DSXchange

Welcome aboard.

Unfortunately you are wrong in your assertion. The change detection stages can take one input directly from the target table. No Lookup stage is required.

I suspect you first learned change detection in a server job environment.

Sounds like someone's been burned in the past. Experience is what you get when what you expected to happen doesn't.

Does Workbench cost extra?

How are you trying to sort? What kind of data are the sort keys?

If you are using dsjob with -jobstatus the exit status of the job is also the exit status of the dsjob command. Therefore you have two possibilities. You can parse the output from the dsjob command itself and do whatever it is you need to do with that (e.g. write to a file). You can interrogate the ...

Correct. There was a one-time-only half-price deal for early adopters.

Yes, why not? Load them when initializing a stage variable. It's surprisingly efficient.

Are you intending to convert the timestamp to a date? This may be the source of your problem if the input column is a timestamp and is not null.

You don't have that much disk space. The limit is approximately (max_file_size * 10000 * partition_count). So, if the maximum file size on your operating system is 2GB and your Data Set is five partitions, that's 100000GB. Most operating systems these size have a larger allowable maximum file size t...

What Craig says is apposite.

Why would you have more than one value in the state file?

The state file is intended to "deliver the next surrogate key value".

Try disabling row buffering completely rather than increasing.

... and whatever it is you're doing fills the scratch disk. Allocate more.

If it's urgent why are you waiting for my advice? You are paying your support provider; let them earn some of it.