| Author |
Message
|
| jefflowrey |
 Posted: Wed Feb 06, 2008 7:01 am Post subject: |
 |
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
Both the XMITQ and the QREMOTE are passive in all of this.
Everything is done by the Queue Manager... 
So it's the Queue Manager that reads the QREMOTE and uses that to create the XQH, which the Queue Manager adds to the message before PUTing it on the XMITQ.
Except it's a little different than that, because it's not the queue manager at all, it's the API code inside PUT...
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
 |
| Vitor |
| Posted: Wed Feb 06, 2008 7:13 am Post subject: Re: Lengthy ... but interesting channel config question. |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| friedl.otto wrote: |
I would say the remote queue resolution has
strapped some metadata to the message before dumping it in the
transmission queue.
|
The remote queue doesn't do anything, that was the perhaps pedantic point I was making. Queues are inert definitions not objects.
You're one of those people aren't you? Who wander over looking bemused, saying that "their queue doesn't seem to be running". Like queues are specific things inside the queue manager?
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| friedl.otto |
| Posted: Wed Feb 06, 2008 7:15 am Post subject: |
|
|
Centurion
Joined: 06 Jul 2007
Posts: 116
|
| jefflowrey wrote: |
| Except it's a little different than that, because it's not the queue manager at all, it's the API code inside PUT... |
Now that is truly scrumptious! So what you're saying is that when I
issue an MQ_PUT from my Java code, com.ibm.mq.jar snoops the
parameters of the QREMOTE then builds the XQH metadata and actually
does an MQ_PUT to the XMITQ, or am I mistaken?
_________________
Here's an idea - don't destroy semaphores unless you're certain of what you're doing!  -- Vitor |
|
| Back to top |
|
|
| friedl.otto |
| Posted: Wed Feb 06, 2008 7:22 am Post subject: Re: Lengthy ... but interesting channel config question. |
|
|
Centurion
Joined: 06 Jul 2007
Posts: 116
|
| Vitor wrote: |
| The remote queue doesn't do anything, that was the perhaps pedantic point I was making. |
QREMOTES and ALIASES we all know are merely 'symlinks' if you will.
| Vitor wrote: |
| Queues are inert definitions not objects. |
Then why is QLOCAL persisted to disk, if it is merely a figment of the
queue manager's imagination? 
| Vitor wrote: |
| You're one of those people aren't you? Who wander over looking bemused, saying that "their queue doesn't seem to be running". Like queues are specific things inside the queue manager? |
Why so acid? Even though our entire department believe that queues can be
running, and refuse to be corrected. Or are just rock-headedly refusing to
understand their messaging software ... I understand that the client API
and the queue manager are the only pieces of executable code.
_________________
Here's an idea - don't destroy semaphores unless you're certain of what you're doing! -- Vitor |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Wed Feb 06, 2008 7:24 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
| friedl.otto wrote: |
Now that is truly scrumptious! So what you're saying is that when I
issue an MQ_PUT from my Java code, com.ibm.mq.jar snoops the
parameters of the QREMOTE then builds the XQH metadata and actually
does an MQ_PUT to the XMITQ, or am I mistaken? |
No.
The Agent process does it, not the Java code you're using. But it does happen "inside" the PUT, and not separately.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| friedl.otto |
| Posted: Wed Feb 06, 2008 7:30 am Post subject: |
|
|
Centurion
Joined: 06 Jul 2007
Posts: 116
|
| jefflowrey wrote: |
| The Agent process does it, not the Java code you're using. But it does happen "inside" the PUT, and not separately. |
Since vitor already thinks I am the thickest organism this side
of Squornshellous Zeta, I may as well press on, and learn something.
"The Agent" ... please explain?
_________________
Here's an idea - don't destroy semaphores unless you're certain of what you're doing! -- Vitor
Last edited by friedl.otto on Wed Feb 06, 2008 7:34 am; edited 1 time in total |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Wed Feb 06, 2008 7:33 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
The Info Center, particularly the Intercommunications guide...
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| friedl.otto |
| Posted: Wed Feb 06, 2008 7:35 am Post subject: |
|
|
Centurion
Joined: 06 Jul 2007
Posts: 116
|
| jefflowrey wrote: |
The Info Center, particularly the Intercommunications guide...
|
I shall work my way through yon PDF. Thanks.
