Mailing-List: contact common-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 22 Nov 2017 17:29:02 +0000 (UTC)
From: "Steve Loughran (JIRA)" <jira@apache.org>
To: common-issues@hadoop.apache.org
Message-ID: <JIRA.13017346.1478110604000.301630.1511371742088@Atlassian.JIRA>
In-Reply-To: <JIRA.13017346.1478110604000@Atlassian.JIRA>
References: <JIRA.13017346.1478110604000@Atlassian.JIRA> <JIRA.13017346.1478110604662@jira-lw-us.apache.org>
Subject: [jira] [Updated] (HADOOP-13786) Add S3A committer for zero-rename
 commits to S3 endpoints
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
archived-at: Wed, 22 Nov 2017 17:29:07 -0000


     [ https://issues.apache.org/jira/browse/HADOOP-13786?page=3Dcom.atlass=
ian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Steve Loughran updated HADOOP-13786:
------------------------------------
       Resolution: Fixed
    Fix Version/s: 3.1.0
           Status: Resolved  (was: Patch Available)

This now committed! Thank you all for your support, insight, testing, revie=
ws!=20

Special mention of : Sanjay Radia, Ryan Blue, Ewan Higgs, Mingliang Liu and=
 extra especially Aaron Fabbri!

Not only does this patch add the committer, it adds (configurable) retry po=
licy to every single call s3a makes of the AWS s3 SDK, with the inconsisten=
t s3 client now configurable to simulate throttling events. Everyone gets t=
o see how their code handles the presence of transient throttle failures.

Finally, I now know more about Hadoop & Spark commit protocols than I never=
 knew I needed to, as well as all those nuances of S3 which matter for that=
. I'll have to make more use of that knowledge, somehow.

> Add S3A committer for zero-rename commits to S3 endpoints
> ---------------------------------------------------------
>
>                 Key: HADOOP-13786
>                 URL: https://issues.apache.org/jira/browse/HADOOP-13786
>             Project: Hadoop Common
>          Issue Type: New Feature
>          Components: fs/s3
>    Affects Versions: 3.0.0-beta1
>            Reporter: Steve Loughran
>            Assignee: Steve Loughran
>             Fix For: 3.1.0
>
>         Attachments: HADOOP-13786-036.patch, HADOOP-13786-037.patch, HADO=
OP-13786-038.patch, HADOOP-13786-039.patch, HADOOP-13786-HADOOP-13345-001.p=
atch, HADOOP-13786-HADOOP-13345-002.patch, HADOOP-13786-HADOOP-13345-003.pa=
tch, HADOOP-13786-HADOOP-13345-004.patch, HADOOP-13786-HADOOP-13345-005.pat=
ch, HADOOP-13786-HADOOP-13345-006.patch, HADOOP-13786-HADOOP-13345-006.patc=
h, HADOOP-13786-HADOOP-13345-007.patch, HADOOP-13786-HADOOP-13345-009.patch=
, HADOOP-13786-HADOOP-13345-010.patch, HADOOP-13786-HADOOP-13345-011.patch,=
 HADOOP-13786-HADOOP-13345-012.patch, HADOOP-13786-HADOOP-13345-013.patch, =
HADOOP-13786-HADOOP-13345-015.patch, HADOOP-13786-HADOOP-13345-016.patch, H=
ADOOP-13786-HADOOP-13345-017.patch, HADOOP-13786-HADOOP-13345-018.patch, HA=
DOOP-13786-HADOOP-13345-019.patch, HADOOP-13786-HADOOP-13345-020.patch, HAD=
OOP-13786-HADOOP-13345-021.patch, HADOOP-13786-HADOOP-13345-022.patch, HADO=
OP-13786-HADOOP-13345-023.patch, HADOOP-13786-HADOOP-13345-024.patch, HADOO=
P-13786-HADOOP-13345-025.patch, HADOOP-13786-HADOOP-13345-026.patch, HADOOP=
-13786-HADOOP-13345-027.patch, HADOOP-13786-HADOOP-13345-028.patch, HADOOP-=
13786-HADOOP-13345-028.patch, HADOOP-13786-HADOOP-13345-029.patch, HADOOP-1=
3786-HADOOP-13345-030.patch, HADOOP-13786-HADOOP-13345-031.patch, HADOOP-13=
786-HADOOP-13345-032.patch, HADOOP-13786-HADOOP-13345-033.patch, HADOOP-137=
86-HADOOP-13345-035.patch, MAPREDUCE-6823-003.patch, cloud-intergration-tes=
t-failure.log, objectstore.pdf, s3committer-master.zip
>
>
> A goal of this code is "support O(1) commits to S3 repositories in the pr=
esence of failures". Implement it, including whatever is needed to demonstr=
ate the correctness of the algorithm. (that is, assuming that s3guard provi=
des a consistent view of the presence/absence of blobs, show that we can co=
mmit directly).
> I consider ourselves free to expose the blobstore-ness of the s3 output s=
treams (ie. not visible until the close()), if we need to use that to allow=
 us to abort commit operations.


--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org