| Author |
Message
|
| elvis_gn |
 Posted: Mon Dec 12, 2005 3:22 am Post subject: [Solved] Stack full |
 |
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
Hi guys,
I have a flow which has an Input node. The node is linked to a message set which has a repeating complex type....
The complex type has a max occurs of 300.
When i run the flow (now in production), i find that when the number of occurances are around 150 the flow works, when its 170 and above the msg does not evn go till the next node.
I placed Trace nodes after each terminal and found that the message does not come anywhere.
When I placed a Throw node after the failure terminal i got a error in the Event log as
| Quote: |
( WBRKBRK.default ) Broker process terminating abnormally: The following diagnostic information will be required when contacting IBM: '
Severe Abend Error detected.
For full details see Abend File: C:/Program Files/IBM/WebSphere Business Integration Message Brokers/errors/WBRKBRK.default.3420.1648.Abend
A summary of the Error follows:
An Unhandled Exception detected in process 3420, on thread 0x670.
Type: EXCEPTION_STACK_OVERFLOW (C00000FD).
Address: 001B:2FE8155D.
Exception Parameters: 2
Param#0: 0x00000001
Param#1: 0x0e502fec
'.
A broker process is terminating abnormally.
Contact your IBM support center. |
and
| Quote: |
( WBRKBRK.default ) Broker process terminating abnormally (stack full): diagnostic information for IBM 'Severe Abend Error: A Stack overflow has occured, further logging may cause a double fault and will be attempted with caution. Abend Files and/or dumps may not be produced.
'.
A broker process is terminating abnormally because its stack is full. For execution group processes, this may be caused by message flows containing loops or more than 500 nodes in series. Otherwise the following information will be required when contacting IBM: 'Severe Abend Error: A Stack overflow has occured, further logging may cause a double fault and will be attempted with caution. Abend Files and/or dumps may not be produced.
'.
Check the definitions of all message flows deployed on the named execution group for loops and excessive path length. If this does not resolve the problem, contact your IBM support center. |
and I dont have 500 nodes in series or anything. It does not have any loops to. Well that anyway cant be the problem as i'm just at the beginning of the flow.
I tried increasing the stack size from 4mb to 8mb....now atleast the message is backed out into the Dead letter Queue....initially even that was not happening.
Any thoughts.
Regards.
Last edited by elvis_gn on Tue Dec 13, 2005 10:11 pm; edited 1 time in total |
|
| Back to top |
|
 |
| mgk |
| Posted: Mon Dec 12, 2005 4:45 am Post subject: |
|
|
Padawan
Joined: 31 Jul 2003
Posts: 1642
|
Hi.
The rest of the Abend file would be interesting for me to look at, but it looks like you need to raise a PMR.
What version anf fix pack the product are you using?
Regards,
_________________
MGK
The postings I make on this site are my own and don't necessarily represent IBM's positions, strategies or opinions. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Mon Dec 12, 2005 4:57 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
From the installation directory, it should be version 5 that elvis_gn is using.
It might be a better idea to install it into a shorter path than the default directory, though. (this is probably not the cause of the problem, but can cause problems).
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| elvis_gn |
| Posted: Mon Dec 12, 2005 6:23 am Post subject: |
|
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
Hi guys,
Thanks for the replies.
I doubt that the directory has anything to with it as the flow works for certain messages.
I would have anyway tried it, but the actual problem is found in the Production machine....and I think I will have to give some solid reasons as to why this can be a problem, and why I did not think of this before 
Can increasing the stack memory lead to a performance issue...
It is currently at 8 Mb.
Regards. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Mon Dec 12, 2005 6:26 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
There are issues with using the default installation directory, because path names can get very long and this can cause deployments to have issues - because path names got too long.
Like I said, I don't think it is part of the problem.
As MGK said, you should open a PMR. Also, you need to be more specific about the version and FixPack/CSD level you are using.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| elvis_gn |
| Posted: Mon Dec 12, 2005 6:31 am Post subject: |
|
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
Sorry, forgot about that
I'm using v5 and fix pack is 4.
Maybe i'll implement the shorter directory in my personal test environment and see.
Will keep u guys updated. Thanks.
Regards. |
|
| Back to top |
|
|
| elvis_gn |
| Posted: Tue Dec 13, 2005 10:10 pm Post subject: |
|
|
Padawan
Joined: 08 Oct 2004
Posts: 1905
Location: Dubai
|
Hey guys,
Got the solution.
The MQSI_THREAD_STACK_SIZE needed to be increased. It is currently at 120000 and I think I need to increase it more, as the messages are huge.
I could not find it in a Windows box, but on AIX, log in and open the .profile and edit the above parameter.
It works.
Thanks for the help.
Regards. |
|
| Back to top |
|
|
|
|
