Mailing-List: contact hbase-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hbase-dev@hadoop.apache.org
Message-ID: <778451834.1206378685130.JavaMail.jira@brutus>
Date: Mon, 24 Mar 2008 10:11:25 -0700 (PDT)
From: "stack (JIRA)" <jira@apache.org>
To: hbase-dev@hadoop.apache.org
Subject: [jira] Commented: (HBASE-536) Remove MiniDFS startup from
 MiniHBaseCluster
In-Reply-To: <1232147798.1206058284624.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HBASE-536?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12581607#action_12581607 ] 

stack commented on HBASE-536:
-----------------------------

What is happening here:

{code}
@@ -38,5 +38,8 @@
   public void testMergeMeta() throws IOException {
     assertNotNull(dfsCluster);
     HMerge.merge(conf, dfsCluster.getFileSystem(), HConstants.META_TABLE_NAME);
+    try {
+      Thread.sleep(10000);
+    } catch (Exception e) {}
{code}

Why the wait without an assert that awatied condition actually occurred?

Convention is that method names start with lower case letters: +  protected void HBaseClusterSetup() throws Exception {

+1 on replacing MiniHBaseCluster(1) w/ version 2.

Remove below

{code}
+  // /** {@inheritDoc} */
+  // @Override
+  // public void setUp() throws Exception {
+  //   super.setUp();
+  // }
{code}

The '\n's in the log message are silly: +    LOG.debug("\n\n\n\n\t\t\tSetup Complete\n\n\n\n");  If we start to allow them, I see a degeneration adding bolding, blinking lights, etc., to make today's log message favorite the loudest

I'm going to take a look at PE exception now.

> Remove MiniDFS startup from MiniHBaseCluster
> --------------------------------------------
>
>                 Key: HBASE-536
>                 URL: https://issues.apache.org/jira/browse/HBASE-536
>             Project: Hadoop HBase
>          Issue Type: New Feature
>          Components: test
>            Reporter: Bryan Duxbury
>            Assignee: Bryan Duxbury
>            Priority: Minor
>             Fix For: 0.2.0
>
>         Attachments: 536.patch
>
>
> Change MiniHBaseCluster to always require a MiniDfs started elsewhere. It will decide where to read and write based on the config passed in to its constructor. 
> HBaseClusterTestCase will be updated to do the MiniDFS spinup and shutdown as part of its normal process.
> Completing this issue will prime us for solving the truly external DFS issue for making the test suite faster. 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.