topic Re: parallel execution problem in Job Conductor and Hash problem in Talend Studio

parallel execution problem in Job Conductor and Hash problem

Anonymous — Tue, 17 Jul 2012 11:59:49 GMT

Hi,
I am a freshman here, I had two question for help.
1. In Job Conductor, I can't run job use an independent process. The same problem is in https://community.talend.com/t5/Design-and-Development/Error-with-component-tRunJob-java-lang-RuntimeException/td-p/111964 . In the last, vgalopin said Uncheck that option. Yes, it can work, but there comes problem 2. I want to know if there is some way to run job in Job Conductor use independent process.??
2. I use parallel execution in one job. And in it's subjob I use HashOutput and HashInput componets. I found When I use independent process it was ok. but uncheck independent process, HashInput output value is wrong or even Exception.
connecting to socket on port 3827
connected
Exception in component tHashInput_1
java.util.ConcurrentModificationException
at java.util.HashMap$HashIterator.nextEntry(Unknown Source)
at java.util.HashMap$KeyIterator.next(Unknown Source)
at org.talend.designer.components.hashfile.common.IteratorHashFile.next(IteratorHashFile.java:68)
at talend_preprocessor.hashoutputtest_0_1.HashOutputTest.tHashInput_1Process(HashOutputTest.java:1320)
at talend_preprocessor.hashoutputtest_0_1.HashOutputTest.tOracleInput_1Process(HashOutputTest.java:1011)
at talend_preprocessor.hashoutputtest_0_1.HashOutputTest$1.run(HashOutputTest.java:1663)
Exception in component tRunJob_1
java.lang.RuntimeException: Child job running failed
at talend_preprocessor.hashtest_main_0_1.HashTest_Main$1tRunJob_1Thread.run(HashTest_Main.java:658)
at routines.system.ThreadPoolWorker.runIt(TalendThreadPool.java:159)
at routines.system.ThreadPoolWorker.runWork(TalendThreadPool.java:150)
at routines.system.ThreadPoolWorker.access$0(TalendThreadPool.java:145)
at routines.system.ThreadPoolWorker$1.run(TalendThreadPool.java:122)
at java.lang.Thread.run(Unknown Source)
java.lang.InterruptedException: sleep interrupted
at java.lang.Thread.sleep(Native Method)
at talend_preprocessor.hashoutputtest_0_1.HashOutputTest.runJobInTOS(HashOutputTest.java:1694)
at talend_preprocessor.hashoutputtest_0_1.HashOutputTest.runJob(HashOutputTest.java:1521)
at talend_preprocessor.hashtest_main_0_1.HashTest_Main$1tRunJob_1Thread.run(HashTest_Main.java:627)
at routines.system.ThreadPoolWorker.runIt(TalendThreadPool.java:159)
at routines.system.ThreadPoolWorker.runWork(TalendThreadPool.java:150)
at routines.system.ThreadPoolWorker.access$0(TalendThreadPool.java:145)
at routines.system.ThreadPoolWorker$1.run(TalendThreadPool.java:122)
at java.lang.Thread.run(Unknown Source)
3334 milliseconds
disconnected
Job HashTest_Main ended at 12:56 17/07/2012.
As I in need to use parallel execution, hash component and Job Conductor. Anyone has the same problem??

Re: parallel execution problem in Job Conductor and Hash problem

Anonymous — Wed, 18 Jul 2012 04:11:44 GMT

Hi
Welcome to Talend Community!
In fact, this is a bug. See this issue on BugTracker.
http://jira.talendforge.org/browse/DOCT-251
Which version of TIS do you use?
Regards,
Pedro

Re: parallel execution problem in Job Conductor and Hash problem

Anonymous — Wed, 18 Jul 2012 08:38:32 GMT

5.0.2 talend enterprise data integration professional edition. I didn't use dynamic job but use independent process in parent job.
I find this http://jira.talendforge.org/browse/TDI-18825 and disable the cache in my jobserver. But now it comes dead lock problem...

Re: parallel execution problem in Job Conductor and Hash problem

Anonymous — Wed, 18 Jul 2012 08:40:31 GMT

Hi
"Use dynamic job" and "Use indepedent process" have so many in common.
And this is a well-known bug.
Sorry. Please report it on BugTracker.
Regards,
Pedro

Re: parallel execution problem in Job Conductor and Hash problem

Anonymous — Wed, 18 Jul 2012 11:45:40 GMT

Hi pedro
I find this http://jira.talendforge.org/browse/TDI-20039?focusedCommentId=229173#comment-229173
Is said this problem is ok on r81162-V4.2.5NB , r81164-V5.0.3NB and r81164-V5.1.0NB ?? Dose it means if I change an edith I can avoid this bug?? Which edith I can choose?? Thanks

Re: parallel execution problem in Job Conductor and Hash problem

Anonymous — Thu, 19 Jul 2012 02:27:13 GMT

Hi
Please try to update TIS to 5.1.0M2.
Hope it has been fixed.
Regards,
Pedro