DSXchange

What are you trying to accomplish? These samples are strings, not BigInt. Therefore they will be left-justified. Please provide more detail of your job design, and any transformation expressions you may be using.

Take a look in the directory RT_CONFIG362 in your project directory. There should be precisely three files: DATA.30, OVER.30 and .Type30 Please advise whether this is the case. If someone, somehow, has put any other file in this directory, it ceases to behave as a hashed file and begins to behave as...

That is an ENTIRELY DIFFERENT QUESTION

Please start a new thread.

ODBC does not use the Oracle OCI.

Why do you want the process ID?

DO NOT KILL a job's process - it will have several child processes. If you kill the job's process these all turn into zombies.

And please put some thought into creating meaningful Subject line, to assist people searching the forum in future!

Why not, thebird? Sometimes it must, for example when doing a lookup from a Lookup File Set or when doing a sparse lookup. The message about dropping a component with the same name can safely be ignored - it's just an alert. You can, except as noted above, eliminate it by effecting a column name cha...

In the job change the default value to $PROJDEF

This instructs the job to read the value from the Administrator, unless overridden when the job is submitted.

Once you've done this, you only need a -param option in dsjob if you want it to run with a different value than the default.

No it's NOT all you need to do. With auto-purge disabled your logs grow and grow and grow and eventually IBM want's to sell you more disk and also consulting services to repair the hashed files (job logs) that you've corrupted when the file system becomes full. This is NOT a resolution - it's a work...

Look here for example server routines. Right click and choose Save As to download.

Have you set up parameters for the shared container, and are you passing values to these from the job in which you have used it?

You must regenerate the account number for every row in which there is only a branch number. You can not load a NULL key into a hashed file.

You can't load any data at all into a hash file, because there's no such thing in DataStage - it's a hashed file.

Pravin needs to edit the Profile at least to enable PM, if not to provide an email address. The email address is not re-sold, and DSXchange itself is the mailer, so it can't be harvested or hacked.

Hopefully what you did worked. It appears to be correct; I've no way of knowing for sure! The pairs of numbers, you probably worked out, are the column locations in the record (start length). Tip: check "Disable Smilies in this post" when posting anything with 8) in it - otherwise it's translated to...

Be careful what you wish for.

The overhead of invoking a JVM for every row is probably something you don't want. Prefer

If you're happy that you have a solution, please mark the thread as Resolved.