DSXchange

Try putting another transform stage before the one that halts and just pass the rows through without any transformation and also put in the "1=2" constraint. Does it abort in the new transform as well? Are you 100% certain that you are not using any stage variables - those are evaluated be...

There are two ways commonly used:
a) start the job through a sequence which call the job and is set to automatically reset the job.
b) reset manually using the "dsjob -mode reset"

Also, if you add a constraint "1=2" to your transform stage, does the job run through to completion?

Use "od" to see what the delimiter character in the source file really is after the transfer to UNIX, then specify that character. This is the probable cause. To test delimeters, just choose a letter or displayable character in your file and set that as delimiter, your error message should...

PX Jobs can use any delimiter; I think that the issue you are seeing is that you are putting the delimiter character in the wrong location in the sequential file definitions. I'm not at a client now so can't do an exact walkthrough, but if you start with a new sequential file stage (to make sure tha...

Ok, let us assume that the transformer is causing it. That would most likely mean that it gobbles up memory and then aborts or does something similar. What are you doing in the transformer? Do you have stage variables (strings that might grow until they consume all memory) or other calls that might ...

I've used state files without any problems at all in the past. Were you perhaps working on a Windows platform?

Could it be a timeout? You could try using the command line FTP with trace turned on to see if you can FTP the file correctly outside of DataStage.

The PADCHAR does not affect the reading of CHAR fields. If you do a "view data" in the Designer of your source file, what is displayed as the value of the second field and what is the correct contents according to your text editor?

Phil - that is a valid point regarding the exports, I have seen that as well. In that case one could re-import into development from production and see if the job still shows the same behaviour.

Tracy - did you actually run a "diff" on the config files?

If you take this step by step you might be able to quickly locate the problem. It seems that your job aborted while writing, both to a hashed file and to an Oracle table. What about if you wrote to a sequential file; does the job complete? Does this error occur in the same place every time? What sta...

It might be a timing issue with the reference lookup, if it takes longer to create the hashed file this represented with 2 stages and it has caching turned on, then you might not have any records in it when it is being loaded for reference. Can you try to use the same hashed file stage for the looku...

While I don't agree with some of the arguments that upul used, I do think that there is no clear-cut answer to this question. It depends upon what your requirements are, I'll just list 2 factors, there are many more: - if speed is the highest priority, go for dataset or filesets - if other applicati...

Export a sample job from each environment and see if the .dsx files are identical.

What happens if you do no character set mapping in the stage but use the DataStage BASIC ASCII or EBCDIC functions?