DSXchange

Good piece of advice. Simple and straight forward. But beware, this methodolgy might cease to exist in the next version.

Its mandatory. It has been discussed here before. Search....

Well if its in a runnable state then you "dsjob -report" will give you either Job status = 1 or 2 or 99. 99 (Not running) will tell you that its in a compiled state. If you get "ERROR: Failed to open job Status code = -1004" That means its not in a runnable state. Thanks for the clarification kumar,...

Welcome Aboard,
Dont overload the transformer then. Modularize your job.
I was not able to understand what you were saying about back end date. If you mean that its not ordered, then you can always sort your data on the date field and then load it.

What is your job design, the metadata? Also copy paste all the warning messages here.

Ok. Lets take a step back. Build a simple routine and call it from within a basic transformer in a px job. Ans = EREPLACE(Arg1, "AAA", "xxx") Build a text file with a few records that have AAA at different places. Run the px job. See if all instances of 'AAA' get replaced...

It should work but right now we are just shooting in the dark. Lets see which arrow hits bullseye. From a glance, nothing seems wrong with the routine and it should work in a px job just as it works in a server job. Write a px routine for it. I remember posting a routine that works with ascii number...

Not char, it should be varchar. Change it to varchar so that you can trim the spaces.

You can probably script it. Do a dsjob -ljobs to get all the jobs, read it one by one and pass it to dsjob -report or dsjob -log and grep for 'Job Status'. You can also check if its anything other than 1 or 2 then you can load the job name in a seperate file. A few lines of scripting.

See if that routine works with the same input in a server job?

What is your job design? Did anything happen to that column on the way to the target?

Welcome Aboard;
Whatever id fires off the job, the job runs under that id. How are you running the jobs?

There is property help. Try that. Also the manual comes in handy right about now.

O comeon trobinson, go easy on the poor guy, it was Maxim

Welcome Aboard, Well I would say, for any error that does'nt require manual intervention falls in the first category. For any error that requires a DataStage analyst to review the error message and act would come under second category. Any enhancements to the code would come under third category. Th...