DSXchange

You are correct Craig. It is control Z. I dont know about dos but in unix it is control z (^Z).

Look at the ICONV conversion code MCD. Thats what you need.

Well just go to the transformer, look the bottom half where the metadata is visible. Examine both input and output visually. You will be able to see which column has a bigger length and is being mapped to a column with lesses length.

Ok. Did some testing. Go to the Seq. File -> outputs tab -> columns. Scroll right to the "Incomplete Column" property and select replace. What replace does is it replaces it with NULL. This was atleast your job wont abort. You can then handle this particular row from not being passed on in the trans...

So what you are trying to say is, when you are picking up the original ftp'd file, there is no box present, but when you copy it a different folder it magically shows up???
How are you copying the file? Is via datastage job or through some other mechanism?

Did'nt you get a similar problem resolved here?

True.
As i mentioned, if it appears at the end of record then MCP conversion code can be used. If it appears in a new line, datastage will treat it as a new record with all the columns missing and hence scream.

Excuse my ignorance. I keep thinking px aggregator works the same way as server. I just double checked and cannot find any place where you can specify the incoming data's sort order.

where are you copying it from and where to. I use to see that character when i would ftp a file from mainframe to windows in ascii format. But if it was ndm'd (ftp'd) directly to a unix server, that 'box' wouldnt show up. Does this box appear at the end of the last character or after the last row on...

I think thats what you are missing, letting the aggregator know that the data is sorted on the specific fields. Just sorting wont help. You will have to inform the aggregator that it is sorted.

Yup, his memory is razor sharp.
Craig, let us in on your secret

Duplicate Post
Another thread is opened here

Trim it. When your re-enter it manually you are omitting the blank spaces. Thats why this behaviour.
Try trimming the incoming data and then perform the lookup.

Here's what i found on google.
http://post.doit.wisc.edu/library/techlib/manuals/adoclib/aixprob/msgdnref/err0481.htm

Scroll down for the error code.

I dont think thats whats happening in the OP's case. I think its more like a datatype problem. beaditya might be expecting a numeric sort on a char field or viceversa.
Even i try not to use the sort stage. I usually adhere to the unix sort. Much faster.