Project split postponed until 6/1

[Owen O'Malley <om...@apache.org>]

We weren't able to resolve the build issues on the project split and  
so need to move it out another week to 6/1. You can track the progress  
by looking at the HADOOP-4687 branch.

-- Owen

Re: Hudson Build Broken

["Tsz Wo (Nicholas), Sze" <s2...@yahoo.com>]

I will commit it.  Thanks, Gary.

Nicholas Sze




----- Original Message ----
> From: Gary Murry <gm...@yahoo-inc.com>
> To: "core-dev@hadoop.apache.org" <co...@hadoop.apache.org>
> Sent: Tuesday, June 2, 2009 2:01:53 PM
> Subject: Hudson Build Broken
> 
> Hi, 
> 
> It looks like hadoop_trunk has been broken for a bit because of a failed unit 
> test .  It looks like HADOOP-5956 
> https://issues.apache.org/jira/browse/HADOOP-5956 resolves the problem.  Or at 
> least it resolved it on my local build.  Can we see about getting this committed 
> so that the build system can get going again?
> 
> Thanks,
> -Gary Murry

Hudson Build Broken

[Gary Murry <gm...@yahoo-inc.com>]

Hi, 

It looks like hadoop_trunk has been broken for a bit because of a failed unit test .  It looks like HADOOP-5956 https://issues.apache.org/jira/browse/HADOOP-5956 resolves the problem.  Or at least it resolved it on my local build.  Can we see about getting this committed so that the build system can get going again?

Thanks,
-Gary Murry

Re: Project split postponed until 6/1

[Nigel Daley <nd...@yahoo-inc.com>]

On Jun 1, 2009, at 6:58 AM, Tom White wrote:

> There are currently 63 issues in the Hadoop patch queue at
> https://issues.apache.org/jira/secure/IssueNavigator.jspa?mode=hide&requestId=12311582 
> .
> Many of them have been sitting in the queue for a long time. I think
> we (committers) should aim to keep the queue short, by not leaving
> patches in the "Patch Available" state for more than a few days. This
> means that new patches need to be reviewed, then either committed
> (after passing Hudson tests, of course), or moved back to the "Open"
> state for feedback to be incorporated.

+1

> So I suggest that we take advantage of the project split to clear out
> the patch queue, and to keep it short in future. It might be a good
> idea to split the queue into three to correspond to the new projects
> (we should be able to use the component field to do this). What do
> people think?

It will naturally be split in 3 as issues move to HDFS, MAPREDUCE or  
stay in HADOOP Jira projects.

Cheers,
Nige


> On Tue, May 26, 2009 at 7:09 PM, Owen O'Malley <om...@apache.org>  
> wrote:
>> We weren't able to resolve the build issues on the project split  
>> and so need
>> to move it out another week to 6/1. You can track the progress by  
>> looking at
>> the HADOOP-4687 branch.
>>
>> -- Owen
>>

Re: Project split postponed until 6/1

[Owen O'Malley <om...@apache.org>]

On Jun 1, 2009, at 8:40 AM, Doug Cutting wrote:

> Tom White wrote:
>> I think
>> we (committers) should aim to keep the queue short, by not leaving
>> patches in the "Patch Available" state for more than a few days.
>
> +100

I agree too. We've let the burden rest on the shoulders of too few for  
too long. We as committers all need to be far better about responding  
to the patch queue.

> Won't we be splitting Jira too, into, e.g., MAPRED, HDFS and either  
> leaving the remaining in HADOOP or renaming it HCORE or somesuch?   
> Jira makes it easy to move issues between projects, and to link  
> between projects, so this should be a painless split.

Of course, I'll move the jiras around and create new subprojects.

-- Owen

Re: Project split postponed until 6/1

[Doug Cutting <cu...@apache.org>]

Tom White wrote:
> I think
> we (committers) should aim to keep the queue short, by not leaving
> patches in the "Patch Available" state for more than a few days.

+100

> So I suggest that we take advantage of the project split to clear out
> the patch queue, and to keep it short in future. It might be a good
> idea to split the queue into three to correspond to the new projects
> (we should be able to use the component field to do this). What do
> people think?

Won't we be splitting Jira too, into, e.g., MAPRED, HDFS and either 
leaving the remaining in HADOOP or renaming it HCORE or somesuch?  Jira 
makes it easy to move issues between projects, and to link between 
projects, so this should be a painless split.

Doug

Re: Project split postponed until 6/1

[Tom White <to...@cloudera.com>]

There are currently 63 issues in the Hadoop patch queue at
https://issues.apache.org/jira/secure/IssueNavigator.jspa?mode=hide&requestId=12311582.
Many of them have been sitting in the queue for a long time. I think
we (committers) should aim to keep the queue short, by not leaving
patches in the "Patch Available" state for more than a few days. This
means that new patches need to be reviewed, then either committed
(after passing Hudson tests, of course), or moved back to the "Open"
state for feedback to be incorporated.

So I suggest that we take advantage of the project split to clear out
the patch queue, and to keep it short in future. It might be a good
idea to split the queue into three to correspond to the new projects
(we should be able to use the component field to do this). What do
people think?

Cheers,
Tom


On Tue, May 26, 2009 at 7:09 PM, Owen O'Malley <om...@apache.org> wrote:
> We weren't able to resolve the build issues on the project split and so need
> to move it out another week to 6/1. You can track the progress by looking at
> the HADOOP-4687 branch.
>
> -- Owen
>