hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Venkatesh Seetharam" <vseetha...@gmail.com>
Subject mapred.system.dir setting
Date Tue, 27 Feb 2007 00:30:13 GMT
Hello there,

Howdy. I've seen in the past that mapred.system.dir needs to be a directory
shared across all the slaves else I get a "No such file or directory"
exception. I'm adding few more slaves to the hadoop cluster and I cannot
have a shared directory across these nodes.

Is there any way that I could set this one to get the system working.

I used to get the following exception and none of the maps on the slaves
work even though I have the directory created on each of the slaves. Only
tasks on Master is successful.

Error initializing task_0001_m_000000_0:
/sv/lucene/tmp/hadoop-svdev/mapred/system/submit_3iroqt/job.xml: No such
file or directory
at org.apache.hadoop.fs.FileUtil.copy(FileUtil.java:110)
at org.apache.hadoop.fs.LocalFileSystem.copyToLocalFile(LocalFileSystem.java
at org.apache.hadoop.fs.FileSystem.copyToLocalFile(FileSystem.java:869)
at org.apache.hadoop.mapred.TaskTracker.localizeJob(TaskTracker.java:331)
at org.apache.hadoop.mapred.TaskTracker.startNewTask(TaskTracker.java:828)
at org.apache.hadoop.mapred.TaskTracker.offerService(TaskTracker.java:511)
at org.apache.hadoop.mapred.TaskTracker.run(TaskTracker.java:857)
at org.apache.hadoop.mapred.TaskTracker.main(TaskTracker.java:1499)

Any help is greatly appreciated.

ENV: Hadoop 0.11.2 on JDK 1.6.0


"Perfection (in design) is achieved not when there is nothing more to add,
but rather when there is nothing more to take away."
- Antoine de Saint-Exupéry

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message