DSXchange

I don't have that particular interval calculation but the routines are here and I'm sure you could adapt something once your business rules are nailed down as Craig indicated.

The best way is the way that's correct for that particular kind of lock. For example if it's showing up in LIST.READU then it's local to the DataStage Engine.

It is a whole different world post 7.5. You actually have to install a suite of software called IBM InfoSphere Information Server. There's no such thing as "just DataStage" any more, even if that's the only licence you have purchased.

On UNIX the Inode is indeed the inode of the file (or the DATA.30 file in a hashed file) in which the lock is held. Most of the shared locks (type RL) are held against the UV.ACCOUNT hashed file in the UV account. Freeing these won't help your problem with locks held by defunct processes. The correc...

Server jobs do not use the default.apt configuration file (or any other configuration file) because they do not execute in the parallel engine.

Just for grins, what does the error code mean? You can research this using the oerr -1406 command.

They are environment variables set for your project, accessible through the Administrator client. Search DSXchange for names such as APT_TRANSPORT_BLOCK_SIZE and APT_BUFFER_MAXSIZE, and then research these in the manuals or the IBM Information Center (on-line help).

There might be a VOC entry from earlier installation. Use the Administrator client command window to execute this command. LIST.ITEM VOC 'HASHED_STORE_DEL' Let us know the result. Your next action will depend on the result of this command. If something is listed, then you can delete it using DELETE ...

You can add my vote to what the others have said.

Integration with source code control systems - to which Roy is alluding - uses Information Server Manager (which you DO have - it's free with Information Server). Direct support exists for IBM source code control systems such as ClearCase - for others you need their Eclipse plug-in. We have made goo...

Or even fieldname - Int(fieldname), multiplied by 10**x where x is the number of digits.

There are several paradigms.

Easiest is to write the header then append the detail and trailer.

Or you could create a dummy field with value 1 to header, 2 to detail and 3 to trailer, and use a sort/merge collector on this field to bring them together.

Just a couple of ideas.

Do you mean they retail "smalls"?

There is a Link Execution Order tool in the Transformer stage toolbar.

Data do not need to be sorted on business key. When the dimension table is loaded into memory the business key column is identified. That is sufficient.