hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14563) Disable zombie TestHFileOutputFormat2
Date Tue, 06 Oct 2015 20:33:26 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14563?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14945723#comment-14945723

stack commented on HBASE-14563:

>From parent issue:
TestHFileOutputFormat2 failed 5 times in last 40 runs. I spent time on it yesterday. Seems
to be a reliance on test order but was having networking issues which complicated my being
able to do diagnosis.... It seems like an ambitious amount of work to get done in a unit test:
 * Simple test for {@link CellSortReducer} and {@link HFileOutputFormat2}.
 * Sets up and runs a mapreduce job that writes hfile output.
 * Creates a few inner classes to implement splits and an inputformat that
 * emits keys and values like those of {@link PerformanceEvaluation}.
Was added a good while ago, here:
commit e4f8a7419fb4bd0102eaf91e9747de6261e0b5c5
Author: jxiang <jxiang@unknown>
Date: Fri Feb 21 20:39:21 2014 +0000
HBASE-10526 Using Cell instead of KeyValue in HFileOutputFormat
git-svn-id: https://svn.apache.org/repos/asf/hbase/trunk@1570702 13f79535-47bb-0310-9956-ffa450edef68

The test stops and starts clusters a few times and then runs MR jobs. Needs shrinking in size
and scope. Needs to be more focused on testing a particular issue. 

HBASE-14564 is issue to reenable.

> Disable zombie TestHFileOutputFormat2
> -------------------------------------
>                 Key: HBASE-14563
>                 URL: https://issues.apache.org/jira/browse/HBASE-14563
>             Project: HBase
>          Issue Type: Sub-task
>          Components: test
>            Reporter: stack
>            Assignee: stack
> Disabling until someone has a chance to look at it.
> I watched it in jvisualvm a while. Its starting and stopping clusters multiple times
and then running mr jobs. Needs a rewrite at least and some shrinking of scope on what is

This message was sent by Atlassian JIRA

View raw message