AWS S3 transferutility上传从不上传,但显示完整且没有错误

时间:2012-11-25 01:46:53

标签: c# amazon-s3 amazon-web-services

我收到很多未上传到我的AWS桶的文件,但下面的代码总是显示完整。有人知道这有什么意义吗?我怎么能知道上传是否失败,如果它总是回来完成的话?我确实看到了一些关于使用transferutility.EndUpload的东西,但我不确定如何使用它。我还将如何实施重试?将对象状态传递给BeginUpload?有什么帮助吗?

public class S3Upload
{
    private string awsAccessKeyId = "XXXXXX";
    private string awsSecretAccessKey = "XXXXXX";
    private string bucketName = System.Configuration.ConfigurationManager.AppSettings.Get("BucketName");
    private Amazon.S3.Transfer.TransferUtility transferUtility;
    private static log4net.ILog log = log4net.LogManager.GetLogger(System.Reflection.MethodBase.GetCurrentMethod().DeclaringType);

    public S3Upload()
    {
        // Initialize log4net.
        log4net.Config.XmlConfigurator.Configure();
        this.transferUtility = new Amazon.S3.Transfer.TransferUtility(awsAccessKeyId, awsSecretAccessKey);
        log.Info("S3 instance initiated");

    }

    public void UploadFile(string filePath, string toPath)
    {

        try
        {
            AsyncCallback callback = new AsyncCallback(uploadComplete);

            log.Info("S3 upload started...");
            log.InfoFormat("S3 filePath: {0}", filePath);
            log.InfoFormat("S3 toPath: {0}", toPath);


            var uploadRequest = new Amazon.S3.Transfer.TransferUtilityUploadRequest();
            uploadRequest.FilePath = filePath;
            uploadRequest.BucketName = bucketName;
            uploadRequest.Key = toPath;
            uploadRequest.StorageClass = Amazon.S3.Model.S3StorageClass.ReducedRedundancy;
            uploadRequest.AddHeader("x-amz-acl", "public-read");
            transferUtility.BeginUpload(uploadRequest, callback, null);
        }
        catch (AmazonS3Exception amazonS3Exception)
        {
              log.ErrorFormat("An Error, number {0}, occurred when creating a bucket with the message '{1}", amazonS3Exception.ErrorCode, amazonS3Exception.Message);    
        }
    }

    private void uploadComplete(IAsyncResult result)
    {
        var x = result;

        if (x.IsCompleted)
        {
            log.Info("S3 upload completed...");

        }
    }
}

2 个答案:

答案 0 :(得分:1)

找到了修复!我添加了transferUtility.EndUpload(ar),如果它出错,它会将异常写入我的日志并重试put请求。

发出请求PutObject时出错。 System.IO.IOException:无法将数据写入传输连接:远程主机强制关闭现有连接。 --->

public class S3Upload
{
    private string awsAccessKeyId = "XXXXX";
    private string awsSecretAccessKey = "XXXXX";
    private string bucketName = System.Configuration.ConfigurationManager.AppSettings.Get("BucketName");
    private Amazon.S3.Transfer.TransferUtility transferUtility;
    private static log4net.ILog log = log4net.LogManager.GetLogger(System.Reflection.MethodBase.GetCurrentMethod().DeclaringType);

    public S3Upload()
    {
        // Initialize log4net.
        log4net.Config.XmlConfigurator.Configure();
        this.transferUtility = new Amazon.S3.Transfer.TransferUtility(awsAccessKeyId, awsSecretAccessKey);
        log.Info("S3 instance initiated");

    }

    public void UploadFile(string filePath, string toPath)
    {

        try
        {
            AsyncCallback callback = new AsyncCallback(uploadComplete);

            log.Info("S3 upload started...");
            log.InfoFormat("S3 filePath: {0}", filePath);
            log.InfoFormat("S3 toPath: {0}", toPath);


            var uploadRequest = new Amazon.S3.Transfer.TransferUtilityUploadRequest();
            uploadRequest.FilePath = filePath;
            uploadRequest.BucketName = bucketName;
            uploadRequest.Key = toPath;
            uploadRequest.StorageClass = Amazon.S3.Model.S3StorageClass.ReducedRedundancy;
            uploadRequest.AddHeader("x-amz-acl", "public-read");
            IAsyncResult ar = transferUtility.BeginUpload(uploadRequest, callback, null);
            transferUtility.EndUpload(ar);
        }
        catch (AmazonS3Exception amazonS3Exception)
        {
              log.ErrorFormat("An Error, number {0}, occurred when creating a bucket with the message '{1}", amazonS3Exception.ErrorCode, amazonS3Exception.Message);
        }
    }

    private void uploadComplete(IAsyncResult result)
    {
        var x = result;
    }
}

答案 1 :(得分:1)

BeginUpload函数启动一个启动并执行上载的新线程。调用EndUpload会导致当前线程阻塞,直到上载完成。此外,EndUpload函数会捕获您从上传中获得的任何异常。

如果你打算调用BeginUpload,你需要从一个单独的线程调用EndUpload(),这样你就不会阻塞你的主线程。如果您打算阻止主线程,只需调用Upload()函数而不是APM版本。

例如,生成一个新线程,该线程将等待上传完成。 (请注意,它不必执行任何特定操作,只需捕获错误。)

public void UploadFile(string filePath, string toPath)
{

    try
    {
        AsyncCallback callback = new AsyncCallback(uploadComplete);

        log.Info("S3 upload started...");
        log.InfoFormat("S3 filePath: {0}", filePath);
        log.InfoFormat("S3 toPath: {0}", toPath);


        var uploadRequest = new Amazon.S3.Transfer.TransferUtilityUploadRequest();
        uploadRequest.FilePath = filePath;
        uploadRequest.BucketName = bucketName;
        uploadRequest.Key = toPath;
        uploadRequest.StorageClass = Amazon.S3.Model.S3StorageClass.ReducedRedundancy;
        uploadRequest.AddHeader("x-amz-acl", "public-read");
        IAsyncResult ar = transferUtility.BeginUpload(uploadRequest, callback, null);
        // transferUtility.EndUpload(ar);
        ThreadPool.QueueUserWorkItem(c =>
        {
             try
             {
                 transferUtility.EndUpload(ar);
             }
             catch (AmazonS3Exception ex)
             {
                 // Retry, log the error, etc.
             }
             catch (WebException ex)
             {
                 // Retry, log the error, etc.
             }
         });
    }
    catch (AmazonS3Exception amazonS3Exception)
    {
          log.ErrorFormat("An Error, number {0}, occurred when creating a bucket with the message '{1}", amazonS3Exception.ErrorCode, amazonS3Exception.Message);
    }
}

此外,该线程还会检查WebException。我不知道你是否可以从Beginupload获得其中的一个,但我正在研究的一些代码似乎在关注那些。 (更安全然后抱歉。)

另外,我发现你可以使用CurrPorts Utility进行测试。该实用程序允许您将开放端​​口杀死为亚马逊(尽管它将在其配置的MaxRetries之前重试)。