| Author |
Message
|
| SANTYP |
 Posted: Tue Nov 17, 2009 11:52 pm Post subject: How to handle multilple messages comming to common subflow |
 |
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
Hi All,
I created an subflow with 4 input nodes,
which are connected to one compute node..
How to handle inputs from these 4 nodes in an unique way.. in the other words..
Can I recongnise from which input terminal I am receiving the message..?
Is there any tree structure that will recognise from which inputnode I am receiving the messages..?
Please respond to me ASAP thnaks in advance... |
|
| Back to top |
|
 |
| AkankshA |
| Posted: Wed Nov 18, 2009 12:35 am Post subject: |
|
|
Grand Master
Joined: 12 Jan 2006
Posts: 1494
Location: Singapore
|
populate some env variable before sending msg to that subflow for each terminal and check for the value of the same in that compute node
_________________
Cheers |
|
| Back to top |
|
|
| smdavies99 |
| Posted: Wed Nov 18, 2009 12:37 am Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003
Posts: 6076
Location: Somewhere over the Rainbow this side of Never-never land.
|
What about setting some identifier before you enter the subflow?
Then you would only need one input terminal...
| Code: |
set Environment.Flow.Identifier = 1 or 2 or 3 or 4
|
depending upon where it is called from.
_________________
WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
| Back to top |
|
|
| SANTYP |
| Posted: Wed Nov 18, 2009 1:13 am Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| smdavies99 wrote: |
What about setting some identifier before you enter the subflow?
Then you would only need one input terminal...
| Code: |
set Environment.Flow.Identifier = 1 or 2 or 3 or 4
|
depending upon where it is called from. |
Hi thanks for your response.. but is there any other way to handle this..?
becasue.. my subflow uses differnt application's message I don't have any control to set environments values for incomming messages... |
|
| Back to top |
|
|
| AkankshA |
| Posted: Wed Nov 18, 2009 1:20 am Post subject: |
|
|
Grand Master
Joined: 12 Jan 2006
Posts: 1494
Location: Singapore
|
| SANTYP wrote: |
Hi thanks for your response.. but is there any other way to handle this..?
becasue.. my subflow uses differnt application's message I don't have any control to set environments values for incomming messages... |
do not change in main flow
in subflow.. when u receive message from each input node.. add a compute node and populate the node preceding it and then merge all 4 nodes into 1 compute and check for the value...
_________________
Cheers |
|
| Back to top |
|
|
| SANTYP |
| Posted: Wed Nov 18, 2009 2:09 am Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| AkankshA wrote: |
| SANTYP wrote: |
Hi thanks for your response.. but is there any other way to handle this..?
becasue.. my subflow uses differnt application's message I don't have any control to set environments values for incomming messages... |
do not change in main flow
in subflow.. when u receive message from each input node.. add a compute node and populate the node preceding it and then merge all 4 nodes into 1 compute and check for the value... |
Its again using 5 computenodes in the subflow right..
Is there any other solution in which I can handle in single compute node |
|
| Back to top |
|
|
| SANTYP |
| Posted: Wed Nov 18, 2009 2:10 am Post subject: |
|
|
Centurion
Joined: 27 Mar 2007
Posts: 142
|
| AkankshA wrote: |
| SANTYP wrote: |
Hi thanks for your response.. but is there any other way to handle this..?
becasue.. my subflow uses differnt application's message I don't have any control to set environments values for incomming messages... |
do not change in main flow
in subflow.. when u receive message from each input node.. add a compute node and populate the node preceding it and then merge all 4 nodes into 1 compute and check for the value... |
Its again using 5 computenodes in the subflow right..
Is there any other solution in which I can handle in single compute node |
|
| Back to top |
|
|
| AkankshA |
| Posted: Wed Nov 18, 2009 2:34 am Post subject: |
|
|
Grand Master
Joined: 12 Jan 2006
Posts: 1494
Location: Singapore
|
| SANTYP wrote: |
Its again using 5 computenodes in the subflow right..
