DSXchange

Get smaller CLOBs. :wink: Seriously, though, consider the volume of data involved in one CLOB, then in one per row for many rows. Construct a test job that extracts all the other data but omits the CLOB - the runtime difference is your cost of extracting the CLOB. If you're using a CAST to convert t...

dveltman1 wrote:But I can't perform the range lookup with that combination.

You should be able to; the range lookup is performed against a virtual data set loaded into memory no matter what the supporting stage type is (except, possibly, Lookup File Set stage).

Fault type is the signal number. 6 is SIGABRT (abort with core dump), which doesn't really add to your knowledge.

Are you trying to load a null into a not nullable column?

Any transformation needs to name the input column(s). That's an absolute rule!

You can't use a Modify stage here, since concatenation is not available as a Modify stage operator/function.

Do you want fries with that? :D

Your input needs to have sufficient location columns defined to cover all possibilities in the data.

There's also logic to detect whether any otherwise/log and reject output links need to be used.

If you reset the job, do you get further diagnostic information logged?

So CSI means "client side includes". I wonder if the makers of the TV series are aware of that?

Hmm, nobody mentioned initialization of stage variables, initialization of loop variables or execution of loop variables.

Can you go to the Job Control tab of Job Properties in the Sequence, and inspect the generated code there? Post it here if it doesn't cause any non-disclosure issues (it shouldn't), surrounded by Code tags to keep it pretty. Check whether every artefact mentioned in any activity actually exists in t...

Columns named in the WHERE clause of the UPDATE statement should be indexed, and the table statistics must be current.

Or it may be that your source has non-printable characters. For example, if this field were at the end of a line in a text file, the line terminator might be read as part of the field.

And, of course, involve your DB2 DBA and DB2 support.

Your Char types are probably padded; you need to remove the pad characters before lookup will be successful with (unpadded) VarChar types. Preferred technique is to use a Modify or Transformer stage to effect the trim operation. You need to be aware of the pad character that is being used in your da...

gayathrisivakumar wrote:But I guess DSxchange users are not only from US

It was the weekend in most of the rest of the world.