I believe the issue is that the Hadoop QA bot runs from a subdirectory of
the project root, i.e. hadoop-common-project, hadoop-hdfs-project, etc.
Your patch only changes things under hadoop-dist, so I don't think it will
work no matter how you format the patch.

For this particular JIRA (HADOOP-8241) you should just comment saying
exactly what testing you did. A reviewer of the patch should probably
manually apply it and test it out before committing it.

--
Aaron T. Myers
Software Engineer, Cloudera



On Tue, Apr 3, 2012 at 10:03 AM, Mostafa Elhemali <mosta...@microsoft.com>wrote:

> Thanks Devaraj. I actually took care to convert the line endings to just
> LINE FEED (0A/^J) before submission, and I don't see ^M characters in
> either patch. I just opened the files in a binary editor to double-check
> and I don't see 0D anywhere.
>
> -----Original Message-----
> From: Devaraj k [mailto:devara...@huawei.com]
> Sent: Tuesday, April 03, 2012 9:51 AM
> To: common-dev@hadoop.apache.org
> Subject: RE: Help a newbie on patch submission
>
> I tried applying your patches in my environment, It works fine.
>
> The issue with Hadoop QA may be because of having <ctl>-M (^M) characters
> in the patch file. Can you try by removing the  <ctl>-M (^M) chars in the
> patch.
>
> Thanks
> Devaraj
> ________________________________________
> From: Mostafa Elhemali [mosta...@microsoft.com]
> Sent: Tuesday, April 03, 2012 9:42 PM
> To: common-dev@hadoop.apache.org
> Subject: Help a newbie on patch submission
>
> Hi,
> I'm trying to submit a patch to fix some build problems in Hadoop trunk on
> Windows, and I'm having a hard time pleasing Hadoop QA. This is for
> Hadoop-8241 (https://issues.apache.org/jira/browse/HADOOP-8241). I
> submitted the first patch and it said it couldn't apply it, so I figured it
> may have to do with the fact that one of the two files fixed is in
> hadoop-hdfs-https which isn't part of the Hadoop Common project. So I
> submitted a second patch just for hadoop-dist/pom.xml which I believe is
> part of the Common project, but that couldn't be applied as well.
> So I come to you: what am I doing horribly wrong and how can I appease the
> almighty Hadoop QA?
>
> Thanks,
> Mostafa Elhemali
>
>
>
>
>
>
>

Reply via email to