| Author |
Message
|
| elsa_ro |
 Posted: Tue Jun 06, 2006 2:42 am Post subject: browse a remote queue |
 |
|
Newbie
Joined: 06 Jun 2006
Posts: 5
|
Hi,
I have a problem with an application that is trying to browse a remote queue and I get an error like "Reconnected to AtoB in browse mode*
*Option not valid for object type*".If I try to put the messages in the remote queue with amqsput, the message is send to the local AtoB on other queue manager.The application is trying to see if there are any messages in the queue(remote queue) and cannot do it.What can I do to make the application browse the remote queue?
Thanks |
|
| Back to top |
|
 |
| wschutz |
| Posted: Tue Jun 06, 2006 2:47 am Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
You can't get or browse messages from remote queues.
Either move the application to the qmgr where the queue is LOCAL, ur use the MQ client to connect to that queue manager.
_________________
-wayne |
|
| Back to top |
|
|
| elsa_ro |
| Posted: Tue Jun 06, 2006 3:58 am Post subject: browse a remote queue |
|
|
Newbie
Joined: 06 Jun 2006
Posts: 5
|
Thanks for the quick response.
There no other trick ?Like the remote queue to put messages in the local queue on a remote queue manager and also on a local queue on the local queue manager?Is it possible? |
|
| Back to top |
|
|
| Vitor |
| Posted: Tue Jun 06, 2006 4:11 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
Well, your application could do 2 puts - one to the remote queue and one to the local. This would of course result in 2 utterly unconnected messages which would then need to be managed in some way as to stay in step.
Complex, tricky, error prone and highly unlikely to be a good idea 
I think a better line of attack is to tell us what you're trying to achieve. Why is the putting application trying to browse this remote queue? What information is it trying to obtain / purpose is it trying to achieve? It's more normal (I use the term guardedly as "normal" is subjective) for the sending and receiving applications to be somewhat uncoupled. Even in a request/reply type model.
Talk us through the design.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| elsa_ro |
| Posted: Tue Jun 06, 2006 4:31 am Post subject: browse a remote queue |
|
|
Newbie
Joined: 06 Jun 2006
Posts: 5
|
The application is trying to browse the queue in order to see if the message is still in the queue.If the message is not in the queue and the application did not get any reply that the message was received,than it will try to resend the message.
Can the remote queue and the local queue have the same name on the same cluster? |
|
| Back to top |
|
|
| sandiksk |
| Posted: Tue Jun 06, 2006 4:40 am Post subject: |
|
|
Centurion
Joined: 08 Jun 2005
Posts: 133
|
Elsa_ro wrote
| Code: |
The application is trying to browse the queue in order to see if the message is still in the queue.If the message is not in the queue and the application did not get any reply that the message was received,than it will try to resend the message.
Can the remote queue and the local queue have the same name on the same cluster? |
You should be checking for messages on the remote Queue managers local queue not on the remote queue of local queue manager. Why would you want to have two diferent type queues with same name. As far as i now its not possible. even if it is, its of no use. |
|
| Back to top |
|
|
| Vitor |
| Posted: Tue Jun 06, 2006 4:45 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
They have the same name, but because they're on 2 queue managers they're not the same queue so that doesn't get you any further.
I think what you're better off looking into is the report confirm_on_delivery options, which will (unsurprisingly) cause a report message to be set when the message is read from the target queue. Possibly also the positive/negative action options. These should give you the sort of information it sounds like you're looking for.
Clearly if the reply to the original message has been held up by some kind of network issue then the report messages will get held in the same way! 
How does the design handle that? Putting application sends message, gets no reply, "browses" queue & determines it's been read off, resends original then the reply to the first messages turns up looking sheepish and muttering about TCP/IP problems?
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
