DSXchange

So where, above the line you posted, is ORACLE_HOME set?

There's no guideline, per se. As noted, it's all about where you can best afford the processing power. Unless those eight connectors come from different instances, I would pretty much always push this back on the database server rather than do this "in job". Unless you've got a mandate to ...

The moment the job completes those files are obsolete, so use something reasonable as the time component for them. Throw them into your "seven days" pile worst case.

Char(10) = LF = UNIX record terminator
Char(13):Char(10) = CR/LF = Windows/DOS record terminator

[/pedantic]

Sad? I think you just earned the "Solved that last problem on my own" achievement. <chime!>

Right... that or something like a Citrix Server.

Off you go to the correct forum for site suggestions.

Search is your friend! The list here is from 2008 so not all inclusive I would imagine but still extensive:

viewtopic.php?t=118110

Down near the bottom is the name of the header file where they should all be defined.

I'm going to leave the recovery steps to people who have far more knowledge in that area, someone like Ray for instance. Doing an exact search on the forums for your 30105 error turned up lots of hits, thought this post might be an example of the kind of help you need: http://www.dsxchange.com/viewt...

I don't really see how "messing up" where the file went could cause folders to disappear. Now, if you were writing the file to the same disk / partition where the projects live and filled it up, that could cause project corruption but again not disappearing folders. Any chance this job did...

Isn't this the relevant part of the error message?

<v1:value>Message part {http://soa.nordea.com/BusinessRelation/ ... DIDRequest was not recognized. (Does it exist in service WSDL?)</v1:value>

Short answer is yes.

Do an exact search here for "multi-row result set".

Use ODBC which allows a 'multi-row result set' to be returned. That or use the UV stage to access your hashed file. Unless things have changed, those were the only two options for a multiple record lookup.

I thought practically all of a Sequence job's compilation 'effort' happened on the client.