| Author |
Message
|
| mehdi_b_y |
 Posted: Mon Jun 12, 2006 12:09 am Post subject: java compute node problems |
 |
|
Novice
Joined: 11 Jun 2006
Posts: 12
|
Hi all,
I want to supervise step by step the execution of my message flow.
How can I do? Debugging? If yes, is there any tutorial for the MQSI v 6.0?
my flow contain a java compute node that makes some problems.
thanks. |
|
| Back to top |
|
 |
| XZ |
| Posted: Mon Jun 12, 2006 12:38 am Post subject: |
|
|
Apprentice
Joined: 22 May 2006
Posts: 45
Location: IBM Hursley
|
Yes, use flow debugger. You can find the relevent info in the "Debugging a message flow" topic in the on line doc. You should be able to debug into a Java ComputeNode (see the Debugging Java topic).
_________________
Regards,
-XZ
WMQI Development |
|
| Back to top |
|
|
| elvis_gn |
| Posted: Mon Jun 12, 2006 1:31 am Post subject: |
|
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
|
| Back to top |
|
|
| mehdi_b_y |
| Posted: Mon Jun 12, 2006 7:05 am Post subject: |
|
|
Novice
Joined: 11 Jun 2006
Posts: 12
|
thanks
my broblem now is how to deploy the message flow.
After adding an other java compute node that it need some jars files
my flow has not deployed into the execution group
an error has occurend :
BIP4502E: Une erreur inattendue est survenue lors du déploiement de la ressource Java.
Une erreur est survenue lors du déploiement de la ressource Java. Il est possible que cet incident ait laissé le courtier dans un état incohérent.
Assurez-vous que la machine qui exécute le courtier dispose de suffisamment d'espace disque disponible. Arrêtez le courtier, supprimez tous les fichiers dans '{0}', redémarrez le courtier et procédez à un redéploiement complet.
sorry I have a french version of the toolkit
but the error is about insufficient disk space
I need your advice.
thanks. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Mon Jun 12, 2006 7:22 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
Well the resolutions seems clear:
| Quote: |
| Assurez-vous que la machine qui exécute le courtier dispose de suffisamment d'espace disque disponible. Arrêtez le courtier, supprimez tous les fichiers dans '{0}', redémarrez le courtier et procédez à un redéploiement complet. |
Make sure that the broker machine has enough disk space. Stop the broker, delete all files in '{0}', restart the broker and do a FULL deployment.
Enjoy 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| elvis_gn |
| Posted: Mon Jun 12, 2006 7:48 pm Post subject: |
|
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
Hi,
fjb_saper
And where would the directory '{0}' be 
mehdi_b_y
I have got this error and some database container utilization threadhold reached errors during some Java development....u could check ur event log for anything more...Restarting the broker and configMgr fixes the issue until the next run of the flow....i guess if the code is not working properly, the memory capacity is being reached...
Regards. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Mon Jun 12, 2006 7:53 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
| elvis_gn wrote: |
Hi,
fjb_saper
And where would the directory '{0}' be
mehdi_b_y
I have got this error and some database container utilization threadhold reached errors during some Java development....u could check ur event log for anything more...Restarting the broker and configMgr fixes the issue until the next run of the flow....i guess if the code is not working properly, the memory capacity is being reached...
Regards. |
Argument 0 passed to the flow from outside ?
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| mehdi_b_y |
| Posted: Tue Jun 13, 2006 12:12 am Post subject: |
|
|
Novice
Joined: 11 Jun 2006
Posts: 12
|
Hi all,
I didn't find the path of the execution group
Please if you have any tutorial that explain how
to make reference to the externel jars files
whene deploying the message flow.
thanks. |
|
| Back to top |
|
|
| dirac |
| Posted: Tue Jun 13, 2006 12:25 am Post subject: |
|
|
Novice
Joined: 31 Mar 2005
Posts: 23
|
{0} should be the JAR path in this error message. For some reason it has not been filled in when throwing your error. If you're on Windows, {0} should be something like:
<workpath>/components/<broker name>/<broker uuid>/config/JAR
You should remove all the files in that JAR directory before attempting another deployment of the JavaComputeNode |
|
| Back to top |
|
|
| mehdi_b_y |
| Posted: Tue Jun 13, 2006 12:48 am Post subject: |
|
|
Novice
Joined: 11 Jun 2006
Posts: 12
|
hi dirac,
I didn't find this directory
the path of the broker is
c:\ibm\mqsi\6.0
I put the jars files in \classes and I add them in the CLASSPATH variable
but the flow is not deployed.
thanks. |
|
| Back to top |
|
|
| dirac |
| Posted: Tue Jun 13, 2006 1:17 am Post subject: |
|
|
Novice
Joined: 31 Mar 2005
Posts: 23
|
I am not sure what happens with external jars, but when deploying a JavaComputeNode flow, a jar is created in the directory I listed as part of the deployment process.
This directory has nothing to do with the directory where the product is installed. On my Windows system, the relevant directory is at:
C:/Documents and Settings/All Users/Application Data/IBM/MQSI/components/<my broker>/<broker uuid>/config/JAR
Once you do a deployment, a jar should appear in that directory as in <jarname>.jar
When you perform repeated delete/deployment of flows, other temporary files are created in that directory... can't remember offhand but it's something like <jarname>.old <jarname>.delete etc.
The error message you have got suggests that something has gone wrong with the commit/rollback process for the JARs when deploy/deleting your flows.
Before doing anything else, you need to delete all the files in that JAR subdirectory. Have a search around your Documents and Settings folder for it. |
|
| Back to top |
|
|
| JT |
| Posted: Tue Jun 13, 2006 7:14 am Post subject: |
|
|
Padawan
Joined: 27 Mar 2003
Posts: 1564
Location: Hartford, CT.
|
Perhaps ?
| Quote: |
The broker or Configuration Manager fails to start because there is not enough space in the Java TMPDIR directory- Scenario: The broker or Configuration Manager fails to start and you see an error message indicating that there is insufficient space.
- Explanation: The broker and Configuration Manager use Java JAR files. When the broker or Configuration Manager starts, the Java run time extracts the JAR files into a temporary directory, the Java TMPDIR directory. On Linux, UNIX and z/OS systems, the TMPDIR directory is usually /tmp; on Windows systems, it is c:\temp. If this directory is not large enough to hold the JAR files, the broker or Configuration Manager does not start.
- Solution: Use one of the following methods to specify the location of this temporary JAR directory:
o use the environment variable TMPDIR
o set the system property java.io.tmpdir
Allow at least 50 MB of space in this directory for broker components and 10 MB of space for Configuration Manager components. More space might be needed if you deploy large user-defined nodes or other JARs to the broker component. |
|
|
| Back to top |
|
|
|
|
