DSXchange

to me it looks like its not doing the lookup all my records are being sent down the didn't find a match link, I've viewed the job in designer and once I@ve run the job there it reports that 0 rows were looked up. I've viewed the hashed file and the records are in there.. am just wondering if I need ...

i've changed the constraints on my outputs to lkReferenceLookup.NOTFOUND and not(lkReferenceLookup.NOTFOUND), but its sending the wrong record to the wrong output and its hangs after the ora-0001

Apologies for such a beginners question, no DS experience: How do I perform a hashed file lookup. Specifically how do I set up my constraints.. I've created a transformer stage with two inputs. The format of both inputs is the same, one is my hashed file for lookup (lkReferenceLookup), the other com...

Hi, Couldnt help asking. Ray you said that keepipng logs of DS for a long time without purge would degrade performance? We keep getting requets of data load dating back to more then 12 months, due to which we dont delete any logs. Would purging them help in performance improvement? Also is there a ...

Having just had a quick scan in my Oracle book, I don't believe that nologging will offer any benefit.

It is a very useful tool, but you need training to be able to use it properly. Hopefully they will not pull the tool because they have not given proper training, that would be the result with any tool, not just DataStage. bit of a chicken and egg situation, I can't convince them to keep it without ...

It's got nothing to do with database 'Archive' errors. And keep in mind the fact that Auto Purge of a job's log will only kick in if the job finishes without warnings. James - you need to take the time to get up to speed on hashed file stages and precheck lookups to determine insert v. update (etc)...

Did you even look at the DataStage log (via Director)? My first question would be why so much stuff is being logged; in an ideal world there should be only a few tens of entries per run, and no warnings at all. Is auto-purge disabled, so that the log grows indefinitely? If so, manually purge the lo...

kcbland wrote:ORAOCI or ODBC, whichever you choose. The bulk loader is the ORABULK stage.

okay, I'll give this a go, presumably there is some tutorial or guide in the documentation to help me get started with this....

thanks! wish me luck!

Job #1: OCI (target table) --> Transformer --> HASH file Job #2: FTP (source data) --> Transformer --> SEQ file Job #3: SEQ file --> Transformer w/reference to HASH file, constraint only allows output rows where unique identifier in source not in HASH reference --> SEQ file Job #4: SEQ file --> Tra...

Perform a lookup against a hash file that contains the target table primary key columns spooled from the table. Use a constraint to only insert rows not in the hash lookup. That will run a heck of a lot faster than using database exception processing to let know the row is already there. thanks for...

thanks for the reply, yes in at the deep end as usual. well I'm not sure what my options are with this table. I can't use 'update existing or insert new' because both columns form the primary key, I tried this and I got an oracle error (sorry can't remember which one - just said about the primary ke...

Hi there, I'm having major performance issues with a DataStage job in our production environment. 2 caveats. 1 I've started to look at DataStage last week, so please excuse my greeness, secondly I've inherited everything on this! Right, heres my problem I've an Oracle 9i table, it has 2 columns, bot...