| Author |
Message
|
| KIT_INC |
 Posted: Tue Dec 01, 2009 10:58 am Post subject: External datasource name |
 |
|
Knight
Joined: 25 Aug 2006
Posts: 589
|
| My environment is broker V61 on AIX. We are trying to create a bar file that can be run in all QA (QA1 to 5) environment. The only parameter different is datasource name (QA1DB, QA2DB....). So we use data of the I/P message to dynamically set the datasource name. However we found that the flow we fail if we leave the datasource name field of the compute node blank. Is this a limitation ? If that is the case, then we can not get away from overriding the bar file before deploying to different environment. Did I miss something? |
|
| Back to top |
|
 |
| Vitor |
| Posted: Tue Dec 01, 2009 11:12 am Post subject: Re: External datasource name |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| KIT_INC wrote: |
| If that is the case, then we can not get away from overriding the bar file before deploying to different environment. Did I miss something? |
IMHO you're better off overriding the bar file. Imagine if someone using your QA2 environment was using QA1 last week, and forgot to change the identifier in the test data.
Also your QA people might not enjoy having 5 sets of test data to maintain. By putting the configuration in the deployment where it belongs, the same data runs against all 5 environments in perfect safety.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| KIT_INC |
| Posted: Tue Dec 01, 2009 12:05 pm Post subject: |
|
|
Knight
Joined: 25 Aug 2006
Posts: 589
|
| I agree with your logic. But if IBM allows dynamic datasource whtat's the point of insisting a valid datasource name in the compute node configuration. |
|
| Back to top |
|
|
| KIT_INC |
| Posted: Tue Dec 01, 2009 12:11 pm Post subject: |
|
|
Knight
Joined: 25 Aug 2006
Posts: 589
|
| One more point. We will most likely put the Datasource name in an external variable, once it is set by a message. So not every message will contain the information needed for datasource name setup. |
|
| Back to top |
|
|
| mgk |
| Posted: Tue Dec 01, 2009 12:23 pm Post subject: |
|
|
Padawan
Joined: 31 Jul 2003
Posts: 1642
|
| Quote: |
| whtat's the point of insisting a valid datasource name in the compute node configuration |
The reason is to be able to generate the ODBC statement strings early. There is a restriction that if you use multiple DSNs in the same compute node, they must all be compatible with the type of the DSN specified on the node. This means all DB2 V9 or all Oracle 11 DSNs for example, you cannot mix and match. This is because the ODBC SQL generated by ESQL is "tuned" to the capabilities of the DSN. And the node's DSN is the one that is used to obtain the capabilities of the DB and generate the SQL. So there is no requirement that you actually use the node's DSN inside the node, only that it is "compatible" with the dynamic DSNs used in the node.
I hope this helps,
_________________
MGK
The postings I make on this site are my own and don't necessarily represent IBM's positions, strategies or opinions. |
|
| Back to top |
|
|
| KIT_INC |
| Posted: Wed Dec 02, 2009 5:08 am Post subject: |
|
|
Knight
Joined: 25 Aug 2006
Posts: 589
|
| mgk, thanks for the explanation. |
|
| Back to top |
|
|
|
|
