DSXchange

Is there a way we can decrypt the encrypted value and encrypt back again before sending it to Oracle? Or is there a work around for the above scenario. If you encrypt the value yourself, you need to have its twin process to decrypt the value. If the password was defined as a job parameter with type...

I think, the reason lies in the very architecture of the two systems and the nature of a shared container. SMP uses a single shared memory irrespective of how many cpus are connected to it. The result of the shared container resides in the shared memory and whatever cpu requires the shared contraine...

Why dont you export the executables as well, if you want to refrain from compiling all the jobs/routines again?

I know, it has happened a lot.
Next, I dont know, lets see whos phrase I use next

kamal kishore wrote:Hi Brian Kernighan,

Wow, he called me Brian Kernighan.
Read the manuals. Its all in there. Or better yet, search here, this has been discussed here before.

Depends upon what compiler you have. I have aCC so i just typed
aCC +Z routine.c

If you job is compiling then its not a compiler issue. Try a force compile and then run a job. Refer to thispost.

Does your password have any special characters like ($) ?

Try the following in execute command stage to check if any files with a particular patter exist. ls <<filedir>> | grep aaam_classif | wc -l where filedir is the directory where the files are. You can run that in execute command stage. If the result is greater than 0, process your job, else send out ...

Welcome Aboard
See if this posthelps.

kamal kishore wrote:I had installed C++ and then also i'm getting the same error can anyone tell what are the environment variables that are need to be set.

Compile the job as ArndW requested.
Four environment variables need to be set also, namely
APT_LINKER
APT_LINEROPT
APT_COMPILER
APT_COMPILEROPT.

You write the code and compile it at the unix level to create the object file (.o). Then give the path to that object file while creating the routine definition.

If you already have a hashed file with distinct values then yes, go for that option. Also, you need distinct values from table A, do a lookup against table B, get all the keys that exist and pass it to the delete statement. OR you can do a lookup on table B and get all the records from table B that ...

One target stage with before sql

update option set to "Insert rows without clearing". Source will be your TABLE A.
But that is going to take time for 2M records.
You need to go for TRUNCATE and BULK loads.

narasimha wrote:Check the sdk routine "UtilityRunJob" (Not sure if it exists in the PX edition)

That routine can run any job, provided its called from within a server job or a Basic routine.