[
https://issues.apache.org/jira/browse/CASSANDRA-1842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jonathan Ellis updated CASSANDRA-1842:
--------------------------------------
Attachment: 1842.txt
This is a bug where the WordCount example incorrectly wraps a byte array that
is subsequently re-used w/o copying. Fix attached.
> ColumnFamilyOutputFormat only writes the first column
> -----------------------------------------------------
>
> Key: CASSANDRA-1842
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1842
> Project: Cassandra
> Issue Type: Bug
> Components: Hadoop
> Affects Versions: 0.7.0 rc 1
> Reporter: Brandon Williams
> Assignee: Jeremy Hanna
> Fix For: 0.7.0
>
> Attachments: 1842.txt
>
>
> In CASSANDRA-1774 we fixed a problem where only the last column was being
> written. However, it appears that we only write the first one now. This is
> easy to observe in the word count example:
> {noformat}
> RowKey: text2
> => (column=word1, value=1, timestamp=1291666461685000)
> {noformat}
> is what the output for text2 looks like, but there should be another column,
> word2. If the word count is run without CFOF it works correctly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