_________________
Here's an idea - don't destroy semaphores unless you're certain of what you're doing! -- Vitor |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Feb 06, 2008 7:36 am Post subject: Re: Lengthy ... but interesting channel config question. |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| friedl.otto wrote: |
| Vitor wrote: |
| Queues are inert definitions not objects. |
Then why is QLOCAL persisted to disk, if it is merely a figment of the
queue manager's imagination?
|
It's not a figment of the queue manager's imagination, nor is a piece of disk storage a running, executable process. QALIAS and QREMOTE objects are not figments of the queue manager's imagination either, and are likewise held on disc for reference.
| friedl.otto wrote: |
| Vitor wrote: |
| You're one of those people aren't you? Who wander over looking bemused, saying that "their queue doesn't seem to be running". Like queues are specific things inside the queue manager? |
Why so acid? |
Around 20 years of hearing the same stupid statements & questions will leave you both corroded and leaking the excess. I've had people who want me to shutdown the queue manager & just start their queue because the box was short of resource. Or claim they're getting 2033 errors because their queue's not running.
| friedl.otto wrote: |
I understand that the client API
and the queue manager are the only pieces of executable code. |
Then how can you say that:
| friedl.otto wrote: |
| a QREMOTE adds "XQH" metadata |
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Feb 06, 2008 7:41 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| friedl.otto wrote: |
Since vitor already thinks I am the thickest organism this side
of Squornshellous Zeta, I may as well press on, and learn something.
|
Are you kidding? In the thickness stakes you're strictly minor leagues!
I offer into evidence a project manager who wanted WMQ installed on a Solaris box; the key issue as I saw it was the lack of any disc mounted on which to create a file system. His comment of "well try the install anyway and see if we get lucky" puts him in my personal top 5 for 2008 to date.
Besides, you started off with a well-formed question. We don't see many of those. 
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| friedl.otto |
| Posted: Wed Feb 06, 2008 7:57 am Post subject: Re: Lengthy ... but interesting channel config question. |
|
|
Centurion
Joined: 06 Jul 2007
Posts: 116
|
| Vitor wrote: |
| Around 20 years of hearing the same stupid statements & questions will leave you both corroded and leaking the excess. I've had people who want me to shutdown the queue manager & just start their queue because the box was short of resource. Or claim they're getting 2033 errors because their queue's not running. |
That is why I endure your bile. I can empathise. You are clearly very
knowledgable, and subsequently jaded and drained of all patience. 
In the OOP paradigm all things definable and/or configurable in MQ would
be objects with executable code in the guise of methods. A QALIAS would
for example merely pass a calling object the reference/pointer to the
QLOCAL that is configured to point to. And it would actually be a method-
call to the QALIAS object that would execute that logic.
So yes, perhaps MQ was developed in C in procedural code, which would
make everything static structs with alterable attributes. But you have to
appreciate that IBM are trying their damndest to serve the user/admin the
illusion that there are indeed objects doing things with and to one another.
So yes, persistence to storage does not per se imply executable
code. But in OOP one typically only bothers with persistence, if the object
is of some special significance and/or frailty.
Thanks for illuminating the stack of generally un-technical reading
material I still have to sift through to attain technical 'gurudom'! 
_________________
Here's an idea - don't destroy semaphores unless you're certain of what you're doing! -- Vitor |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Feb 06, 2008 8:08 am Post subject: Re: Lengthy ... but interesting channel config question. |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| friedl.otto wrote: |
| Vitor wrote: |
| Around 20 years of hearing the same stupid statements & questions will leave you both corroded and leaking the excess. I've had people who want me to shutdown the queue manager & just start their queue because the box was short of resource. Or claim they're getting 2033 errors because their queue's not running. |
That is why I endure your bile. I can empathise. You are clearly very
knowledgable, and subsequently jaded and drained of all patience. |
This is not bile. This is me on an average day. My bile is much worse than this.
But that's.....
| friedl.otto wrote: |
In the OOP paradigm ... there are indeed objects doing things with and to one another.
|
IMHO you need to distinguish the user's view of the product, which in Java & C# is OO, from the administrative view of the product, where the channels and the definitions live.
But that's a bit as well, or at least the subject of a new thread.
The path to gurudom is long but I'm confident you'll achieve it.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
