DSXchange

Your original post had "GetFName", not "GetFNames". The job name must be exactly correct. The error message, after all, was failure to attach job.

Welcome aboard! :D I agree with Ken. Chances are that your routine code could be streamlined, for example with judicious use of the Ereplace() function. You can post your code here if you'd like us to look at it, and it does not violate any non-disclosure agreement you may have in place. If you do, ...

I have seen this kind of disconnection occur with the DRS stage when an attempt is made to access an unsupported data type. In my case, it was a CLOB in Oracle. Changing the column derivation to CAST(columnname AS VARCHAR2(4000)) resolved the error. Nonetheless, it ought not to break the connection;...

If the job name is "GETFNAMES" then this is the name you must use in DSAttachJob, not "GetFNames". Job names are case-sensitive.

Yes, you must install the client software. It's not automatic, most times folks want to separate client and server machines.

You have already purchased some Ascential product, which is how you can have evaluations of other Ascential product. I had wondered.

DSAttachJob failed says it all, really.

You invoked DSAttachJob("GetFName", DSJ.ERRFATAL)
DataStage could not find a job called GetFName in a suitable state for attaching; for example the spelling or casing is incorrect, or the job has been deleted, or the repository indexing is inaccurate.

Your requirement is not something that a hashed file is geared to do; the whole raison d'etre of a hashed file is to return THE row whose key is given. You would not even be able to populate your proposed hashed file as you wish; you would end up with one row for each key value; the most recently wr...

Don't forget also to check any column name you might have typed (mis-typed) into a WHERE or other clause on the Selection tab. And also check that you have spelled the table name correctly (it, too, is an identifier).

Do you have the Oracle CLIENT software installed too? At run time DataStage is a client to the Oracle server, so the Oracle client software must be installed on the DataStage machine, and properly configured, even if the Oracle server is on the same machine. Make sure also that you are using the 32-...

DRS was introduced in 7.1 (perhaps even 7.0) when Peoplesoft elected to switch to DataStage. It allowed Peoplesoft to deploy canned jobs where the only change between their different customers might be database type.

Seems to have crept in since Ascl took over the maintenance of Version Control from our friends at DCS!

Yes, stop even believing that hashed files connect to anything. They won't help you to connect to Sybase. Hashed files are nothing more than an implementation mechanism of a database table, using in databases such as UniVerse, UniData, D3 and others. And, of course, DataStage Engine. You need either...

The message tells you exactly what's wrong. You are not authorised to use the at command on the UNIX server. Contact your UNIX administrator; you may need to be added to cron.allow, or you may need to be granted write access to the atjobs and/or cronjobs directory (or added to a group that has such ...

No, what I meant was that you should define an ODBC data source name (DSN) that uses the ODBC driver for text files to connect to the text file in question. This is one of the drivers supplied in the branded_odbc directory; consult the help there to determine which one (I don't have access to UNIX d...

What you will need is a stand-alone job that consists solely of job control code. This code loops through all jobs in the project, determining the name of the log file, then purging that log. Open "DS_JOBS" To DSJobs.fvar Then ClearSelect 9 SSelect DSJobs.fvar To 9 Loop While ReadNext JobN...