0

We are using WebSphere MQ and mmx , however we are facing issues while trying to connect with queue:

[2/10/12 13:24:51:861 CST] 00000011 SystemOut     O 13:24:51,861 INFO  [ListenerThread]  - Retry [=1] reconnecting to JMS Queue/Topic
[2/10/12 13:24:51:864 CST] 00000012 SystemOut     O 13:24:51,864 INFO  [ListenerThread]  - Retry [=1] reconnecting to JMS Queue/Topic
[2/10/12 13:24:51:874 CST] 00000012 SystemOut     O 13:24:51,874 INFO  [JMSListener]  - init() failed with JMSException during initializing JMS access: xxsvclnk.queue.ISEEOutboundQueue
[2/10/12 13:24:51:875 CST] 00000011 SystemOut     O 13:24:51,875 INFO  [JMSListener]  - init() failed with JMSException during initializing JMS access: xxreqctr.queue.ISEEInboundQueue

Please let us know possible causes for this issue . we have done all the relevant changes (host name:port ) etc.

JoshMc
  • 10,239
  • 2
  • 19
  • 38
xslguy
  • 91
  • 1
  • 11
  • You haven't provided us much to go on. Can you print the linked exceptions? If it's an MQ issue, the linked exceptions will contain completion and reason codes that will pinpoint exactly the cause. Also need to know what version of WMQ client and what version of WMQ server. You also mention doing "relevant changes" which implies this worked before? What has changed? – T.Rob Feb 10 '12 at 20:40
  • Thanks for reply ,We have a shared infrastructure environment , in this environment unfortunately we don't have control on many things . I don't see any error code/stack trace apart from this . We are using websphere 7 as web server. Any clues. – xslguy Feb 10 '12 at 21:18

1 Answers1

0

Per the update in the comments, the apps don't print linked exceptions and you do not have access to fix them. JMS exceptions are a multi-level data structure. The linked exceptions exist to hold vendor-specific diagnostic codes. If your apps do not print the entire JMS exception then the apps should be reported to the vendor or programmers as containing sev-1 defects. There is no possible justification for a JMS application to not print all levels of a multi-level diagnostic data structure. This post is a brilliant example of exactly why such practices are unjustified. Without a stack trace and linked exceptions, there is no data with which to diagnose production problems. The week (if that) saved on delivery of the code will be paid for many times over with an extended outage.

Your only other option here is to do tracing. Exactly what options are available depends on the versions of WMQ client and WMQ server that are installed. You do have enough access and/or support to find that out, yes?

T.Rob
  • 31,522
  • 9
  • 59
  • 103