DSXchange

I've dealt with it, but not directly. My client recently shutdown their M204 system, but before that we were sending information back and forth. Only problem was it was pretty much treated as that crazy inmate permanently stuck in solitary confinement. We never met face to face, sometimes I'd slide ...

What database? Are you saying you are using a SERIAL field to auto-generate surrogate keys and sometimes you get duplicates? In any case, be more clear about what you are doing and how you are generating these 'serial' values. I've got no idea if you have a DataStage job design issue or a database p...

kduke wrote:You used to get a warning if you changed the value of an argument. Now it is encouraged.

Really? I thought it still threw warnings when you compiled routines that did that... I'll have to go check that out, Mr Kim. :D

5)Unconnected lookups. Meaning if the lookup of a source row gives more than 1 row. it should pass on both the rows instead of 1. [i can understand "one row at a time" is the basis of datastage. At least it makes sense to pass 2 rows in case of a lookup] Available in parallel jobs. And Server jobs ...

vmcburney wrote:... while Oracle will charge per core, even for the weaker x86 chips.

What a shocker.

If not, please enlighten us in regards to what you mean by 'objects'. Then you might get a better answer... or some even more groovilicious diagrams.

I know we've had this conversation before here, so a search may turn something up. I'm sure you've already figured out that the Sequential File stage isn't up to this particular task. One way would be to bone up on your BASIC and write a job control routine to accomplish this. There you can control ...

Write caching disabled is a normal informational message and isn't a problem. Something else caused your issue. You need to post the actual message from the log - don't paraphrase or type it here by hand, cut and paste it please. Just saying "it aborted" doesn't tell us anything.

Two million characters is only 2MB, give or take. I've used this mechanism to process 150MB files without issue, so not sure what's going on here. Perhaps it's just a matter of turning off row buffering? How big is any given flattened row?

Batch, Sequence, Server, PX... doesn't matter. If you can run it from Director, you can run it from the command line using dsjob. We occassionally pass in parameters at the command line (typically processing dates) but the jobs mostly read their own from config files. If whatever you are calling for...

Uncheck both of them Actually, the only one you need to uncheck is the 'Delete file before create' one. The create one is safe to leave alone as it just gives you the option of overriding defaults when the file is created. And it is only created if it doesn't already exist. Which actually happens w...

If 'append' means 'add new records', just write to it. If you mean something else, be more specific.

It's perfectly feasible. Any of the methods in the stage should support it. I don't use SQL Server - are you not aware of the proper 'left outer join' syntax you'd need? Check with your DBA if not.

That's old style Oracle syntax - you sure that works in Sql Server?

Because that 'just' uses cron on a UNIX system. We have lots of dependancies between processes and sometimes across servers. You need an Enterprise Scheduler to do that and we have Control-M. It also handles messaging by linking to our 'paging and escalation' system without having to script it in. I...