You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Arun C Murthy (JIRA)" <ji...@apache.org> on 2007/10/13 11:21:50 UTC
[jira] Created: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
--------------------------------------------------------------
Key: HADOOP-2051
URL: https://issues.apache.org/jira/browse/HADOOP-2051
Project: Hadoop
Issue Type: Bug
Components: mapred
Affects Versions: 0.15.0
Reporter: Arun C Murthy
Assignee: Arun C Murthy
Priority: Blocker
Fix For: 0.15.0
The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12535223 ]
Hadoop QA commented on HADOOP-2051:
-----------------------------------
+1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12367808/HADOOP-2051_2_20071016.patch
against trunk revision r584939.
@author +1. The patch does not contain any @author tags.
javadoc +1. The javadoc tool did not generate any warning messages.
javac +1. The applied patch does not generate any new compiler warnings.
findbugs +1. The patch does not introduce any new Findbugs warnings.
core tests +1. The patch passed core unit tests.
contrib tests +1. The patch passed contrib unit tests.
Test results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/955/testReport/
Findbugs warnings: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/955/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Checkstyle results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/955/artifact/trunk/build/test/checkstyle-errors.html
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/955/console
This message is automatically generated.
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch, HADOOP-2051_2_20071016.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Devaraj Das (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Devaraj Das updated HADOOP-2051:
--------------------------------
Resolution: Fixed
Status: Resolved (was: Patch Available)
I just committed this. Thanks Arun!
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch, HADOOP-2051_2_20071016.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Hudson (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12535526 ]
Hudson commented on HADOOP-2051:
--------------------------------
Integrated in Hadoop-Nightly #275 (See [http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Nightly/275/])
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch, HADOOP-2051_2_20071016.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy updated HADOOP-2051:
----------------------------------
Status: Open (was: Patch Available)
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy updated HADOOP-2051:
----------------------------------
Status: Patch Available (was: Open)
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch, HADOOP-2051_2_20071016.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy updated HADOOP-2051:
----------------------------------
Attachment: HADOOP-2051_1_20071013.patch
Here is a patch which ignores all exceptions (after logging) in {{JobTracker#TaskCommitQueue#run}}, similar to other long-running threads in the {{JobTracker}}.
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy updated HADOOP-2051:
----------------------------------
Status: Patch Available (was: Open)
Submitting patch for review.
Thanks to Devaraj and Christian for their help on this one!
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Arun C Murthy updated HADOOP-2051:
----------------------------------
Attachment: HADOOP-2051_2_20071016.patch
Here is a new patch incorporating the received feedback... I've created HADOOP-2062 as a follow-up.
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch, HADOOP-2051_2_20071016.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12534538 ]
Hadoop QA commented on HADOOP-2051:
-----------------------------------
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12367681/HADOOP-2051_1_20071013.patch
against trunk revision r584336.
@author +1. The patch does not contain any @author tags.
javadoc +1. The javadoc tool did not generate any warning messages.
javac +1. The applied patch does not generate any new compiler warnings.
findbugs -1. The patch appears to introduce 1 new Findbugs warnings.
core tests +1. The patch passed core unit tests.
contrib tests +1. The patch passed contrib unit tests.
Test results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/943/testReport/
Findbugs warnings: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/943/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Checkstyle results: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/943/artifact/trunk/build/test/checkstyle-errors.html
Console output: http://lucene.zones.apache.org:8080/hudson/job/Hadoop-Patch/943/console
This message is automatically generated.
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Devaraj Das (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12534887 ]
Devaraj Das commented on HADOOP-2051:
-------------------------------------
Arun, could you please modify the patch to do "return" when InterruptedException is thrown by queue.take().
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Devaraj Das (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12534583 ]
Devaraj Das commented on HADOOP-2051:
-------------------------------------
Exception in thread "Task Commit Thread" java.lang.NoClassDefFoundError: org/apache/hadoop/mapred/JobHistory$HistoryCleaner
at org.apache.hadoop.mapred.JobHistory$JobInfo.logFinished(JobHistory.java:439)
at org.apache.hadoop.mapred.JobInProgress.isJobComplete(JobInProgress.java:877)
at org.apache.hadoop.mapred.JobInProgress.completedTask(JobInProgress.java:827)
at org.apache.hadoop.mapred.JobInProgress.updateTaskStatus(JobInProgress.java:400)
at org.apache.hadoop.mapred.JobTracker$TaskCommitQueue.run(JobTracker.java:1977)
In Christian's jobtracker.out file, I found this exception. This would make the thread go away.
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Doug Cutting (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12534960 ]
Doug Cutting commented on HADOOP-2051:
--------------------------------------
> FindBugs is complaining that I'm catching Exception, which is precisely the point of this jira.
Maybe you should catch Throwable instead?
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-2051) JobTracker's TaskCommitQueue is
vulnerable to non-IOExceptions
Posted by "Arun C Murthy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12534796 ]
Arun C Murthy commented on HADOOP-2051:
---------------------------------------
FindBugs is complaining that I'm catching Exception, which is precisely the point of this jira. *smile*
{noformat}
REC_CATCH_EXCEPTION: Exception is caught when Exception is not thrown
This method uses a try-catch block that catches Exception objects, but Exception is not thrown within the try block, and RuntimeException is not explicitly caught.
It is a common bug pattern to say try { ... } catch (Exception e) { something } as a shorthand for catching a number of types of exception each of whose catch blocks is identical, but this construct also accidentally catches RuntimeException as well, masking potential bugs.
{noformat}
> JobTracker's TaskCommitQueue is vulnerable to non-IOExceptions
> --------------------------------------------------------------
>
> Key: HADOOP-2051
> URL: https://issues.apache.org/jira/browse/HADOOP-2051
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Priority: Blocker
> Fix For: 0.15.0
>
> Attachments: HADOOP-2051_1_20071013.patch
>
>
> The {{JobTracker#TaskCommitQueue#run}} method only handles {{IOException}}s. Christian Kunz ran into a scenario where a job was stuck with all tasks in {{COMMIT_PENDING}} state and the stack traces showed that the "Task Commit Thread" wasn't even around.
> The work-around is to model {{TaskCommitQueue#run}} along the lines of other long-running threads in the {{JobTracer}} ({{ExpireLaunchingTasks}}, {{ExpireTrackers}} etc.) to catch, log and ignore any {{Exception}} in a loop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.