| Author |
Message
|
| SANTYP |
 Posted: Tue May 03, 2011 5:58 pm Post subject: No of instances problem |
 |
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
Hi Gurus,
I have a major problem in one of my flows, which has message frequency high as a result I made it run 5 times parllaly(I changed no of instances to 4), but if one of the thread haulted due to some error... remaing threads also not picking up the messages..
this flow mainly to extract data from the database.. and prepare files.. files sizes are varys b/w 100k to 4mb or some times more...
flow has input node ... java compute node -- and output node..
any help on the topic highly appreciated.. |
|
| Back to top |
|
 |
| SANTYP |
| Posted: Tue May 03, 2011 10:38 pm Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| any help on the issue.. please.. |
|
| Back to top |
|
|
| smdavies99 |
| Posted: Tue May 03, 2011 10:44 pm Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003
Posts: 6076
Location: Somewhere over the Rainbow this side of Never-never land.
|
Please have a little patience.
your original post was added at 1:58AM GMT. For most of the people who post here it is the middle of the night and they are (hopefully) fast asleep. I'm sure some of the IBMer's who are on the same TZ as myself will be along in a while.
_________________
WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
| Back to top |
|
|
| mqjeff |
| Posted: Wed May 04, 2011 3:49 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
It seems most likely that the problem is in your javacompute node.
In general, different instances of a message flow do not contend with each other unless you have taken specific steps to make them wait for a given resource. |
|
| Back to top |
|
|
| khudania |
| Posted: Wed May 04, 2011 4:12 am Post subject: |
|
|
Apprentice
Joined: 30 Nov 2004
Posts: 43
|
what is the current JVM heap size ?
If there are java compute nodes with intensive java processing, an increase in the jvm heap size should be considered specially when there are multiple instances.
_________________
If the doors of perception were cleansed, everything would appear as it is - infinite |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed May 04, 2011 4:23 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| smdavies99 wrote: |
| Please have a little patience. |
Remember this is a volunteer board with no SLA on responses. Remember also that some of us will see your repeated pleas while half asleep, inadequately caffinated and kinda grumpy.
| smdavies99 wrote: |
| I'm sure some of the IBMer's who are on the same TZ as myself will be along in a while. |
And if you need them along earlier, raise a PMR. You don't need a crashed broker before you call.
Even in this state I'm not going to attempt to answer a Java question except to echo the comment of my associate. Threads are not linked unless you link them so prime suspect has to be anything shared between the JCNs in the various instances (shared structures and so forth).
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| joebuckeye |
| Posted: Wed May 04, 2011 4:29 am Post subject: |
|
|
Partisan
Joined: 24 Aug 2007
Posts: 365
Location: Columbus, OH
|
| khudania wrote: |
what is the current JVM heap size ?
If there are java compute nodes with intensive java processing, an increase in the jvm heap size should be considered specially when there are multiple instances. |
Now granted the OP doesn't mention what kind of error is being encountered, just 'some error', but unless the error is an out of memory error I would think that just increasing the JVM heap size is more of a prayer to see if it works than any sort of realistic solution to the issue. And even if it is a memory issue increasing the JVM heap size may simply delay the problem and not fix it.
I would say the board can't really help the OP unless/until they find out what the original error is. |
|
| Back to top |
|
|
| SANTYP |
| Posted: Wed May 04, 2011 5:14 pm Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| khudania wrote: |
what is the current JVM heap size ?
If there are java compute nodes with intensive java processing, an increase in the jvm heap size should be considered specially when there are multiple instances. |
I given enough heap size for that execution group..
min heap size is 256 mb and made max to 1gb |
|
| Back to top |
|
|
| SANTYP |
| Posted: Wed May 04, 2011 6:12 pm Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| mqjeff wrote: |
It seems most likely that the problem is in your javacompute node.
In general, different instances of a message flow do not contend with each other unless you have taken specific steps to make them wait for a given resource. |
I didn't made any such arrangement.. its happening only when any hault occurs.. remaing normal cases its running parallely.. |
|
| Back to top |
|
|
| khudania |
| Posted: Thu May 05, 2011 2:09 am Post subject: |
|
|
Apprentice
Joined: 30 Nov 2004
Posts: 43
|
| mqjeff wrote: |
It seems most likely that the problem is in your javacompute node.
In general, different instances of a message flow do not contend with each other unless you have taken specific steps to make them wait for a given resource. |
I agree, 
Have you taken care of the exception case scenario [ 'hault' as per your description]. What do you when the hault case happens. How do you handle that ? Do you throw it ?
I would have considered using different execution groups because they all run within separate jvm of their own.
_________________
If the doors of perception were cleansed, everything would appear as it is - infinite |
|
| Back to top |
|
|
| Vitor |
| Posted: Thu May 05, 2011 4:20 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| SANTYP wrote: |
| mqjeff wrote: |
It seems most likely that the problem is in your javacompute node.
In general, different instances of a message flow do not contend with each other unless you have taken specific steps to make them wait for a given resource. |
I didn't made any such arrangement.. its happening only when any hault occurs.. remaing normal cases its running parallely.. |
Then it sounds like you need a PMR
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
