DSXchange

Celeste, VSG is "visual schema generator", which is a part of UniData that makes a schema that ODBC clients and other SQL applications can access. It presents the data in a normalized form, since this is what these environments expect. For example, multi-valued fields are not recognized by ODBC; ins...

To clarify my comments on hash files and the copying of data between them: * The permanent store of data is the data warehouse. Hash files are used to expedite the lookup and allocation of surrogate keys. * I simply use the data copying method I described as a way to pre-populate hash files in diffe...

Neil Thanks for the link, Damien Kavney MIS Technical Support Department Of Natural Resources & Mines PH: 38963989 -----Original Message----- From: Neil Beckwith [mailto:neilbeckwith@hotmail.com] Sent: Thursday, 4 July 2002 3:57 PM To: datastage-users@oliver.com Subject: Re: archive link for Oliver....

Thanks Ray, and by the way, good luck with your future endeavors! Rays correct, I assumed when doing a hash copy of the data and overflow files that you used DataStages default DYNAMIC hash files. Theres a hidden file ".Type30" in every DYNAMIC hash files directory. Also, I assumed Chris was doing a...

MONTHLY POSTING - DataStage Users List As of 07 July 2002 the DataStage Users discussion list had 661 subscriber addresses. A few of the addresses are to organizations who then redistribute the postings to more than one reader. ================== Posting Guidelines ================== 1) Non-moderate...

A word of caution about copying DATA.30 and OVER.30. Two in fact. (1) Make sure the target directory also has a zero-length file called .Type30 in it. (2) Make very sure that the source hashed file is not in use (the ANALYZE.SHM -d command can be used for this). Kenneth Bland cc: Subject: Re: extern...

In Administrator (project properties > log purge), you can set the default rule for all jobs in the project (for example, purge entries more than 1 day old, or all entries except the last run). In Director, (Job > Purge Log) you can change the default for the currently selected job. "m_musumeci@inwi...

Stacy, I would seriously consider keeping your development on a separate server to production. We ran for over 2 years on a single server and as a result, were not able to upgrade DataStage, as we did not have an environment in which to test the upgrade. There is no way I would upgrade my production...

> Anybody know if DS 5.2 for AIX can receive and process a stored > procedure system-return-code at run time ? > > Previous postings on this list stated lesser versions could not. > > Its understood the most common DataStage use for stored procedures is > to read or write rows in a database, and Dat...

Hello, I would like to purge log entries for all the Jobs in a Job Batch and I would like to do this at every Job Batch run. May I set the default purge settings to the existing job from the Administration Window ? If not, may I create a routine for purging ? Thank you for interest Massimiliano

Chris, I would argue first of all that you should never store "production" data within your tool (DataStage or whatever). If you need a persistent and static lookup, put it in your warehouse! That way you can refresh your hash file whenever, because hash files can corrupt, etc. As far as copying has...

Your metadata should be stored in DataStage in the table definitions. You want a system of record of all metadata stored inside DataStage. Using the dictionary created by DataStage doesnt necessarily reflect the true content of the hash file. This is because failure to delete and recreate the hash f...

All I am faced with a dilemma - we currently have two instances of DataStage running for two different departments. Both have development/test/production folders. As we plan for growth, we are wondering if it is not better to build one ETL server that handles ETL for all departments and then assign ...

All I am trying to set up the security for DataStage on a Unix machine. I know that on NT, security can be assigned to NT user groups and users assisgned to those groups. I am not familiar with DataStage security on UNIX. What is the best way to handle this and at what level is security assigned. An...

Hi Lawrence, If you are upping the clientside user limit, then my understanding is that you will need to do a client reinstall on each of your machines. Machines left with a lower client licence will probably fail to connect when that limit is reached. I dont think you will need to uninstall first t...