ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Broker on Mainframe

Post new topic  Reply to topic
 Broker on Mainframe « View previous topic :: View next topic » 
Author Message
gabbar
PostPosted: Wed Sep 03, 2003 6:52 am    Post subject: Broker on Mainframe Reply with quote

Acolyte

Joined: 10 Dec 2002
Posts: 50

Hi,

We have a test broker installed on 390. It was working fine until recently. But for the past few days, the broker is dying after processing a few messages. Everytime we restart the broker, it's processing fewer messages and going down. It now cannot even process 5 messages.

The only error that we consistently see in the trace log is BIP2235W. We think the broker is short of resources that it is using, which may be the log file size or maybe something else.

Did anyone face similar problems with broker on 390 or for that matter on any platform.

Thanks in advance.
Back to top
View user's profile Send private message Visit poster's website
Craig B
PostPosted: Wed Sep 03, 2003 8:22 am    Post subject: Reply with quote

Partisan

Joined: 18 Jun 2003
Posts: 316
Location: UK

Hi,

The BIP2235W message is indicating that the broker could not read its LIL directory and therefore would not be able to load any user plugins or user parsers that flows may call. You cannot be using a plugins in your flows in yor broker because at deploy time the flows would attempt to find the registered parser classes and/or Node Factorys for these user nodes, and your deploy would fail. On this basis the BIP2235W is not related to the crashing of the broker.

It is possible that your broker is running short of resources, but after a clean start, and only processing 5 messages you would not expect it to have a resource shortage after that time? What size input messages are these flows taking in? Do you have anything special in your flows such as looping etc?

The broker does not write any logs of its own (like an MQSeries queue manager) and the main use of resources for writing out would be if tracing had been enabled on the execution group or message flows. This would cause a large amount of writing to the filesystem, and if this is full, this could cause your broker to crash.

However, if it is abending, you should be seeing some kind of errors in the system log or local log which would indicate a reason why it was terminating. Do you have any idea of what is being processed at the time when it actually terminates? Do you have any database interaction in these flows?

Could you look for any errors in the logs and/or give a fuller description of what you are doing in your flows. If the DataFlowEngine address spaces are terminating you should get some information in the log that at least gives an abend code and a module and offset that it was in. Could you post these details on this thread. Thanks
_________________
Regards
Craig
Back to top
View user's profile Send private message
Tibor
PostPosted: Tue Feb 08, 2005 2:02 am    Post subject: Reply with quote

Grand Master

Joined: 20 May 2001
Posts: 1033
Location: Hungary

Just a hint for BIP2235W, independently by platform:

Broker's LilPath (<mqsidata>/registry/<broker>/LilPath) won't work when a linefeed is on the end of line: [url=http://www-1.ibm.com/support/docview.wss?rs=977&context=SSHQUD&context=SSMKUK&context=SS2GTQ&context=SSKMAB&context=SS9H3T&context=SS433P&context=SSKM8N&context=SS3GH2&context=SSBJDG&context=SSBJCK&context=SSSRR3&context=SS2KS6&context=SSCL2Z&context=SSEK2R&context=SSSRD5&context=SSSRKD&context=SSFKTZ&context=SSBTEG&context=SSFKUX&context=SSVLA5&context=SSGTPH&q1=bip2235*&uid=swg21175142&loc=en_US&cs=utf-8&lang=en]Look at the IBM support site[/url]

We have fixed a similar error right now

Tibor
Back to top
View user's profile Send private message
Andi.MQ
PostPosted: Fri Feb 11, 2005 7:36 am    Post subject: Reply with quote

Novice

Joined: 10 Dec 2003
Posts: 11
Location: Switzerland

Hi gabbar
We have some Brokers running on z/OS in Version 5.1. We have problems with
memory allocation in some executiongroups. That's a known problem and the only way to solve this problem is to split the flows on additional executiongroups.

I'm sure that you must get some errormessages. Please try to post your MSGLOG output or anything.

Regards,
Andi.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  Reply to topic Page 1 of 1

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Broker on Mainframe
Jump to:  



You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
Protected by Anti-Spam ACP
 
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.