DSXchange

Ray, I just checked as well - U2 files can't be accessed readily using ODBC in EE/PX jobs. Not even at version 8. Strange that more esoteric ODBC connections are available but not UV. I wonder why this wasn't done, perhaps to dissuade developers from keeping hashed files when converting from server ...

Ray - I thought it had already been moved.
iamnagus - Please ignore my post until the moderator moves the thread

I'm not sure if one cannot declare a hashed file in an ODBC stage in PX - I can't check right now, but would assume that this ought to be a simple matter to include in the ODBC configuration.

The number of Nodes in an APT_CONFIG configuration and the number of CPUs on a system have absolutely nothing to do with each other. You can run a 100 node configuration on a single CPU system or a 1-node configuration on a 64 CPU system. The configuration file tells DataStage how many levels of par...

Off the top of my head:

(a) Full moon and Venus is visible,
(b) Computer is hungover and cannot remember where table is
(c) The data is too heavy for the CPU and cannot be lifted.

Seriously, though - the most common cause of looup failure is when the lookup fails.

I think that your ulimit settings for root are limited, since that user spawns the other processes this is where you limitations are inherited from at runtime.

Put in an additional column for testing, setting it to 1 if "In.Code_val_c=' '" and 0 otherwise and you'll know the answer.

I doubt that you correctly set the nullable attribute in the DB column, otherwise failed lookup column will be coming back as null.

Question 1 You can look at the source code of KeyMgtGetNextValue in the manager and you can see where it generates the starting value. Make your own copy of this routine and then modify the starting value to be 100000. Question 2 Define your column as a numeric one, DataStage will do the conversion...

Set the column Code_val_c to be nullable from your lookup and try again.

Welcome to DSXchange, arunselvaraj. (that red is very difficult to read) If the column coming from the lookup is declared as nullable=Yes and your lookup is set to "continue" on failed lookups then the value will be null. What are your nullability settings in your job and please explain which column...

DSGetStageInfo() doesn't take a DataSet name, it takes a stage name as input. You can get the row information from the DSGetLinkInfo() routine, using the input link name and stage name.

I really should have looked at the thread. There is a bug in DataStage regarding attaching to the job that is running. The documentation states you can't do it, but you actually can --- the side effect is that the job "hangs" for a long time a while doing this. I had this problem a couple of years a...

Bingo - Craig, you might have gotten it there. A difference between testing the routine from the manager and running the job from a sequence is just that - that the DS_JOBS entry for the sequence itself is locked in the latter case. Normally, I cannot see how a read lock should affect this simple SE...

Can you replace the 3 parameters 1 by 1 to see which one is causing the error? My guess would be something to do with the password...

Apart from looking at the missing dll error message in your log, if you can compile a job without a transform stage but cannot compile one with a transform stage then either you don't have the correct compiler installed or it is not configured correctly.