You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by Tom White <to...@cloudera.com> on 2010/04/09 20:10:16 UTC
Plans for a 0.21 Hadoop Release
I wanted to send an update on my plans for doing a 0.21 Hadoop Release.
Next Friday (16 April) is the feature freeze date for the new 0.21
branch. On that date I will create new 0.21 branches for Common, HDFS,
and MapReduce (I will rename the old branches so history is not lost),
and update the CHANGES.txt files and JIRA to reflect the new branch
contents. If there are issues that are not blockers that you wish to
see in the 0.21 release then you should endeavour to get these
committed in the next week.
After the branches have been created the next step will be to close
the blockers before cutting a release candidate. I hope it will be
possible to do this in a matter of weeks. The current blockers for
0.21 are: http://bit.ly/9UaOTe (Common), http://bit.ly/crOG7I (HDFS),
http://bit.ly/cCS7zw (MapReduce); although this list will change as
issues that are not true blockers get removed, and issues that crop up
during build and testing are added.
Please let me know if you have any questions or concerns about this plan.
Cheers,
Tom
Re: Plans for a 0.21 Hadoop Release
Posted by Allen Wittenauer <aw...@linkedin.com>.
On Apr 12, 2010, at 4:28 PM, Tom White wrote:
> On Mon, Apr 12, 2010 at 4:14 PM, Allen Wittenauer
> <aw...@linkedin.com> wrote:
>> So, does this mean that everything that has been committed for 0.22 is actually going to be in 0.21? For example, symlinks were committed to trunk for 0.22 according to JIRA.
>
> Yes. Note that symlinks to date only work with the FileContext API, so
> they are not yet integrated with the shell, or MapReduce for example.
>
>> Hmm...... because of JIRA, would it be easier to actually call this release 0.22?
>
> That's what I meant by updating JIRA - bulk changing everything fixed
> in 0.22 before the freeze date to be marked as being fixed in 0.21.
OK, thanks for clearing that up. :)
*goes back to working on some things he wants to get in before freeze*
Re: Plans for a 0.21 Hadoop Release
Posted by Tom White <to...@cloudera.com>.
On Mon, Apr 12, 2010 at 4:14 PM, Allen Wittenauer
<aw...@linkedin.com> wrote:
> So, does this mean that everything that has been committed for 0.22 is actually going to be in 0.21? For example, symlinks were committed to trunk for 0.22 according to JIRA.
Yes. Note that symlinks to date only work with the FileContext API, so
they are not yet integrated with the shell, or MapReduce for example.
> Hmm...... because of JIRA, would it be easier to actually call this release 0.22?
That's what I meant by updating JIRA - bulk changing everything fixed
in 0.22 before the freeze date to be marked as being fixed in 0.21.
Cheers,
Tom
RE: Plans for a 0.21 Hadoop Release
Posted by Zl...@barclayscapital.com.
> So, does this mean that everything that has been committed for 0.22 is actually going to be in 0.21? For example, symlinks were committed >to trunk for 0.22 according to JIRA.
+1 for symlinks in next stable release and the associated refactoring of isFile, isDirectory, etc.
_______________________________________________
This e-mail may contain information that is confidential, privileged or otherwise protected from disclosure. If you are not an intended recipient of this e-mail, do not duplicate or redistribute it by any means. Please delete it and any attachments and notify the sender that you have received it in error. Unless specifically indicated, this e-mail is not an offer to buy or sell or a solicitation to buy or sell any securities, investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Barclays. Any views or opinions presented are solely those of the author and do not necessarily represent those of Barclays. This e-mail is subject to terms available at the following link: www.barcap.com/emaildisclaimer. By messaging with Barclays you consent to the foregoing. Barclays Capital is the investment banking division of Barclays Bank PLC, a company registered in England (number 1026167) with its registered office at 1 Churchill Place, London, E14 5HP. This email may relate to or be sent from other members of the Barclays Group.
_______________________________________________
Re: Plans for a 0.21 Hadoop Release
Posted by Allen Wittenauer <aw...@linkedin.com>.
So, does this mean that everything that has been committed for 0.22 is actually going to be in 0.21? For example, symlinks were committed to trunk for 0.22 according to JIRA.
Hmm...... because of JIRA, would it be easier to actually call this release 0.22?
Re: Plans for a 0.21 Hadoop Release
Posted by Tom White <to...@cloudera.com>.
Hi Owen,
Thanks for your offer to review the work I've been doing on
compatibility. If pushing back the freeze date will help us stabilize
the public API, then I'm OK moving it to 30 April.
On the serialization changes, I'm +1 to backing out the changes since
0.20 rather than trying to get new changes in for 0.21.
Cheers,
Tom
On Sun, Apr 11, 2010 at 12:07 AM, Owen O'Malley <ow...@gmail.com> wrote:
> On Fri, Apr 9, 2010 at 5:16 PM, Doug Cutting <cu...@apache.org> wrote:
>
>> Is this just HADOOP-6685? Or does it also include MAPREDUCE-1183?
>>
>
> I just meant h-6685.
>
> My instinct would rather be to revert the serialization API changes made
>> since 0.20 (mostly HADOOP-6165) if that's not an API we intend to support,
>> and not add new APIs at the last minute.
>>
>
> I have the patch almost ready, but it is a fair point that reverting the
> previous change is lower risk. I will file an issue. I'm still asking for
> the two week extension to look at the other APIs, especially since I'll
> largely be offline this week.
>
> -- Owen
>
Re: Plans for a 0.21 Hadoop Release
Posted by Owen O'Malley <ow...@gmail.com>.
On Fri, Apr 9, 2010 at 5:16 PM, Doug Cutting <cu...@apache.org> wrote:
> Is this just HADOOP-6685? Or does it also include MAPREDUCE-1183?
>
I just meant h-6685.
My instinct would rather be to revert the serialization API changes made
> since 0.20 (mostly HADOOP-6165) if that's not an API we intend to support,
> and not add new APIs at the last minute.
>
I have the patch almost ready, but it is a fair point that reverting the
previous change is lower risk. I will file an issue. I'm still asking for
the two week extension to look at the other APIs, especially since I'll
largely be offline this week.
-- Owen
Re: Plans for a 0.21 Hadoop Release
Posted by Doug Cutting <cu...@apache.org>.
Owen O'Malley wrote:
> and hopefully to get the new generic serialization api in before the
> branch.
Is this just HADOOP-6685? Or does it also include MAPREDUCE-1183?
My instinct would rather be to revert the serialization API changes made
since 0.20 (mostly HADOOP-6165) if that's not an API we intend to
support, and not add new APIs at the last minute. Tom's proposed that
the serialization API is "evolving", so we can technically change it
going forward, but we should still be cautious about promoting new APIs
without a clear consensus, and serialization APIs have proven
controversial in the recent past. It might be also better to get some
experience with a new API in trunk before we include it in a release.
Doug
Re: Plans for a 0.21 Hadoop Release
Posted by Owen O'Malley <om...@apache.org>.
On Apr 9, 2010, at 11:10 AM, Tom White wrote:
> Next Friday (16 April) is the feature freeze date for the new 0.21
> branch.
I've been buried by security work and would really love to have some
more time to review API annotations, look at the jdiffs for the apis,
and hopefully to get the new generic serialization api in before the
branch. Can I ask for a feature freeze data of 30 April?
Thanks,
Owen