You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Amareshwari Sriramadasu (JIRA)" <ji...@apache.org> on 2009/06/24 09:35:07 UTC

[jira] Commented: (HADOOP-4975) CompositeRecordReader: ClassLoader set in JobConf is not passed onto WrappedRecordReaders

    [ https://issues.apache.org/jira/browse/HADOOP-4975?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12723481#action_12723481 ] 

Amareshwari Sriramadasu commented on HADOOP-4975:
-------------------------------------------------

bq. I made the change to the Parser class, but I think the question is still open about whether the change should go into the Configuration(Configuration) constructor.
Yes. I figured out that the solution is to move the changes to Configuration(Configuration), when I'm trying to change join package to use new api. Raised HADOOP-6103 for the Configuration changes.

> CompositeRecordReader: ClassLoader set in JobConf is not passed onto WrappedRecordReaders
> -----------------------------------------------------------------------------------------
>
>                 Key: HADOOP-4975
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4975
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.19.0
>            Reporter: Jingkei Ly
>            Assignee: Jingkei Ly
>             Fix For: 0.21.0
>
>         Attachments: break-wrapped-rr-test.patch, HADOOP-4975-1.patch, HADOOP-4975-2.patch
>
>
> I am using a custom ClassLoader which I set in my JobConf via setClassLoader(). The ClassLoader is loaded key and value classes which are required to read records from SequenceFiles that were written out in a previous MapReduce job. 
> However, I am getting a ClassNotFoundException when using the CompositeInputFormat to create a RecordReader to read these SequenceFiles from HDFS. It occurs when the SequenceFile.Reader tries to create an instance of the Key/Value classes, presumably because the class loader SequenceFile.Reader is using is not the one I set with JobConf.setClassLoader. Below is an example of the stack trace I get:
> {code}
> Caused by: java.io.IOException: WritableName can't load class
> 	at org.apache.hadoop.io.WritableName.getClass(WritableName.java:73)
> 	at org.apache.hadoop.io.SequenceFile$Reader.getKeyClass(SequenceFile.java:1596)
> 	... 33 more
> Caused by: java.lang.ClassNotFoundException: org.apache.hadoop.MyWritableClass
> 	at java.net.URLClassLoader$1.run(URLClassLoader.java:200)
> 	at java.security.AccessController.doPrivileged(Native Method)
> 	at java.net.URLClassLoader.findClass(URLClassLoader.java:188)
> 	at java.lang.ClassLoader.loadClass(ClassLoader.java:307)
> 	at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
> 	at java.lang.ClassLoader.loadClass(ClassLoader.java:252)
> 	at java.lang.ClassLoader.loadClassInternal(ClassLoader.java:320)
> 	at java.lang.Class.forName0(Native Method)
> 	at java.lang.Class.forName(Class.java:247)
> 	at org.apache.hadoop.conf.Configuration.getClassByName(Configuration.java:673)
> 	at org.apache.hadoop.io.WritableName.getClass(WritableName.java:71)
> 	... 34 more
> {code}
> I'll attach a unit test that can demonstrate this more clearly....

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