DSXchange

It's conceptually easy to design - you have to keep track of how far you have successfully processed. You also have to design so as to be able to process beginning from a given point. Quite a few techniques exist. It's easiest if your overall design includes staging areas after extraction and before...

For hashed files, you probably need to add physical storage overhead, since this affects the number of rows that you can actually store in a group and effective transfer rate (since entire groups are transferred). Storage overheads for hashed files are: (a) three pointers (each 32-bits or 64-bits) (...

I often use 3.5 inch "floppy" disks for these transfers, and have had no bad experiences.
Try importing on the desktop (even if not a "real" import) to determine whether the export file is corrupted. If so, export again.

You can do this with a Hashed File stage. The sort phrase is augmented with the FMT field qualifier. You don't need EVAL if using a Hashed File stage.

The best function to use is Index() or Convert(). Index(searchcharacters, mycharacter, n) searches for the nth occurrence of mycharacter in searchcharacters, returning 0 if not found. Convert(searchcharacters, "", mystring) removes all of the searchcharacters from mystring ...

Job status comes from RT_STATUSnnn so the most likely thing to work is "clear status file". However, this will reset the status to "compiled". You should be able to determine the real status by inspecting the log.

What happens when you try to import into a non-NLS DataStage a file exported from an NLS-enabled installation? Can you fiddle with the CharacterSet property in the file header?

To connect to the databases jobs use the Dynamic Relational Stage, with the access protocol set appropriately (Oracle, Sybase, IBM DB2, etc).

Table definitions have been preloaded for tables that contain PeopleSoft data.

There's no such thing as "DataStage groups" - I suspect you're referring to DataStage roles. You attach DataStage roles to operating system groups in the Administrator client. Definitely do-able. Create one operating system group for each project, and another (say dstage) for all DataStage users. Ma...

Check for link variables such as LASTERR and DBMSCODE.

Hey anybody is there??????????? If you want rapid service, buy priority 7x24 support (and learn the true cost of "urgent"). That no-one had replied in 40 minutes should not be a cause for concern; this is an all volunteer site, contributors log in and post when they can. I'm guessing as to where yo...

Anyone care to suggest what this thread has to do with the site?

The job probably isn't really running. What the Director status view shows you is the most recently recorded status of the job, which is read from the job's RT_STATUSnnn table. If the job was killed without the opportunity to update RT_STATUSnnn you will still be reading a status of "running". That'...

As I understood it, only validating the job or choosing the Create check box cause a hashed file to be created. That knowledge is from 5.2, however; have not verified it since.

There is a function DSGetLogEntry for capturing the detail of a logged event. This requires the Event ID (a sequential number) of the event in question. You can determine the Event ID, for example by calling the function DSGetNewestLogId (which can filter on "warning" type). You can find out more in...