SQL Server load problems

[ShaneMuir]

I have a simple job (sql -> lookup -> transform -> sql) which was working fine with a sequential file as the target but when I change the target to a SQL server stage it falls in a heap.

The transformer in question maps the source columns to the target columns (ie the column names are completely different) with a small amount of manipulation. The target column meta data was imported. However the transform seems to pass not only the target columns but also the raw data to the db. I was able to confirm this by changing the Write Mode from Truncate to Replace. When I then checked the db it contained the nicely converted data but also all the columns and raw data from the source

I have rebuilt the job again and the same thing is happening.

I am assuming that this is not normal behaviour, so if anybody else has had similar problems I would love to hear if you were able to solve it.

Thanks in advance

Shane

[ray.wurlod]

That definitely sounds like a bug. Have you reported it through your support provider? Clearly you can generate a reproducible case. When you say "SQL Server stage" I presume you mean "SQL Server Enterprise stage"? Are you selecting from the same table that you are updating?

[ShaneMuir]

That definitely sounds like a bug. Have you reported it through your support provider? Clearly you can generate a reproducible case. When you say "SQL Server stage" I presume you mean "SQL Server Enterprise stage"? Are you selecting from the same table that you are updating?

Hi Ray,

A reproducible case seems random - I played around a little yesterday, inserted a peek stage between the transform and the sql sever enterprise stage to have a look what was being passed and suddenly it started working, removed the peek stage and all was fine still. Although a colleague here had the same problem yesterday in a new job that they had created. (she got around it by setting the truncate column option).

Will note it down and report it.

Thanks