DSXchange

You could certainly run an external program to PGP encrypt it (using ExecSh or ExecDOS) from withing DataStage. As 1) an after stage routine, 2) an after job routine, 3) from a Job Sequencer or batch job).

Good Luck,
Tony

Louise, The Oracle client must be installed and configured correctly on the Solaris server on which DataStage is running. In addition, the directories containing the Oracle libraries must be added to some Environment variables in the 'dsenv' file. The 'dsenv' file is in the ../DataStage/DSEngine dir...

Craig, You're probably aware of this, but incase you're not... I just noticed that I can use Job Parameters in different stages in a Job Sequencer, but I can't put pound signs ('#') around them... ROOT_PATH : "/" : PROJECT_PATH: "/config_files/JobControl.cfg" ROOT_PATH and PROJEC...

Chetan, I suspect that Parallel Extender was not installed when the server was installed. Parallel extender doesn't not automatically get installed with DataStage server. If your company bought parallel extender, I believe that you can install it without wiping out your current DataStage jobs, etc.,...

heh! I see stuff like this all the time and it cracks me up. Just shows that the customer's requirements may be "negotiable". Have a good weekend, Vincent!

Tony

Srinivas,

If you do not have the same number of rows returning from each link to the collector, it will hang waiting for the "next" link to process.

Hope this helps,
Tony

Kenneth, I hadn't considered a batch job. That would take care of it and I could handle all the conditions that make the other approach somewhat brittle. Am I thinking correctly that having several developers put hash files under the project directory is just asking for trouble (unless someone coord...

I would like to be able to access a hash file with the UV stage, but have the hash file in a directory other than a project directory. I know that when I create a hash file, I can put it in another directory by using the "Use directory path" option. However, if I do this, I cannot access t...

Emma, When I ran into this need, I created an extra field on the hash file that was a "composite" index, i.e. it contained the contents of both "key" fields, concatenated together and separated by some character (I used a tilde). So creating the field looked something like this: ...

Abhishek Sharma, You can write the data to a sequential file and use the bulk load stage or sqlloader to load the data into Oracle. Are you doing any lookups in your job (server job, not PX)? If so, are they live database connections or hash files? Live database lookups can slow your job down signif...

Guillermo,

Have you used the 'release job', 'package job(s)', 'install package' functionality in Manager? This can be useful in these situations.

Tony

Can you see what the '?' chars are? I did a hex dump and saw that they were char(25) (hex 19). I never did figure out where the char(25) was coming from. Those columns were empty in the input flat file. I handled it something like this: In the column derivation I had a transform NullToSpace NullToSp...

What I found was that I had some columns that were NULL. I was putting them into a hash file and reading them back out. OracleOCI8 -> Transformer -> hash -> Transformer -> sequential file I would get errors writing to the sequential file. I would NOT get these errors if I just wrote the input data i...

In the Columns tab on the Input tab of the Aggregator stage you should see a column in the grid marked Sort. Put a 1,2,3,4, etc in this field to designate the sort order.

hope this helps,
Tony

Stephan, You could create one new field - it would contain the contents of the two columns (separated by something like a tilde). This could be done when you create the hash file. Then build two new indexes. One on the two column key field and one on the single column that you need to loosen to. You...