Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop >> mail # dev >> Logistics for releasing 2.4

Copy link to this message
Re: Logistics for releasing 2.4
Thanks for the list, Jason.  I scanned through them and except HDFS-5356, all jiras were currently targeted for 2.4.0. I changed the target version of HDFS-5356 to 2.4.0, so now all are targeted for 2.4.0.

Assuming "2.3.0" label is clean and doesn't need any mass/batch update, I think we can go over these jiras and re-target for 2.3.0, if one needs to be fixed in 2.3.0. 


On Friday, January 24, 2014 10:07 AM, Jason Lowe <[EMAIL PROTECTED]> wrote:
I just committed the addendum patch for HADOOP-9652 which should resolve
the performance issue.

Speaking of critical issues to fix for 2.3, I do wonder what to do with
all the Blocker/Criticals targeted for 2.4 (the old branch-2 HEAD
release) which apparently is now going to be 2.3 (the new branch-2 HEAD
release).  We should either get these resolved before the 2.3 release or
decide which ones aren't really blockers or that critical for 2.3 and
target them for a future release (e.g.: 2.4). According to my
calculations, there's currently 24 JIRAs that need to be scrubbed


On 01/23/2014 06:53 PM, Sangjin Lee wrote:
> I also wanted to bring your attention to HADOOP-9652. As this would address
> a major performance issue with the local filesystem, it would be good to
> get this in for 2.4 (or is it 2.3?).
> Regards,
> Sangjin
> On Thu, Jan 23, 2014 at 2:23 AM, Steve Loughran <[EMAIL PROTECTED]>wrote:
>> the JIRA I referenced is ready to go -it just needs review
>> On 22 January 2014 20:04, Andrew Wang <[EMAIL PROTECTED]> wrote:
>>> Thanks for the comments everyone.
>>> Vinod, if you think YARN-149 isn't ready yet, we can leave it out.
>>> Alternatively, we could release note it as "beta" with said known issues,
>>> and let people kick the tires. It looks like a bunch of the core
>>> functionality is already in place.
>>> Unless anyone else objects, I plan to cut a 2.4 branch later this week.
>>> Steve, Stack, if you think those two JIRAs can wrap up in that time frame
>>> we can include it. Since this more experiment will hopefully go well, we
>>> can include them in a 2.5 next month.
>>> Best,
>>> Andrew
>>> On Wed, Jan 22, 2014 at 11:28 AM, Stack <[EMAIL PROTECTED]> wrote:
>>>> Thanks Andrew for bringing this up.
>>>> +1 on more frequent releases and an effort at (roughly time-based
>>> release.
>>>> We are working to get 'HDFS-5776 Support 'hedged' reads in DFSClient'
>> to
>>>> land in time for 2.4 (but don't hold up the release for us!)
>>>> St.Ack
>>>> On Tue, Jan 21, 2014 at 3:51 PM, Vinod Kumar Vavilapalli <
>>>> [EMAIL PROTECTED]> wrote:
>>>>> If the timeline is to cut one next week, I don't think we can ship
>>>>> YARN-149 as part of that and call it stable. There are a bunch of
>> major
>>>>> things that are still missing there: YARN-1202, YARN-1410, YARN-1525
>>> and
>>>>> YARN-1611/YARN-1459.
>>>>> We need to start labeling individual features alpha/beta/stable now
>>> that
>>>>> we have a stable 2.2 base.
>>>>> Thanks
>>>>> +Vinod
>>>>> On Jan 21, 2014, at 1:26 PM, Andrew Wang <[EMAIL PROTECTED]>
>>>> wrote:
>>>>>> So, as per my earlier email, I think a 2.4 with just HDFS-4949,
>>>>> HDFS-2832,
>>>>>> and YARN-149 would be an attractive and stable release, and is
>>>> something
>>>>> we
>>>>>> could actually cut this week and vote on before the month is out.
>> The
>>>>> other
>>>>>> stuff we can ship in Feb or March when it's gotten a chance to bake
>>>> for a
>>>>>> bit, and culturally speaking, the fact that it's in 2.5 rather than
>>> 2.4
>>>>>> shouldn't be a big deal.
>>>>>> Thanks,
>>>>>> Andrew
>>>>>> On Tue, Jan 21, 2014 at 11