hadoop-hdfs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Wangda Tan <wheele...@gmail.com>
Subject Re: About reset branch-3.1 to trunk before release.
Date Mon, 19 Mar 2018 18:37:46 GMT
Done JIRA fix version update:

Moved all JIRAs with fixVersion = 3.2.0 to 3.1.0 except following few fixes
(which committed after 49c747ab187d0650143205ba57ca19607ec4c6bd)

    YARN-8002. Support NOT_SELF and ALL namespace types for allocation tag.
(Weiwe
i Yang via wangda)

    HADOOP-15262. AliyunOSS: move files under a directory in parallel when
rename a directory. Contributed by Jinhu Wu.

    MAPREDUCE-7066. TestQueue fails on Java9

    YARN-8028. Support authorizeUserAccessToQueue in RMWebServices.
Contributed by Wangda Tan.

    YARN-8040. [UI2] New YARN UI webapp does not respect current pathname
for REST api. Contributed by Sunil G.

Thanks,
Wangda

On Mon, Mar 19, 2018 at 11:12 AM, Wangda Tan <wheeleast@gmail.com> wrote:

> Thanks Akira for the additional vote,
>
> With help from Apache Infra Team (Daniel Takamori), we just reset
> branch-3.1 to trunk (SHA: 49c747ab187d0650143205ba57ca19607ec4c6bd). Will
> update JIRA fix version shortly.
>
> - Wangda
>
> On Sun, Mar 18, 2018 at 6:10 PM, Akira Ajisaka <ajisakaa@oss.nttdata.co.jp
> > wrote:
>
>> +1 for resetting branch-3.1.
>>
>> Thanks,
>> Akira
>>
>>
>> On 2018/03/18 12:51, Wangda Tan wrote:
>>
>>> Thanks for sharing your thoughts.
>>>
>>> We have done build and single node cluster deploy / test for the latest
>>> trunk code (commit: 49c747ab187d0650143205ba57ca19607ec4c6bd). Since
>>> there
>>> are no objections, so I will go ahead to do the branch replace.
>>>
>>> Since we don't have force push permission to release branches. I just
>>> filed
>>> https://issues.apache.org/jira/browse/INFRA-16204 to get help from
>>> Apache
>>> infra team.
>>>
>>> Please hold any commits to branch-3.1, will keep this email thread
>>> posted.
>>>
>>> Best,
>>> Wangda
>>>
>>> On Wed, Mar 14, 2018 at 3:14 PM, Vinod Kumar Vavilapalli <
>>> vinodkv@apache.org
>>>
>>>> wrote:
>>>>
>>>
>>> I see one new feature: https://issues.apache.org/jira/browse/YARN-7626:
>>>> Allow regular expression matching in container-executor.cfg for devices
>>>> and
>>>> named docker volumes mount.
>>>>
>>>> There are 21 sub-tasks. There are three feature-type JIRAs in those -
>>>> https://issues.apache.org/jira/browse/YARN-7972,
>>>> https://issues.apache.org/jira/browse/YARN-7891 and
>>>> https://issues.apache.org/jira/browse/YARN-5015. These should be okay -
>>>> not major disrupting features.
>>>>
>>>> Everything else is either a bug-fix or an improvement so we should be
>>>> good.
>>>>
>>>>  From the list, it doesn't look like resetting will destabilize 3.1, +1
>>>> for
>>>> doing this.
>>>>
>>>> Thanks
>>>> +Vinod
>>>>
>>>> On Mar 14, 2018, at 1:54 PM, Wangda Tan <wheeleast@gmail.com> wrote:
>>>>>
>>>>> Hi mapreduce/yarn/common/hdfs-devs,
>>>>>
>>>>> As of now, we have all blockers done for 3.1.0 release [1]. The release
>>>>>
>>>> is running behind schedule due to a few security-related issues.
>>>> Because of
>>>> this and since branch-3.1 is cut 5 weeks before on Feb 8, trunk 3.2 is
>>>> already diverging. There're 64 commits in trunk but not in branch-3.1.
>>>> [2]
>>>>
>>>>>
>>>>> I took a quick scan of them, most of them are good fixes which we
>>>>> should
>>>>>
>>>> bring to 3.1.0 as well. And this can also reduce differences between
>>>> 3.2.0
>>>> and 3.1.0 release for less maintenance burden in the future.
>>>>
>>>>>
>>>>> Unless anyone objects, we will reset branch-3.1 to trunk in 1-2 days
>>>>> and
>>>>>
>>>> cut RC after that.
>>>>
>>>>>
>>>>> Thoughts?
>>>>>
>>>>> - Wangda
>>>>>
>>>>> [1] project in (YARN, HADOOP, MAPREDUCE, HDFS) AND priority in
>>>>> (Blocker,
>>>>>
>>>> Critical) AND resolution = Unresolved AND "Target Version/s" = 3.1.0
>>>> ORDER
>>>> BY priority DESC
>>>>
>>>>> [2] project in (YARN, HADOOP, MAPREDUCE, HDFS) AND fixVersion in
>>>>> (3.2.0)
>>>>>
>>>> AND fixVersion not in (3.1.0)
>>>>
>>>>
>>>>
>>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message