DSXchange

That's not enough. Square is simply how your data viewer is displaying a character that it doesn't know how to display. You have to discover what these characters are before you can do anything about them. One way to do this is to write the data to a text file (maybe using DB2 export facility) then ...

Why not invest in a premium membership (less than Rs12 per day) that would allow you to read the entirety of premium posts and, as a bonus, contribute to keeping DSXchange alive. Premium memberships are the model by which the hosting and bandwidth costs of DSXchange are met. If you are a DataStage ...

Then you can loop through the number constructing its binary equivalent into a string which can be written into a text file.

Iteratively divide the remainder by 2 until there is no remainder, using integer arithmetic.

There is no such thing as "some junk character" in ETL. Every character in the data is part of your client's data, and you have no authority to change that.

You must process the data using an NLS character map that can handle every character in the data.

Re-start job automatically is not possible; it is only possible to restart sequences and, even then, they must have restart capability enabled. You need something external (possible within Management Console) monitoring which DataStage jobs are running, something local that can detect network interr...

Just curious, have you tested it with more than one table? Also, are there likely to be any "troublesome" columns, such as BLOB data type, that might bite you? The short answer is: run it multiple times. If you require these to be simultaneous, then create the job as being enabled for mult...

No. If you're really interested use server side tracing to see how the Deep Search does it (even if you don't understand what each actual function call is doing, you'll get some appreciation of the vast number of function calls that are actually made - even though the actual recursive search is perf...

You're getting exactly what a right outer join is supposed to return.

Your options include removing duplicates upstream of the Join stage or filtering records downstream of the Join stage.

0x20 is a shorthand way to specify the hexadecimal number 20, which is the same as 32 using a decimal base. The character whose ASCII code is 32 (in decimal) or 20 (in hexadecimal) is the space character (" "). If your data contains Vertical Tab characters use Convert() function to remove ...

The Sequential File stage for server jobs is much better equipped for handling missing columns. Get this working in a Server job then, if you feel it's necessary for other reasons, encapsulate the server Sequential File stage and its output link in a server Shared Container, which you can use in a p...

I would continue along the path you are on - analyse the logic and reproduce it using DataStage. There may be conversion tools (I am not aware of any, but IBM are always keen to nab Oracle sites, so may be prepared to help). In my experience, however, the results produced by conversion tools, even w...

You can find everything you need at the IBM Information Center from which you can find a link to the documentation, from where you can download the Planning, Installation and Configuration Guide. The Information Center treatment is better than earlier versions, as you get separate documents for each...

Allocating more scratch space (new configuration file) will certainly help. If you allocate enough, it will solve your problem. Allocating buffer pools in scratch disk probably won't have any impact on tsort scratch space. Any effect would be negative, as you'd be forcing buffers to use the same scr...

There is no entry for spinlocks. Spinlocks are a different place to wait for events - rather than just waiting on a semaphore placidly, spinlocks are a "busy wait" that keep on re-trying immediately. Doing so with many locks can chew up quantities of CPU. By all means take alternative advi...

What do you do in the presentation layer when valid but formerly out of range values begin to appear? This problem has just occurred here. (It's hot!)