DSXchange

Start your investigation with "port attributes are empty"

Make very sure that they're junk! Don't assume anything.

Already answered here

Posting the same question in multiple forums is frowned upon, as it wastes server storage space. Please refrain from doing so in future.

Yes, but exactly how depends on your actual requirement. For example if you always want the latest file, capture its name with an operating system command like ls -lrt dirpath | head -1 or DIR /OD/4 dirpath | SORT /R in an Execute Command activity (in a job sequence), then capture that output to fee...

So what is the workaround? Does this problem occur only on your client, or all clients using this standard network install?

Added URL to my earlier Ernie post. Don't have an answer for shaonli.

U hasn't logged in for quite a while, and may not even have the knowledge that you have requested him to provide.

The second person personal pronoun is spelled "you" in English.

Only you can determine what. But check permissions where your DataStage client is installled for a start.

What's changed? ("Nothing" clearly is not the correct answer.)

Len() function is applicable to string data types, not to date data types.

The same is true of the Trim() function.

Over and above all of that, almost all parallel Transformer stage functions are intolerant of NULL values.

If you have Runtime Column Propagation enabled it is sufficient; the job design may have completely empty column grids. Is this, perhaps, the case?

Obvious question, but someone has to ask. Have you PUT any metadata onto the link (that is, have you loaded any table definition into the Columns grid on either end of the link)?

Can you see the columns on the passive stage end of the link?

Welcome aboard.

Had you created default values for these environment variables in the Administrator client in the new (7.5.2) project. Even if it is the same project, upgrading may have overwritten the DSParams file in which these environment variables are stored.

In an upstream Transformer stage split a1 into 50 columns and preserve a2 and a3. Downstream of that use a Pivot stage.

Create a key (perhaps the per-partition row number) on each input to the Lookup, Join or Merge stage, probably using a Column Generator stage.