Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Fri, 1 May 2015 23:19:06 +0000 (UTC)
From: "Xiaoyu Yao (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12826667.1430520964000.53828.1430522346374@Atlassian.JIRA>
In-Reply-To: <JIRA.12826667.1430520964000@Atlassian.JIRA>
References: <JIRA.12826667.1430520964000@Atlassian.JIRA>
 <JIRA.12826667.1430520964800@arcas>
Subject: [jira] [Updated] (HDFS-8309) Skip unit test using
 DataNodeTestUtils#injectDataDirFailure() on Windows
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HDFS-8309?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xiaoyu Yao updated HDFS-8309:
-----------------------------
    Attachment: HDFS-8309.00.patch

Attach a patch that ensure all tests that use DataNodeTestUtils.injectDataDirFailure() are skipped for Windows and update comments. 

> Skip unit test using DataNodeTestUtils#injectDataDirFailure() on Windows
> ------------------------------------------------------------------------
>
>                 Key: HDFS-8309
>                 URL: https://issues.apache.org/jira/browse/HDFS-8309
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>    Affects Versions: 2.7.0
>            Reporter: Xiaoyu Yao
>            Assignee: Xiaoyu Yao
>            Priority: Minor
>         Attachments: HDFS-8309.00.patch
>
>
> As [~cnauroth] noted  in HDFS-7917 below, DataNodeTestUtils.injectDataDirFailure() won't work for Windows as rename will fail due to open handles on data node dir. This ticket is opened to skip these tests for Windows. 
> bq.Unfortunately, I just remembered that the rename isn't going to work on Windows. It typically doesn't allow you to rename a directory where there are open file handles anywhere in the sub-tree. We'd have to shutdown the DataNode before doing the rename and then start it up. By doing that, we'd be changing the meaning of the test from covering an online failure to covering a failure at DataNode startup, so I don't think we can make that change.
> Below are the two test cases that need to be fixed:
> # TestDataNodeVolumeFailure#testFailedVolumeBeingRemovedFromDataNode
> # TestDataNodeHotSwapVolumes.testDirectlyReloadAfterCheckDiskError


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)