DSXchange

The DataStage users need to be able to write to the same files for job designs and log files and many more. Usually I do a root install on UNIX. Before I do then I create a dstage group and a dsadm user. The dsadm user will own most of the files created by the install. All developers are added to th...

No it does not. We had 7.5.1A and it had a bug like this. We had to get a patch to fix it.

ray.wurlod wrote:Serks

Tell them Ray man.

Let me get in on this. Above?

You can pay me now or pay me later but quality matters.

I guess my answer was too direct or too short. I wanted the shortest answer. How'd I do?

You do not need SETFILE if you use the local account. You do this by leaving the Account name blank or the path name blank. UV stage treats hash files as tables and you use SQL to extract the columns. I can't of any advantage to update a hash file with a UV stage. They are sort of interchangable. Yo...

There is a patch for Designer on XP. We could not change routine activities. Not sure it will fix this error.

Probably nulls.

Ross

Ken Bland posted a set of jobs to do this. I think it was even developed in version 6 or earlier so it should work fine for most posters. It is very well documented and several DSX users actual have this in production.

Also add some DsLogInfo() calls so you can tell where it fails.

Call support then post your code. Maybe we can help.

How wide are these rows? That is slow. What about mb/sec?

The problem is the metadata. I do this all the time. Part of EtlStats has jobs like this. You have to use a parameter. When I reseed the SDKsequences then I use:

select max(#ColName#) from #TableName#

The SQL in one job is:

#SqlStmt#

I would download Ken Bland's routines. He has a real nice methodology. Otherwise you need multiple sequences. I would suggest no more than 10 jobs in a sequence. A sequence should be by subject area with related jobs. Do not run all extract jobs in a sequence then all transform jobs then all load jo...