[
https://issues.apache.org/jira/browse/HBASE-607?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12593483#action_12593483
]
stack commented on HBASE-607:
-----------------------------
The punctuation added to key doesn't belong in multiregiontable. It belongs in
a test of its own. But don't put this in the way of committing since the
punctuation was there previous.
You might make a method that does following:
{code}
+ region.close();
+ region.getLog().closeAndDelete();
+ return region;
{code}
... rather than repeat it multiple times.
I tried the patch on branch and it passed TTI and TTMR.
+1 on the branch patch.
In trunk patch, you do this:
{code}
- conf.set("fs.default.name", fs.getName());
+ // Set the hbase.rootdir to be the home directory in mini dfs.
{code}
Is that right?
Otherwise patch looks good to me.
> MultiRegionTable.makeMultiRegionTable is not deterministic enough for
> regression tests
> --------------------------------------------------------------------------------------
>
> Key: HBASE-607
> URL: https://issues.apache.org/jira/browse/HBASE-607
> Project: Hadoop HBase
> Issue Type: Test
> Components: test
> Affects Versions: 0.2.0, 0.1.2, 0.1.1
> Reporter: Jim Kellerman
> Assignee: Jim Kellerman
> Fix For: 0.2.0, 0.1.2
>
> Attachments: 607-0.1-V2.patch, 607-0.1.patch, 607-trunk.patch
>
>
> MultiRegionTable.makeMultiRegionTable is too non-deterministic for regression
> tests. It fails more often than the tests that use it.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
