hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Matteo Bertozzi (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-7388) Snapshot branch 12/18 rebase broke TestSnapshotFromMaster#testSnapshotHFileArchiving
Date Fri, 21 Dec 2012 17:03:12 GMT

    [ https://issues.apache.org/jira/browse/HBASE-7388?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13538213#comment-13538213
] 

Matteo Bertozzi commented on HBASE-7388:
----------------------------------------

{code}
// load the table (creates 4 hfiles)
UTIL.loadTable(new HTable(UTIL.getConfiguration(), TABLE_NAME), TEST_FAM);
{code}
How can you be sure about that? there're no tuning parameters, there're no flush in the code...
What happens if we turn on by default compression or encoding? 
Maybe change or write a new loadTable() method, calling flush() to ensure the number of files
that you want?
                
> Snapshot branch 12/18 rebase broke TestSnapshotFromMaster#testSnapshotHFileArchiving
> ------------------------------------------------------------------------------------
>
>                 Key: HBASE-7388
>                 URL: https://issues.apache.org/jira/browse/HBASE-7388
>             Project: HBase
>          Issue Type: Sub-task
>          Components: snapshots, test
>    Affects Versions: hbase-6055
>            Reporter: Jonathan Hsieh
>            Assignee: Jonathan Hsieh
>             Fix For: hbase-6055
>
>         Attachments: hbase-7388.patch
>
>
> {code}
> Running org.apache.hadoop.hbase.master.cleaner.TestSnapshotFromMaster
> Tests run: 4, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 30.232 sec <<<
FAILURE!
> Results :
> Failed tests:   testSnapshotHFileArchiving(org.apache.hadoop.hbase.master.cleaner.TestSnapshotFromMaster):
Archived hfiles [] is missing snapshot file:hdfs://localhost:36177/user/jon/hbase/.snapshot/snapshot/8b7235fe93a521daee2a446915d87bfe/fam/01a951c716a445dea55698bdd79c294c
> {code}
> Something in trunk changed the number of hfiles generated such that only four are generated.
 The previous default was a min of 5 hfiles before compaction actually started so files were
never compacted, and thus never archived.  

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message