accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mike Drob (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-2126) Hadoop profile names should be consistent across branches
Date Fri, 03 Jan 2014 23:01:51 GMT


Mike Drob commented on ACCUMULO-2126:

Some thoughts on the patches a as whole:
# Does the version of ZK we depend on not change between Hadoop 1 and 2? It seemed odd that
the version of slf4j did change but nothing else.
# Can we attempt to detect and provide some helpful output when a user attempts to build with
the old commands? Currently, {{mvn package -Dhadoop.profile=2.2}} results in a bunch of arcane
looking compilation errors.
# nit: References to Hadoop 1.0.x and 2.0.x should probably become "Hadoop 1.x" and "Hadoop
2.x" in the multitude of {{pom.xml}} files

> Hadoop profile names should be consistent across branches
> ---------------------------------------------------------
>                 Key: ACCUMULO-2126
>                 URL:
>             Project: Accumulo
>          Issue Type: Improvement
>          Components: build
>            Reporter: Sean Busbey
>            Assignee: Sean Busbey
>            Priority: Minor
>             Fix For: 1.4.5, 1.5.1, 1.6.0
>         Attachments: ACCUMULO-2126.1-1.5.1-SNAPSHOT.patch.txt, ACCUMULO-2126.1-1.6.0-SNAPSHOT.patch.txt,
> ACCUMULO-1419 changed the hadoop pofile names to reflect the major components used. This
makes documenting how to build against a major hadoop line across all of our branches more
difficult than necessary.
> Since the changes in what we need to do vary with the Hadoop line, e.g. Hadoop 1 and
Hadoop 2, we should name them appropriately.
> Additionally, our current activation scheme for hte default doesn't allow activating
it by name.

This message was sent by Atlassian JIRA

View raw message