DSXchange

No poster is compensated in any manner, including the Inner Circle. Next time come to IBM Live and meet us at the DSXchange booth, or take the classes being offered.

You folks do realize there are costs associated with this site (hardware, software, electricity, etc)? Again, your questions and answers are NOT BLOCKED. Share all you want.

Do you have any concept of how much time the top 10 posters have spent answering questions? How about hosting the hardware and making the embellishments to the site source code? Who do you think pays the electricity bills?

Not to mention that when a posters knowledge becomes as potent as Rays or Craigs or Arnds, their answers are not Premium content. You have a competing free resource. But, do you think they are spending the same effort to answer?

kaps wrote:DSExport command can also export individual jobs.

Thanks

Care to share the exact command line syntax for how you accomplished this?

Notice that short and concise answers are not premium content hidden. But if you look at the history of the forum, with long lengthy posts, you see the value there. Notice this post continues under the Premium answer because I'm windbagging. As far as going forward, we're constantly getting the same...

Not to mention we'll be giving summaries in the top 3 lines, providing detail behind the button. So, if you're needing in-depth, time-consuming answers, consider the effort on our part.

gmorey wrote:If it is free, why do I have to pay to read your posts?

Premium answers from the Inner Circle and the ETL Experts Panel. Other posters answers are still FREE.

There's a maximum number of readu locks allowed. I see this a lot and yes performance is impacted with every write. By the way, sequential files also incur readu locks, but only on the file. readu locks on hash files are at the row level.

Is it 5 seconds to get the first row or the last row of data? Many people confuse the quick response of the first found rows as the indication the entire process takes a short time. Try using an ORDER BY in your SQL query tool to cause all rows to be computed first and see if the time changes.

Search the forum, use google. This one has been answered enough times.

Probably permissions.

I'm a little curious about the /* PDB - AnsattFraStage */ comment in the table name. You should remove that, otherwise the SQL looks correct for the Oracle stage. It might be the data, consider if NULL values are being passed in, or a mismatch with the column definitions, typical of user-defined SQL...

Several methods. By duplicate, do you mean exactly the same, every bit on every byte on every column of data is identical? Or do you mean the "key" columns appear multiple times, and you need to consolidate the rows, potentially choosing different values from different occurences of the rows to arri...

This is not a tool issue. The industry practice is called surrogate keys, and they require a natural key for assignment. The use of a changeable value as a primary key is a fundamental mistake in data modeling. The unique columns that identify a row are called the "natural key". You need to talk to ...