DSXchange

Telnet and "ls -al" or "du -s" will do the trick, if it is a mount point then "df" will also help confirm the missing directory contents so you are probably correct in assuming that your engine directory has gone back to the big bit bucket in the sky! If this is a development site and not production...

senthilt1 - Hint: The "@INROWNUM" and "@OUTROWNUM" mnemonics are available in the Transform stage.

Actually, are you certain that the directory is empty? I've had cases where changes in access rights and using a tool such as WinSCP to browse UNIX directories showed empty directories. As Ray said, some of the single files in that directory are almost impossible to delete while the system is being ...

Someone did a "rm -rf" on the directory. Or perhaps it was on it's own disk and someone unmounted it. Is the project directory safe? If yes, make a quick backup of that just in case.

What is "the access problem"?

Did you specify "unicode" in the extended attribute for the string fields?

"DSU.ExecuteScript" is a user-written DataStage BASIC routine. Since someone at your site wrote it and you didn't post the code, you will have little chance of getting a solution as it stands.

Since you posted this as a PX problem, the routine in question can ony be a before- or after- subroutine.

That should not be possible. If you have a job with input file A and output file B, both declared as UTF-8 and declare the character columns (I think that the extended attribute is "unicode") the same in source and target then both file will be identical.

On your pc, the PDF document is titled QualityStageParallel Job Developer Guide.

Make a reject link with a row limit of 1 come out of the transform stage and only put records down that link when you wish to abort the job. [edit] I just realized that this post is in the wrong forum, please only post relevant topics to "DSXChange Testimonials". [/edit] Moderator - please move to t...

You cannot delete single rows from a sequential file. You will need to copy the original file to another and skip the records that you don't want.

The "view data" has always been an issue with multibyte data and the representations of special characters will mostly be incorrect. Use an OS tool that supports different character sets to view data like this.

You need to set the NLS to that of the file. If several languages can occur, it is best to write (and read) the file with UTF-8

What happens if you try to specify "unicode" in the extended attributes column for strings?

[edit]
I just saw that you used the term "urgent" - If I had seen that before responding I would have urgently gone on to another topic.
[/edit]

Chandru - expanding on the reply isn't going to help much in this case, there are a number of ways to go about what you are intending to do. If you were to look into the documentation and read the chapters for the stages listed above you will get a better idea of what options are available to you. O...