DSXchange

As you can be sure that the job isn't running anymore, you can do a Job->Clear Status File in director.

Stephan

To accelerate lookups on hash files, the option 'Preload file to memory' can be checked in the hash file stage. Some of our lookups are rather small, containing less then 20 records. Is it advisable to preload these files, though they allocate 128 MB (or whatever is adjusted on the system) as the la...

If I get you right, you have serveral outputs, but you want them all to write to one sequential file. According to my experience, it is better to write to seperate files first and then concatenate all these files to one, e.g. using something like 'cat file1 file2 file3 > file4' in an after job subro...

Is there a way to 'uncomment' (I don't know how you would express this in english, what I mean is 'mark as comment') a larger group of lines in basic source code other than preceding each single line by an asterix? Is there a way to continue basic code on the next line, like it can be done in ubix s...

Is it possible to have a unix shell script attaching to a DataStage project and querying the project properties using basic functions like, I guess it's name was DSGetProjektInfo? If yes, how would it look like?

Stephan

Craig

isn't there a 'dsenv' file in $(cat /.dshome) where, among othersm the PATH variable can be set?



Stephan

Ken, Ray again: I assume, that unchecking the 'preload file to memory option' is what is meant by not using read cache. And again, the job fails to work as intended. Less records are marked as used in the hash than the monitor pretends to have found (in the hash) and updated, both in the database ta...

Ray

I assume, that it is sufficient to uncheck the 'Preload file to memory option' to not use the read cache you are talking about?

Nevertheless, DataStage keeps refusing to do as you suggested.


Stephan

John

most likely, 2 of your records are equal in their key values. If this is the case, the earlier record is overwritten by the later. You can only have one record per key expression in a hashfile.

Stephan

Say, you had a hash lookup with 3 key columns. You would want to gradually loosen the matching criteria from all the 3 columns to 1 column, but still using the same hashfile. How would you have to specify the key expression in a transformer, if you wanted to unconditionally evaluate the matching aga...

Cal it's essential to compose the command string the right way. I never used sequencer, but did something similar as ExecSH -> before-job sobroutine, where you wouldn't have the "" around the command. If this is no solution, maybe you have to concatenate(:) the elements of the command like...

Shiv

take a look in here:

http://www.tools4datastage.com/forum/to ... C_ID=84667

Stephan

Cal my experience is that only the invocations are deleted if you compile the job. The logs will be still there in the very job (without invocation) unless they are deleted manually or by auto purge. View the logs of the job without invocation and drag the column border in the heading between '>On d...

Ken I already tried like you suggested, marking the used records, but this didn't work (I tried the options 'Disabled' and 'Disabled - lock for update' within the hash file stages). What I want to do is as follows: there is a datasource outside our dwh, which is fed from the same source, a flat file...

Is there an easy way to provide that records from a hash lookup are picked only once, even if subsequent records from the stream input would match the same row again?

Stephan