DSXchange

You can't. Server edition does not have the concept of data types. You will need to perform any necessary validation testing in the Transformer stage, and create an output link from that which can handle any rows that are not passed to the third stage. Server jobs do not have the concept of Data Set...

There are several change capture products that work by "scraping" the transaction logs. You might like to investigate these. Otherwise you need a "before image" of the table before any of the insert, update or delete operations has been applied, and some way to compare the new wi...

There is no way to manipulate an Excel workbook with UNIX scripting because Excel does not run on UNIX machines. In my experience the only effect of the leading single quote is to force Excel to accept the following value as text. There are no real data types within Excel - if that value can partici...

Type 1, Type 2, Type 2/3 or Type 3 slowly changing dimensions? Change detection for dimension involves comparing newly arriving rows with what's already in the table. The easiest way to do this is to load the candidate rows (currently active rows) and columns (columns on which "change" is ...

Even one of the IBM experts needed more than ten attempts to upgrade from 8.0.1 to 8.1. Wild horses could not extract that expert's name from me.

The best scenario for scenario 3 (need more than one Join stage) is to use more than one Join stage. Provided your system has the resources, there is no limit on the number you can have. But that's a major proviso.

Edit the values file and have the Parameter Set reference the values file.

The best approach is not to measure the thing being measured. Wait till the job finishes, then do the measuring.

The other best approach is to let someone else do the work for you. Download ETLstats from Kim Duke's website for free.

Then try rebuilding the index.

No, .odbc.ini is also UNIX only. On Windows the ODBCINI Registry entry is read. Likewise $DSHOME can be found from the Registry.

The uvodbc.config file in the project is there to filter which DSNs are visible from that particular project (and, by elimination, which ones are not).

You would only have missed a word or two, so I removed the premium status. Premium membership is one of the ways in which the hosting and bandwidth costs incurred by DSXchange are defrayed. For this reason I do not wish to create any precedent of undermining that mechanism - in particular I respond ...

Configure it IN the Parameter Set, not in the job or connector. The Add Environment Variable button is available in the Parameter Set editor dialog.

If you create values files, these will appear in the drop down list as well as "(As pre-defined)".

A job run request from a DataStage client is actually made by that client's agent process (dsapi_slave) on the DataStage server. Agent processes inherit their environment from dsrpcd then, on UNIX, modify it by any statements in the $DSHOME/dsenv script. Project-based environment variables are set u...

Where did you get those numbers? I know why the 18 character limit exists, but it's not directly documented anywhere. It stems from the fact that the local repository is a schema in the DataStage Engine database, which is compliant with SQL 92 standards, which include an 18 character limit on schema...

There is a DELETE.INDEX command. Syntax is as you'd expect. You can use ALL in place of indexname.

If it's a bug it's probably in the data browser. Try a query against the hashed file (use either LIST hashedfilename or SELECT * FROM hashedfilename;).