DSXchange

The first thing you need to do is to figure out in which stage this problem is occurring - from the error message it's almost certainly a Modify or Transformer stage. Disabling operator combination, either at the stage level or at the job level, will help to prove this. Ultimately, the error message...

A Transformer stage with a single output link can not process rejected rows. You need at the very least two outputs. Otherwise all you can do is to drop the invalid rows. The last-executed output link can be an Otherwise/Log link, in which case it handles any row that has not met the output constrai...

I need to connect the DB2 database throgh script. Why? You have a range of perfectly usable DB2 stages available to you, for which someone presumably has paid a great deal of money. Therefore, since that person is also paying you, ought you not to be using the tool? Have you been engaged as a DataS...

It would have been better had the rep given you the reasons for waiting for Hawk. MetaStage ceases to exist with Hawk; is replaced by a tool called Information Analyzer. A lot of the MetaStage-style reporting will be (is?) available directly on/from the Designer canvas. That said, it might also be w...

The reference Data Set for a Lookup stage must fit into available physical memory. It may be this amount of space that you are exhausting. If this is the case, consider whether you are carrying unnecessary data in the reference Data Set and, if not, whether you can solve your requirement using a Joi...

You already have the method. Set preserve partitioning to Clear and/or set downstream partitioning to Auto.

A small correction, it's not the .uvodbc.config file, it's the uvodbc.config file. There is one in each project, against the possibility that you may need different ODBC data sources to be visible from different projects. You may also benefit by installing and configuring Oracle client software on t...

Given that you tried it logged in as root, I think that you need to involve your support provider. The first question is whether you are running on a supported platform. In the meantime, DataStage will almost certainly still work OK using the default start points for shared memory segments. These ar...

Don't try to read a four-partition Data Set with a job (or stage) that is constrained to operate in only a single partition. That's what the error text is telling you that you tried to do, by forcing a downstream stage to use the same partitioning as an upstream stage that was possible executing in ...

Ereplace is not present in parallel jobs, except in the BASIC Transformer stage.

It should be whatever it is. But sendmk asserts that there is no compiler on the production server, which is the question that is really driving this thread. If they didn't have a compiler called CC, then the operating system would have generated a "command not found" message. This did not occur, wh...

Not unless you create one.

You could, of course, use a BASIC Transformer stage, which would give you access to the Ereplace() and Change() functions, but these are not natively available in parallel jobs (Modify or parallel Transformer stage).

If you need to generate them there (but why?) there is only one solution, and that is to get a compatible C++ compiler. Otherwise you have to distribute the executable components (assuming everything, such as O/S, is compatible). These are all stored in the project directory, in a subdirectory calle...

Exception

If, for some reasons, you have hard-coded the mapping between server name and IP address in the hosts file on your client machine, you will need to correct this.

This link says it all!