                else:
                    # Calculate BlobSize in this month
                    daily_size_container += blob.size
                    # blobSize_Daily[count_name] += blob.size  # /(1024*1024)  # content_length - bytes

            blobSize_Total[count_name] += total_size_container / (1024 * 1024)
            blobSize_Daily[count_name] += daily_size_container / (1024 * 1024)

    return None


if __name__ == '__main__':
    # connect string
    Connection_String_List ="DefaultEndpointsProtocol=https;AccountName=<storagename>;AccountKey=<key>;EndpointSuffix=core.chinacloudapi.cn"
    # for i in Connection_String:
    start = datetime.datetime.now()
    print(start)

    # 定义全局变量 - blobSize_Daily & blobSize_Total
    blobSize_Daily = {}
    blobSize_Total = {}

    now_date = datetime.datetime.now().strftime("%Y%m%d")

    print("开始计算")
    calculateBlob(Connection_String_List, 1)
    print("计算完成")

    print("统计当前新增大小")
    print(blobSize_Daily)
    print("统计Blob总大小")
    print(blobSize_Total)
    end = datetime.datetime.now()
    print(end)

如运行是没有Azure blob模块，可以使用 pip install azure-storage-blob 安装。以上代码运行结果如下：

 

 

如果有多个Storage Account，可以考虑加入多线程的方式来运行，在代码中增加一个myThread类，然后在 __main__ 中把 calculateBlob(Connection_String_List, 1) 运行替换为 many_thread(Connection_String_List) 即可。

class myThread(threading.Thread):

    def __init__(self, threadID, name, connection_string):
        threading.Thread.__init__(self)
        self.threadID = threadID
        self.name = name
        self.connection_string = connection_string

    def run(self):
        print("开始线程：" + self.name)
        calculateBlob(self.connection_string, self.threadID)
        print("退出线程：" + self.name)


def many_thread(Connection_String_List):
    threads = []
    for i in range(len(Connection_String_List)):  # 循环创建多个线程
        t = myThread(i, "Thread-" + str(i), Connection_String_List[i])
        threads.append(t)
    for t in threads:  # 循环启动线程 一个线程对应一个连接字符串
        t.start()
    for t in threads:
        t.join()

 

遇见问题

在多线程执行时，可能会遇见问题：("Connection broken: ConnectionResetError(10054, 'An existing connection was forcibly closed by the remote host', None, 10054, None)", ConnectionResetError(10054, 'An existing connection was forcibly closed by the remote host', None, 10054, None))，出现此问题大都是由于客户端使用了已经断开的连接导致所导致的。所以一定要仔细调试多线程关闭代码。是否是把还需要运行的线程给关闭了。导致了以上的错误消息。

 

附录一：多线程计算Blob的完整代码

import os, uuid, datetime, threading
import logging
from azure.storage.blob import BlobServiceClient, BlobClient, ContainerClient, __version__


def calculateBlob(connect_string, count):
    try:
        blob_service_client = BlobServiceClient.from_connection_string(connect_string)
    except Exception as e:
        messages = str(count) + "Connect_String Error, Messages:" + e.args.__str__()
        print(messages)
        logging.info(messages)
    else:
        all_containers = blob_service_client.list_containers()
        for c in all_containers:
            count_name = c.name
            print(count_name)
            if count_name not in blobSize_Total:
                blobSize_Total[count_name] = 0
            if count_name not in blobSize_Daily:
                blobSize_Daily[count_name] = 0
            container_client = blob_service_client.get_container_client(count_name)
            generator = container_client.list_blobs()

            total_size_container = 0
            daily_size_container = 0

            for blob in generator:
                total_size_container += blob.size
                blob_create_time = blob.creation_time.strftime("%Y%m%d")
                if blob_create_time != now_date:
                    continue
                else:
                    # Calculate BlobSize in this month
                    daily_size_container += blob.size
                    # blobSize_Daily[count_name] += blob.size  # /(1024*1024)  # content_length - bytes

            blobSize_Total[count_name] += total_size_container / (1024 * 1024)
            blobSize_Daily[count_name] += daily_size_container / (1024 * 1024)

    return None

class myThread(threading.Thread):

    def __init__(self, threadID, name, connection_string):
        threading.Thread.__init__(self)
        self.threadID = threadID
        self.name = name
        self.connection_string = connection_string

    def run(self):
        print("开始线程：" + self.name)
        calculateBlob(self.connection_string, self.threadID)
        print("退出线程：" + self.name)


def many_thread(Connection_String_List):
    threads = []
    for i in range(len(Connection_String_List)):  # 循环创建多个个线程
        t = myThread(i, "Thread-" + str(i), Connection_String_List[i])
        threads.append(t)
    for t in threads:  # 循环启动线程 - 一个线程对应一个连接字符串
        t.start()
    for t in threads:
        t.join()


if __name__ == '__main__':
    # connect string
    Connection_String_List =  ['DefaultEndpointsProtocol=https;AccountName=<your storage account 1>;AccountKey=<Key 1>;EndpointSuffix=core.chinacloudapi.cn', 'DefaultEndpointsProtocol=https;AccountName=<your storage account 2>;AccountKey=<Key 2>;EndpointSuffix=core.chinacloudapi.cn']
    # for i in Connection_String:
    start = datetime.datetime.now()
    print(start)

    # 定义全局变量 - blobSize_Daily & blobSize_Total
    blobSize_Daily = {}
    blobSize_Total = {}

    now_date = datetime.datetime.now().strftime("%Y%m%d")

    many_thread(Connection_String_List)
    print("Main Thread End")

    print(blobSize_Daily)
    print(blobSize_Total)
    end = datetime.datetime.now()
    print(end)

运行效果：

 

 

 

参考资料

快速入门：使用 Python v12 SDK 管理 blob ：https://docs.azure.cn/zh-cn/storage/blobs/storage-quickstart-blobs-python

Python 列表(List) : https://www.runoob.com/python/python-lists.html

BlobServiceClient Class : https://docs.microsoft.com/en-us/python/api/azure-storage-blob/azure.storage.blob.blobserviceclient?view=azure-python