| Author |
Message
|
| ivanachukapawn |
 Posted: Mon Oct 13, 2008 7:45 am Post subject: distributed queueing and open input count |
 |
|
Knight
Joined: 27 Oct 2003
Posts: 561
|
| If an application connected to a remote queue manager opens up a queue (defined with a remote queue definition) for output, does the local queue exhibit an open-input-count > 0? |
|
| Back to top |
|
 |
| markt |
| Posted: Mon Oct 13, 2008 8:25 am Post subject: |
|
|
Knight
Joined: 14 May 2002
Posts: 508
|
| why don't you try it? can't take more than 30 seconds to test. |
|
| Back to top |
|
|
| ivanachukapawn |
| Posted: Mon Oct 13, 2008 8:29 am Post subject: |
|
|
Knight
Joined: 27 Oct 2003
Posts: 561
|
| I have MQ 7.0 Server on my laptop, no remote queue managers on the network, consequently no sender/receiver channels etc. so I can't test it in my current environment. I was hoping that some kind soul out there who does have the appropriate environment available would check this out for me. |
|
| Back to top |
|
|
| zpat |
| Posted: Mon Oct 13, 2008 9:00 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5866
Location: UK
|
MQ is Asynchronous and decoupled.
Opening a remote queue has no effect on the destination local queue. All you are doing is opening the xmit queue.
Putting a message will of course cause the channel that receives it to open the destination local queue for a time. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Mon Oct 13, 2008 9:04 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9472
Location: US: west coast, almost. Otherwise, enroute.
|
| Quote: |
| If an application connected to a remote queue manager opens up a queue (defined with a remote queue definition) for output, does the local queue exhibit an open-input-count > 0? |
First, an application (with the exception of the mq client) connects to a LOCAL qmgr - a qmgr in the same o/s instance. Once connected, by definition, all other qmgrs are REMOTE.
For an application that has opened a QRemote definition, it is the LOCAL xmit queue that the application puts messages to.
| Quote: |
| ...exhibit an open-input-count > 0? |
Yes.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| markt |
| Posted: Mon Oct 13, 2008 10:00 am Post subject: |
|
|
Knight
Joined: 14 May 2002
Posts: 508
|
| Quote: |
| no remote queue managers on the network |
Use 127.0.0.1 |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Mon Oct 13, 2008 10:05 am Post subject: Re: distributed queueing and open input count |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
| ivanachukapawn wrote: |
| If an application connected to a remote queue manager opens up a queue (defined with a remote queue definition) for output, does the local queue exhibit an open-input-count > 0? |
Guys you have to read a little bit more carefully.
The question sounds too much like a test question so I won't answer.
Just look at the play of words: open for OUTPUT show INPUT count....
This is all too obvious... 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| mqjeff |
| Posted: Mon Oct 13, 2008 10:19 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
ivanachuckapawn has generally shown a posting history that suggests the confusion between input and output is more likely just that - a confusion, rather than an attempt to ask a test question.
Also, that would be a relatively poor test question. |
|
| Back to top |
|
|
| ivanachukapawn |
| Posted: Mon Oct 13, 2008 10:31 am Post subject: |
|
|
Knight
Joined: 27 Oct 2003
Posts: 561
|
GM Saper,
You are wrong about your surmise that it is a test question. I am writing a spec for a program which must determine if remote users are actively using distributed queueing to put messages to a local queue. However it doesn't really matter because only one responder understood the question (the one who said that the channel would have the local queue open for input "for a time"). That is the answer I was looking for. For this program I will not be able to use open for input count to determine if the remote users are actively distributing queueing. Maybe I'll have to AMQSGEN queue statistics on the queue in question to detect any traffic.
I know it is not rational for me to resent your surmise that this was a "test" question, but I do.
You have been very helpful to me in the past and I hope that this will continue. Incidentally, I am IBM certified MQ 6.0 Admin despite the fact that I don't remember if by some miracle MQ would keep the local queue open for input as a function of the remote user keeping the xmit queue open for input.
I'm sorry for the confusion and hope that you harbor no bad feelings about this incident. |
|
| Back to top |
|
|
| mqjeff |
| Posted: Mon Oct 13, 2008 10:35 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
Specifically, the local queue is open as long as any incoming channel instance that has written to it is still running.
You can use DIS CONN(*) to see if channel instances have local queues open. |
|
| Back to top |
|
|
| ivanachukapawn |
| Posted: Mon Oct 13, 2008 10:38 am Post subject: |
|
|
Knight
Joined: 27 Oct 2003
Posts: 561
|
Thank you Jeff. Your suggestion is very helpful.
Ivanachukapawn |
|
| Back to top |
|
|
| kevinf2349 |
| Posted: Mon Oct 13, 2008 1:00 pm Post subject: |
|
|
Grand Master
Joined: 28 Feb 2003
Posts: 1311
Location: USA
|
Ivan
What is your intent? Is the intent to stop them from using the queue remotely? Or is it to monitor the remote usage of the queue?
I am a little unsure as to what it is you are trying to achieve or why you are trying to achieve it. |
|
| Back to top |
|
|
| ivanachukapawn |
| Posted: Mon Oct 13, 2008 1:32 pm Post subject: |
|
|
Knight
Joined: 27 Oct 2003
Posts: 561
|
| Its a DR situation, the host has moved and the remote sender channel hopefully had its IP switched via DNS resolution. I need to programatically detect the remote usage of the queue in a recovery scenario. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Mon Oct 13, 2008 1:42 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
| ivanachukapawn wrote: |
| Its a DR situation, the host has moved and the remote sender channel hopefully had its IP switched via DNS resolution. I need to programatically detect the remote usage of the queue in a recovery scenario. |
Can you please be a little more specific? I don't quite understand what you mean. I would have thought that the monitoring would tell you the information you seek here. Or is this to verify that the DR software has quicked in as it is supposed to??
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Oct 13, 2008 2:01 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
| ivanachukapawn wrote: |
| Its a DR situation, the host has moved and the remote sender channel hopefully had its IP switched via DNS resolution. I need to programatically detect the remote usage of the queue in a recovery scenario. |
Ivan (or is it Ivana),
This is a very indirect way of determining whether the DNS name has been repointed to the DR IP. It may indicate the repoint happened, but other things can have the q open, or the repoint could have happened but the RCVR channel no longer has the q open, for a variety of reasons. Lots of false negative and false positive possibilities.
By the way, when I first touched MQ clustering I too had the same questions about testing on one laptop. You can put multiple QMs on your laptop and have them each on a different listener port, and still test the vast majority of how and why MQ does what MQ does.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
