[
https://issues.apache.org/jira/browse/HDFS-2293?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13092457#comment-13092457
]
Uma Maheswara Rao G commented on HDFS-2293:
-------------------------------------------
Todd,
sorry, you are right. looks it is working on trunk. I tested with
Minidfscluster. Because of this when it is trying to getFileSystem, it will
return file system from cache. So, it is taking defaults for me. In command
line that will not be the case. Every time fs will be initialized.
Just i cleaned all fs objects in cache and executed commandline option & tried
to set block size to 120.
then it displays the below message:
put: io.bytes.per.checksum(512) and blockSize(120) do not match. blockSize
should be a multiple of io.bytes.per.checksum
So, it is proven that, block size is getting effected.
I will post the tests for this issue for more coverage.
thanks
Uma
> Add the ability to set dfs.block.size for a given file from the command line
> ----------------------------------------------------------------------------
>
> Key: HDFS-2293
> URL: https://issues.apache.org/jira/browse/HDFS-2293
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs client
> Affects Versions: 0.23.0
> Reporter: Aaron T. Myers
> Assignee: Uma Maheswara Rao G
> Labels: newbie
>
> See the thread
> [here|http://www.mail-archive.com/[email protected]/msg01459.html],
> in which Allen correctly points out that since dfs.block.size is a
> client-side config, one could just set a different value in a custom
> HADOOP_CONF_DIR. We might as well make that easy.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
