Chris,
Thank you very much. It does help, a lot. I had a feeling that it's
something like that.
I wish you all the best,
Marko
Quoting Chris Nauroth <[email protected]>:
Based on this stack trace, I'm guessing that you're running a 1.x version
of Hadoop.
The TaskTracker uses a set of local directories on the node for storage of
submitted job files during the task's execution. This is configured in
mapred-site.xml in the property named mapred.job.local.dir. The
DiskErrorException means that even after trying all directories configured
in mapped.job.local.dir, the TaskTracker couldn't find a place to store
the files. Possible root causes are misconfiguration, permissions on the
local directories blocking access, disks are full, or disks have failed
and gone into read-only mode.
I hope this helps.
--Chris Nauroth
On 5/21/15, 3:51 AM, "Marko Dinic" <[email protected]> wrote:
I'm new to Hadoop and I'm getting the following exception when I try to
run my job on Hadoop cluster:
org.apache.hadoop.util.DiskChecker$DiskErrorException: Could not find
any valid local directory for jobcache/job_201409031055_3865/jars/job.jar
at
org.apache.hadoop.fs.LocalDirAllocator$AllocatorPerContext.getLocalPathFor
Write(LocalDirAllocator.java:376)
at
org.apache.hadoop.fs.LocalDirAllocator.getLocalPathForWrite(LocalDirAlloca
tor.java:146)
at
org.apache.hadoop.fs.LocalDirAllocator.getLocalPathForWrite(LocalDirAlloca
tor.java:127)
at
org.apache.hadoop.mapred.JobLocalizer.localizeJobJarFile(JobLocalizer.java
:268)
at
org.apache.hadoop.mapred.JobLocalizer.localizeJobFiles(JobLocalizer.java:3
80)
at
org.apache.hadoop.mapred.JobLocalizer.localizeJobFiles(JobLocalizer.java:3
70)
at
org.apache.hadoop.mapred.DefaultTaskController.initializeJob(DefaultTaskCo
ntroller.java:232)
at org.apache.hadoop.mapred.TaskTracker$4.run(TaskTracker.java:1381)
at java.security.AccessController.doPrivileged(Native Method)
at javax.security.auth.Subject.doAs(Subject.java
Can anyone please tell me what seems to be the problem?
Best regards,
Marko
