hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gang Li (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-9753) wrong offset for the first record in index file created by fix function in MapFile
Date Sun, 21 Jul 2013 21:42:48 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-9753?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Gang Li updated HADOOP-9753:
----------------------------

    Hadoop Flags:   (was: Incompatible change)
    
> wrong offset for the first record in index file created by fix function in MapFile 
> -----------------------------------------------------------------------------------
>
>                 Key: HADOOP-9753
>                 URL: https://issues.apache.org/jira/browse/HADOOP-9753
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: io
>    Affects Versions: 1.0.4
>         Environment: ubuntu 12.04
> jdk 7.0
> hadoop-1.0.4
>            Reporter: Gang Li
>              Labels: easyfix
>   Original Estimate: 1h
>  Remaining Estimate: 1h
>
> I want to convert a sequence file into a map file. So i used the fix function provided
by MapFile to create the index file. Also i set the indexInterval as 1 since i want to index
each record in the data file. However, i found that the position of the first record in the
data file was incorrectly indexed by the index file. And it turns out that the fix function
sets the position of the first record as 0. This is wrong because the reader of the data sequence
file has already read the head and the position should be the length of the head. So, i tried
to set the position as the correct value and it now works.

--
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