OSH generation problem in dboptions (password part) with OCI

brwalid · Post by **brwalid** » Fri Mar 23, 2007 9:35 am

Hello All,

We're trying to read an Oracle table with these parameters in an OCI Stage :
Data Source Name = PRODUCTION_DATABASE
User = USERNAME
Password = 1FRM

After the compilation, DS generate the Orchestrate code like

## Operator options
-nchar_cs [&DSProjectMapName]
-db_cs [&DSProjectMapName]
-dboptions '{user=[&User],password=[&Password],arraysize=1500}'
-server '[&PAR_ORA_PRODUCT_ALIAS]'

The problem is that the Orchestrate code add a & before the parameters that makes password=&1FRM.
Oracle is waiting for a value substitution because of the &1.
The easy issue is to change the password but it's not always possible.
I know that \ annihilate the & in SQL Plus.

So, is there a way to annihilate the & from Datastage\Orchestrate coding generation in the dboptions ?

Thanks for your help.

ray.wurlod · Post by **ray.wurlod** » Fri Mar 23, 2007 5:42 pm

Welcome aboard. :D

Does this occur if you use auto-generated DBoptions, supplying user ID and password through explicit properties? (Yes, I know the array size will default in this case, but let's get the password issue solved first.)

If this works, what is in the DBoptions property? In particular, are the user ID and password references separately quoted? Maybe that's what your DBoptions should look like.

CLOPES · Post by **CLOPES** » Sat Mar 24, 2007 10:13 am

ray.wurlod wrote:Welcome aboard. :D

Does this occur if you use auto-generated DBoptions, supplying user ID and password through explicit properties? (Yes, I know the array size will default in this case, but let ...

Same thing :
Orchestrate generate automatically the code with something like password=&[my_password].
I have to try with a number for the first character for the user and the dsn to see if it'll do it too.

CLOPES · Post by **CLOPES** » Wed Apr 18, 2007 6:57 am

It's a PX bug.
IBM have opened an engineering case for that.

ray.wurlod · Post by **ray.wurlod** » Wed Apr 18, 2007 3:57 pm

Do you happen to know the eCase number?
If so could you post it here please?

CLOPES · Post by **CLOPES** » Thu Apr 19, 2007 6:41 am

ray.wurlod wrote:Do you happen to know the eCase number?
If so could you post it here please?

Our case was Case No - 503782*FR but IBM have closed it to open an engineering case and they will reopen it as soon as the engineering have
resolved this ecase the 16 April.
Don't know much more.

CLOPES · Post by **CLOPES** » Wed May 16, 2007 7:50 am

CLOPES wrote:
ray.wurlod wrote:Do you happen to know the eCase number?
If so could you post it here please?
Our case was Case No - 503782*FR but IBM have closed it to open an engineering case and they will reopen it as soon as the engineering have
resolved this ecase the 16 April.
Don't know much more.

IBM have made a patch for the "&" problem in an Oracle password.
:D