| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Designing for flexible common infrastructure |
« View previous topic :: View next topic » |
| Author |
Message
|
| alechko |
 Posted: Sun Jan 31, 2010 6:25 am Post subject: Designing for flexible common infrastructure |
 |
|
Apprentice
Joined: 12 Jan 2005
Posts: 37
|
Hi there,
My organization has several message flows which are very similar but each is doing a different business logic.
We've designed the flows so that they all make use of infrastructure subflows which expose several properties (promoted properties). These subflows include the start of the message flow, logging, error handling and the end of the message flow.
The problem is that once we make a change to the infrastructure (the subflows), we have to redeploy all the msgflows that make use of it - quite annoying!.
We thought about 2 solutions which none is perfect:
1. Rewrite the compute nodes to javaCompute and then we can redeploy only the jar file once we make a change. The problem is that this solution only helps with compute nodes.
2. Seperate the infrastruture to a different message flow. The problem is that it breaks the UOW of the entire interface into several small UOWs.
Do you have any other suggestion?
Is there a way to call a subflow which resides in a compiled object in the runtime environment instead of compiling it into the bar file (doing a "link" rather then "call")?
Thanks,
Alik |
|
| Back to top |
|
 |
| Gaya3 |
| Posted: Tue Feb 02, 2010 6:00 am Post subject: Re: Designing for flexible common infrastructure |
|
|
Jedi
Joined: 12 Sep 2006
Posts: 2493
Location: Boston, US
|
| alechko wrote: |
We thought about 2 solutions which none is perfect:
1. Rewrite the compute nodes to javaCompute and then we can redeploy only the jar file once we make a change. The problem is that this solution only helps with compute nodes.
2. Seperate the infrastruture to a different message flow. The problem is that it breaks the UOW of the entire interface into several small UOWs.
|
What is the problem in the First option, even this is the same logic we implemented in one of the project over here.
_________________
Regards
Gayathri
-----------------------------------------------
Do Something Before you Die |
|
| Back to top |
|
|
| Vitor |
| Posted: Tue Feb 02, 2010 6:23 am Post subject: Re: Designing for flexible common infrastructure |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| alechko wrote: |
The problem is that once we make a change to the infrastructure (the subflows), we have to redeploy all the msgflows that make use of it - quite annoying!.
|
Only if you're doing the deploy by hand. Once you automate it then it can a better option. Even a script with the mqsideploy commands will take the sting out significantly, and with some moderate effort you can use ant (or similar) to build the bar files for you.
Used exactly this sort of technology for exactly these reasons; logging & error handling in common sub flows referenced by 20 odd main flows. The bar files in this example were built with ant as the 20 flows only did routing and message input/output; the compute nodes & good stuff were in sub flows which in turn used the common flows for logging (if you see what I mean).
With this much code sloshing round it was safer to rebuild the bars for each deploy from the source code repository so we could guarantee which version of code was where.
My 2 cents.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
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
|
|
|
|
