JPPF Issue Tracker
Please log in to bookmark issues
CLOSED  Bug report JPPF-105  -  Temporary files disappear when setting a large MemoryMapDataProvider as data provider in a JPPFJob
Posted Dec 18, 2012 - updated Dec 27, 2014
icon_info.png This issue has been closed with status "Closed" and resolution "RESOLVED".
Issue details
  • Type of issue
    Bug report
  • Status
  • Assigned to
  • Progress
  • Type of bug
    Not triaged
  • Likelihood
    Not triaged
  • Effect
    Not triaged
  • Posted by
  • Owned by
    Not owned by anyone
  • Category
  • Resolution
  • Priority
  • Reproducability
    Not determined
  • Severity
  • Targetted for
    icon_milestones.png JPPF 3.2.x
Issue description
The problem arises when a large MemoryMapDataProvider (>2Gb. In my case) is set in a JPPFJob as data provider with long running task. Basically, the JPPF driver fails to find the JPPFJob associated file , as a result it does not assign the job to a node. Furthermore, the driver drops the connection with the node. This is the exception thrown by the JPPF driver:
2012-12-18 11:48:26,021 [DEBUG][org.jppf.server.protocol.ServerTaskBundleNode.taskCompleted(198)]: received exception for ServerTaskBundleNode[name=BinaryRelevanceJob 5 Counter: 0, uuid=null, initialTaskCount=6, taskCount=6, requeued=true, cancelled=false] : java.lang.Exception: /tmp/jppf5419575781347962408.tmp (No such file or directory)
  at org.jppf.server.nio.nodeserver.AbstractNodeContext.handleException(
 at java.util.concurrent.Executors$
  at java.util.concurrent.FutureTask$Sync.innerRun(
 at java.util.concurrent.ThreadPoolExecutor.runWorker(
  at java.util.concurrent.ThreadPoolExecutor$
Caused by: /tmp/jppf5419575781347962408.tmp (No such file or directory)
  at Method)
 at org.jppf.server.nio.PlainNioObject.write(
  at org.jppf.server.nio.AbstractNioMessage.writeNextObject(
  at org.jppf.server.nio.AbstractNioMessage.write(
  at org.jppf.server.nio.nodeserver.AbstractNodeContext.writeMessage(
  at org.jppf.server.nio.nodeserver.SendingBundleState.performTransition(
  at org.jppf.server.nio.nodeserver.SendingBundleState.performTransition(
 ... 6 more
Particularly, I am creating 10 JPPFJobs that uses this kind of data providers. This jobs' task might take to execute as long as 4 hours. This is important because when the task are short (<2 mins), the bug does not happen. The driver is running in a dedicated server with 8Gb. of RAM and 6-core AMD processor. The nodes are running in 6 nodes with a similar configuration. Operative system: Ubuntu 11.10 64 bits. Java Version: java version "1.6.0_22" OpenJDK Runtime Environment (IcedTea6 1.10.4) (6b22-1.10.4-0ubuntu1~11.04.2) OpenJDK 64-Bit Server VM (build 20.0-b11, mixed mode)

Steps to reproduce this issue
The following link contains a JUnit test case that can reproduce the issue as well as a patch. Patch+TestCase

Comment posted by
Dec 20, 07:36
Hi Juan Manuel,

It seems the file pointed to by the link is somehow corrupted. I'm able to extract the .tar from the gzip file, but I cannot extract the files in the .tar. Could you please compress the files again and attached the resulting archive to thsi bug report?


Comment posted by
Dec 20, 09:14
A file was uploaded. Reproducing JUnit test caseicon_open_new.png This comment was attached:

This test case reproduces the FileNotFoundException on the temporary file, before any fix is implemented.
Comment posted by
Dec 20, 10:13
Fix committed to SVN:

The issue was updated with the following change(s):
  • This issue has been closed
  • The status has been updated, from New to Closed.
  • This issue's progression has been updated to 100 percent completed.
  • The resolution has been updated, from Not determined to RESOLVED.
  • Information about the user working on this issue has been changed, from lolo4j to Not being worked on.