DSXchange

Please open up your DS/Basic manual and search for the conversion codes "MO" and "MO0C" to see how DataStage can convert between octal and numeric/string representations.

How about using the attribute for "incomplete column"?

Doing "view" data on Japanese characters does not always show what you expect. In your example, you put in 2 sets of 2 kana characters which are encoded in the extended part of Latin-I and are not necessarily depicted correctly. You will need to cut-and-paste Kanji characters and you will need to kn...

Alex, there is no builtin function to accurately do this in DataStage. There are 3rd party product for name cleansing out there (Trillium software comes to mind). In the past I've programmed my own logic in short routines, using spaces, commas, and periods as delimiters and using a list of known pre...

Posting an identical request 3 times in as many minutes won't get you a response any faster, nor will it make you any friends.

Moderator - please delete this triplicate post

Isn't this identical to the post just made by bhaskarreddy?

Moderator - please delete this identical post

If you re-run a job with the same parameters and inputs/outputs, don't you expect the same results with lots of warnings? Are you looking to get rid of the warnings, or to set the job so it aborts after a given amount of warnings? btw, This quote is attributed to Albert Einstein: Insanity: doing the...

If you get that message then the process(es) are probably still active at UNIX. Can you identify the ones running for 10 days and a do a "kill {pid}" on them? The "kill -9 {pid}" should only be done as a last resort, since it leaves locks open in DataStage - perhaps you have already done this and th...

Sometimes the Director shows a process as still running when the actual process is no longer active. Have you checked from UNIX to see if there is still a 10-day old running process? If not, a reset or recompile of the job sequence will do the trick for you.

DB/2 has no turncate table command, when DataStage executes a "CLEAR TABLE" it actually will perform a delete under the covers and this can fill up the rollback. You can use the "import from /dev/null..." DB/2 syntax in a before stgae to quickly truncate the table.

DataStage should be able to handle this type of file, so knowing what errors you are getting would help.

Could you tell us what the "various errors" are/were? Is a variable block file one with different record types?

Can you not do your A and B join in the database and live with just one reference?

What is your source? Is it a database or a sequential source? Are your joins A->B->D 1:1? Is your source number of rows many times the size of the reference, equisized or much smaller? These factors do play a role in the design phase of a job.

Ravindra, I don't know what is wrong with your job, but the design is quite inefficient with the hashed file lookup being performed for each input row {I hope that the hashed file is at least loaded to memory}. You should redesign so that your job is called from a sequence which gets the row count (...