DSXchange

The DIR /X command will show you what the 8.3 name of a Windows file is, sometimes when there are duplicates the ending is not ~1. The DataStage engine came from operating systems that did not allow spaces in the path and I have never felt comfortable using such filenames as they didn't always funct...

This is odd, I've never seen the compiler itself hang. Perhaps it, too, is waiting on a lock. Can you monitor the activity of the process that is doing the compile from UNIX - to check whether it is using CPU cycles? Does PORT.STATUS show any activity for this process?

Ray corrected me in a nice way - CHAR() and SEQ() are opposite functions and I was only 50% off

DataStage uses one character as a field mark - that is ASCII 254; if you have this in a string in a hashed file record it will be interpreted as the effective column separator. This might be causing your error. Other than that you should be able to store any of the other 255 ASCII values in a field....

There can be a number of reasons for a sequential write to fail, but I don't think that the error code of "0" is correct or is going to help. Can you please try redirecting the output .dat to write to /tmp. Does it work? If yes, then most likely you have a permissions or a filespace issue in the ori...

Reset your job in the director and then look at the message titled "from last run", this might give you more information. If your job were not compiled you would be getting a very different error message.

Veni,

are you aware that you can call up UNIX commands from sequences and jobs? This will allow you to take care of your file deletion.

Poorna, start off with something simple, like an Excel sheet with only one column and see if you can get that to work correctly in view data. Then you have an idea of how it functions and can analyze your problem. Since Excel allows unstructured data that is often the source of errors. In your case ...

ravij, the only systems you will find which will have identical runs on separate runs are mainframe jobs and some other realtime operating systems. In your case the measurement times of 3/4 seconds are so small as to make any statistics absolutely worthless. Your minimum snapshot should be a minute ...

The first error is the only one that matters - if the DataStage engine is incapable of using (i.e. attaching) to the disk shared segment 0xade... then nothing else can or will work. These shared memory areas are used for all interprocess communication in the engine as well as for holding all locks a...

How do you know that the link partitioner is only going down one path? It would simplify your debugging to make the two link partitioner outputs write to sequential files and remove the shared containers from the equation completely. If the containers use sequential files for writing then the constr...

I want to increase the load data by dateseeta, your question is still unintelligible. We understand you are loading from a sequeuntial file to an oracle database. But what does I want to increase the load data by date mean in this context? You want to increate the value of a date column? You want to...

What's a Holiday? Is that some new stage?

Can you post two or three lines of your sequential file and the resulting lines after FTPing to this thread? I don't know what else to say right now without further information.

It looks like insufficient access rights; there should be more text to the error message. You can also execute this command manually and see if more error text is generated. I doubt that this is the first error message; either your engine failed to install & start or you have access issues.