DSXchange

Good point, Ray. Most of them did, but not all, and the primary group wasn't 'dstage', but a department group. We made all the files and directories that we access through DataStage owned by the dstage group. That explains why we had to set the 's' bit on the group.

Thanks,
Tony

I'm not sure if this is normal, something specific to HP, or something we're not smart enough to figure out a way around, but we had to set the 's' permission on the group for the project directory and any directories where we wrote hash files or data. This is supposed to make any directories or fil...

Make sure you have your permissions and umask set right...

Tony

Andrew,

Look in the help files for the 'dsjob' command. That's the easiest way to call any DataStage job from the operating system.

Good Luck,
Tony

Also remember that DataStage basic is case sensitive, so a variable named TonysVariable is not the same as Tonysvariable. You may be seeing an issue where one of the variables that you're using is misspelled.

Good Luck,
Tony

In a second transformer parse the row into columns using substring() or field().

Or use a Row Splitter stage instead of the second transformer to split the data out into columns (v7.1).

Tony

SonShe, Design your job like this: HashFile1 | | Hash_lu IN V SeqFile------->XFormer------->HashFile1----->XFormer----->SeqFile2 Make Fld1 the key when reading and writing in the hash file stages. For the derivation of Field2 put: If IsNull(Hash_lu.Fld1) Then IN.Fld2 Else Hash_lu.Fld2 : " "...

Also consider the scalability of a RoutineActivity stage per job parameter. If you have 20+ parameters in a job (paths, credentials, dates, etc), then this becomes an nasty job. Yes it does, no argument from me. I wish we could stuff all these in a container, but there is no container stage in Sequ...

Kenneth, I think that this answers Poorna_76's question, but since you said that the Sequencer can't do this, I'm not sure if I understood his question correctly. Can you review the scenario I describe below and let me know if I'm off base? Thanks! Poorna_76 We do this with most of our job sequencer...

You can do the same thing in DataStage, but I don't know if it will be any faster: 1) Create a new server job. 2) Add a transformer and a sequential file stage. 3) Create an output link from the transformer to the sequential file stage. 4) Create a new stage variable in the transformer stage. Call i...

If a job truly is "released" you should not have to specify the job name plus the release number. At least the last time I tried it.

Tony

Byron,

I usually write a new Routine and call all the after job routines from this one routine. Obviously, I just set this one new routine as the After Job routine.

HTH,
Tony

Yeah, I did. Sorry.

With proper indexing, I would think this would beat the pants off your ODBC lookups. Maybe even without proper indexing.

No, you pretty much have to have proper indexes when doing stuff like this unless the amount of data is very small. But it's pretty simple to add extra indexes.

Tony

If you're talking about the OCI8To9.b tool that Ascential distributes, I've used it a couple of times. It worked fine with no problems.

Tony