DSXchange

You need to follow the directions: In the designer: 1. right mouse click on the Pallette text at the top of the pallette box 2. choose "customize Pallette" 3. In the top-left Repository window, select Parallel -> Processing -> Basic Transformer 4. after activating, click on the right arrow to add it...

Rameshrr3, I think that solution is not to stop DSXchange from linking to external sites, since it doesn't seem to affect the 10,000+ other users, including myself who is at a site that severely restricts internet access. Perhaps you could see that your censor software acts less choosy? When followi...

Shrinivas, please post in the appropriate forum - either Server or PX for this question (although it does look like PX). What is your problem with the error? If you have a column defined as char(16) and try to put the varchar text "Hello World, my name is Arnd" into it you will get this error message.

Suresh - He might have, but the PX compiler would have balked at that, even if the red highlighting in the transform stage wouldn't have given it away [but in Server that would have worked just fine!]

Koti, it is easy since the conversion is done automatically. You define your input file's columns and datatypes in the stage, then you can output from that stage to a flat file or database of your choice. If you have any doubts or questions you can look at the documentation for the Complex Flat File...

Maintenance of the &PH& directory log files is/was an oversight in DataStage since version 1. You can remove the contents from UNIX, so a simple script could remove any files older than a week or two.

You need to know your column and line separators and the data types (and max lengths) for all your columns. Put those into the stage and you will be all set.

Does the project still show up in the drop-down list of projects in any of the client tools? If it is no longer in the UV.ACCOUNT file then it will not be in that list and if the directory still remains you can delete it. This problem can occur when you have had a user create or compile or reset a j...

The modify stage doesn't allow combining commands

How long are the columns on the source side? 1 or 999? Does the length come back as "0" or perhaps as null? Do these columns contain binary data? Are you using a 1-node configuration file or a multiple node one (just asking in order to cover all the possibilities)? If you know Server job programming...

Add a new column that is derived from "LEN(YourBigColumnName)" and see what is reported. VarChar(1000) is nothing special for DataStage so the error most likely lies elsewhere. How long is the data in those columns that show up as empty?

A Column length of 1000 is nothing particularly long. From you description you are missing 3 records, not colums; or did I misunderstand? Could you post any warnings or error messages you are getting so that it might be easier to understand the issue?

You mentioned a commit frequency of 500. If you change that to a commit frequency of 1 and run the job with a 4-way configuration, does the error disappear, remain the same, or change?

What have you specified as your commit frequency?

DataStage "DATE" conversions only cover the date portion and not the time. You will need to specify . That will give you the time accurate to the nearest second without milliseconds.