hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stack <st...@duboce.net>
Subject VOTE: Merge new Assignment Manager AMv2), HBASE-14614
Date Tue, 30 May 2017 16:09:46 GMT
Moving on from the DISCUSSION thread [4], lets vote on whether to merge the
new Assignment Manager (AMv2) out on branch HBASE-14614 to master.

Unit tests pass. Tests at scale show that the new AM is at least as good as
the old.

AMv2 is a long-standing effort addressing the most common source of trouble
operating an hbase cluster at scale, region assignment.

AMv2 is the last hurdle to our being able to branch for hbase2.

Please vote +1 if in favor of merge and -1 if agin.

See below if you need more background on what we're voting on.


The new AssignmentManager on HBASE-14614 branch promises new levels of
scaling, insight, resilience and assignment performance. A skeleton dev
overview is available here [1].

There are outstanding items to complete but they can come in post-merge.
Critical is reenabling a set of tests disabled because they depend on
workings no longer present or they depended on a broken behavior in AMv1
(Disabled tests and the list of TODOs are in [1]).

Currently I am working on polish making it pass the outstanding failing
unit tests, fixing white space, and findbugs but it is ready for commit.
I've put up an RB patch that is absent protobufs so the bulk is less
intimidating [3].

This patch is mostly the work of Matteo Bertozzi (inception, bulk of
implementation). Others have contributed including Stephen Yuan Jiang.
Umesh Agashe and Appy also have contributed fixes that got added to the
HBASE-14614 branch.

1. https://docs.google.com/document/d/1eVKa7FHdeoJ1-
2. https://docs.google.com/document/d/1QLXlVERKt5EMbx_
3. https://reviews.apache.org/r/55117/

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