Is there any other solution in which I can handle in single compute node |
No broker property can give you the name of the previous node.... which the message followed
_________________
Cheers |
|
| Back to top |
|
|
| kimbert |
| Posted: Wed Nov 18, 2009 3:06 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
| Why is it important to use a single Compute node? |
|
| Back to top |
|
|
| smdavies99 |
| Posted: Wed Nov 18, 2009 4:25 am Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003
Posts: 6076
Location: Somewhere over the Rainbow this side of Never-never land.
|
| kimbert wrote: |
| Why is it important to use a single Compute node? |
Perhaps some people have an aversion to them?
There have been a few posts here in the recent months where the OP was very reluctant to use one when it makes sense to do so. Is there a school of 'Minimalist Flow Design' that exists somewhere?
_________________
WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
| Back to top |
|
|
| sirsi |
| Posted: Wed Nov 18, 2009 5:21 am Post subject: |
|
|
Disciple
Joined: 11 Mar 2005
Posts: 177
|
| usuage of nodes depends on ur business logic... in your case you have to use few compute nodes to fit into ur requirement |
|
| Back to top |
|
|
| Mercator |
| Posted: Thu Nov 19, 2009 9:26 am Post subject: |
|
|
Apprentice
Joined: 21 Jul 2009
Posts: 34
|
| smdavies99 wrote: |
| Is there a school of 'Minimalist Flow Design' that exists somewhere? |
Yes there is. I stumbled across a document at my current client. It was a recent recommended practices document given to them by a consultant. This consultant came from a very large company. In his recommendations he states the following:
- No more than 10 nodes per flow
- No more than 5 nodes per sub flow
- Avoid using Compute and Java nodes
I'm still baffled by this document. I'm also a bit frightened by it. |
|
| Back to top |
|
|
| agrawalamit166 |
| Posted: Thu Nov 19, 2009 10:54 am Post subject: I created an subflow with 4 input nodes, |
|
|
Voyager
Joined: 17 Aug 2009
Posts: 78
Location: NY, US
|
| If it is MQInput, I think, we can find the source name through InputRoot.MQMD.SourceQueue... |
|
| Back to top |
|
|
| smdavies99 |
| Posted: Thu Nov 19, 2009 11:21 am Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003
Posts: 6076
Location: Somewhere over the Rainbow this side of Never-never land.
|
| Mercator wrote: |
| smdavies99 wrote: |
| Is there a school of 'Minimalist Flow Design' that exists somewhere? |
- No more than 10 nodes per flow
- No more than 5 nodes per sub flow
- Avoid using Compute and Java nodes
I'm still baffled by this document. I'm also a bit frightened by it. |
Ouch
Many of the projects I have worked on over the years would most certainly fall foul of those 'rules'.
There are far to many reasons to state here why I find them impossible to agree with especially avoiding ESQL & Java compute nodes but here are a few.
- Compute Nodes: These are (IMHO) the most flexible nodes in the toolkit.
- My standard error handler sub-flow has 9 nodes including Zip, Email & FileOutput nodes.
- If you are using multiple databases if different types (eg DB2 & ORacle) that is TWO nodes. You will soon use up the quota of 10.
- These rules/guidelines might have been relevant in V2.x days but now?
I certainly would not be happy developing flows using these rules.
Just my 0.02p worth.
_________________
WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
| Back to top |
|
|
| Mercator |
| Posted: Thu Nov 19, 2009 11:41 am Post subject: |
|
|
Apprentice
Joined: 21 Jul 2009
Posts: 34
|
| smdavies99 wrote: |
Ouch
Many of the projects I have worked on over the years would most certainly fall foul of those 'rules'.
There are far to many reasons to state here why I find them impossible to agree with especially avoiding ESQL & Java compute nodes but here are a few.
- Compute Nodes: These are (IMHO) the most flexible nodes in the toolkit.
- My standard error handler sub-flow has 9 nodes including Zip, Email & FileOutput nodes.
- If you are using multiple databases if different types (eg DB2 & ORacle) that is TWO nodes. You will soon use up the quota of 10.
- These rules/guidelines might have been relevant in V2.x days but now?
I certainly would not be happy developing flows using these rules.
Just my 0.02p worth. |
Agreed. Thats why I told my client to throw the doc in the trash if they want to be succesful. |
|
| Back to top |
|
|
|
|
