[
https://issues.apache.org/jira/browse/HADOOP-10783?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14052416#comment-14052416
]
Steve Loughran commented on HADOOP-10783:
-----------------------------------------
presumably this is a reference to apache commons-lang.
2.6 is the last "legacy" release; there is now a commons-lang 3 release. This
does look more modern, but an upgrade seems traumatic as [it requires every
import to be
changed|http://commons.apache.org/proper/commons-lang/article3_0.html].
There's benefits in doing this from the project's improvements, but there's
100+ imports that would need changing. That's a significant enough change that
it would have to go into trunk and then be backported to branch-2 when people
were happy. We'd also have to think about retaining the -2.6 jar on our
classpath for the entirety of the hadoop 2.x lifetime to avoid breaking
downstream projects that depend on it.
If you can supply a patch for trunk we to do this we can review it
(As you are clearly trying to get Hadoop working on BSD, you should switch to
trunk because that is where fixes will go in. Consider also writing an
Uber-JIRA "get Hadoop working on BSD" to which you can make these sub tasks or
dependencies. Please also put "FreeBSD" in the environment field")
> apache-commons-lang.jar shipped with hadoop does not support FreeBSD
> --------------------------------------------------------------------
>
> Key: HADOOP-10783
> URL: https://issues.apache.org/jira/browse/HADOOP-10783
> Project: Hadoop Common
> Issue Type: Bug
> Affects Versions: 2.4.1
> Reporter: Dmitry Sivachenko
> Assignee: Steve Loughran
>
> Hadoop-2.4.1 ships with apache-commons.jar version 2.6.
> It does not support FreeBSD (IS_OS_UNIX returns False).
> This is fixed in recent versions of apache-commons.jar
> Please update apache-commons.jar to recent version so it correctly recognizes
> FreeBSD as UNIX-like system.
> Right now I get in datanode's log:
> 2014-07-04 11:58:10,459 DEBUG
> org.apache.hadoop.hdfs.server.datanode.ShortCircui
> tRegistry: Disabling ShortCircuitRegistry
> java.io.IOException: The OS is not UNIX.
> at
> org.apache.hadoop.io.nativeio.SharedFileDescriptorFactory.create(SharedFileDescriptorFactory.java:77)
> at
> org.apache.hadoop.hdfs.server.datanode.ShortCircuitRegistry.<init>(ShortCircuitRegistry.java:169)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.initDataXceiver(DataNode.java:583)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.startDataNode(DataNode.java:771)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.<init>(DataNode.java:289)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.makeInstance(DataNode.java:1931)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.instantiateDataNode(DataNode.java:1818)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.createDataNode(DataNode.java:1865)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.secureMain(DataNode.java:2041)
> at
> org.apache.hadoop.hdfs.server.datanode.DataNode.main(DataNode.java:2065)
--
This message was sent by Atlassian JIRA
(v6.2#6252)
