You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@jclouds.apache.org by "Andrew Gaul (JIRA)" <ji...@apache.org> on 2014/07/23 19:55:38 UTC

[jira] [Comment Edited] (JCLOUDS-639) Provide methods to get the progress of a running upload to a blobstore

    [ https://issues.apache.org/jira/browse/JCLOUDS-639?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14072028#comment-14072028 ] 

Andrew Gaul edited comment on JCLOUDS-639 at 7/23/14 5:54 PM:
--------------------------------------------------------------

You can hack this in with a InputStream payload with a CountingInputStream wrapper.  Note that InputStream payloads disable retries on network errors, so you would really want to create a CountingByteSource implementation instead.  I believe that this is the best you can do with jclouds as-is; jclouds really needs proper async support for these long running tasks, since transferring large multi-part blobs can take hours or days.


was (Author: gaul):
You can hack this in with a InputStream payload with a CountingInputStream wrapper.  Note that InputStream payloads disable retries on network errors, but I belileve that this is the best you can do with jclouds as-is.  jclouds really needs proper async support for these long running tasks, since transferring large multi-part blobs can take hours or days.

> Provide methods to get the progress of a running upload to a blobstore
> ----------------------------------------------------------------------
>
>                 Key: JCLOUDS-639
>                 URL: https://issues.apache.org/jira/browse/JCLOUDS-639
>             Project: jclouds
>          Issue Type: Improvement
>          Components: jclouds-blobstore
>    Affects Versions: 1.7.3
>         Environment: doesn't matter
>            Reporter: Peter
>
> The jclouds library should provide a way to get progress information about a running upload to any blobstore (information like how many bytes where already transferred and optionally average speed of the transfer).
> In _JavaUrlHttpCommandExecutorService_, the outputStream is already being wrapped in a _CountingOutputStream_, but there is no way to call the _getCount()_ method from outside the method. 
> There should be some way to access the current count of bytes already sent to calculate the status of the upload. Additionally a way to get the average speed would be nice, but I guess this is more optional because you can calculate the speed on your own if you know how many bytes where already transferred.
> Another option to solve this, is to provide access to provider-specific features like the ProgressListener in the S3-API (see http://docs.aws.amazon.com/AWSJavaSDK/latest/javadoc/com/amazonaws/services/s3/model/ProgressListener.html), but I guess that not all vendors have something similar in their APIs, anyway the first approach would be a more generic one. 



--
This message was sent by Atlassian JIRA
(v6.2#6252)