DSXchange

I'm sorry but I for one don't understand your problem or what "is changing from single/double/triple" might mean. :( On the subject of creating multiple output files, there's no mechanism in PX that I am aware of to accomplish that, especially when you add in naming the file dynamically ba...

Yup, command output is a dynamic array so you need to remove the included (invisible) Field Mark or specify only the first element of the array. And as noted, discussed here many times with full syntax examples.

Why are you using CREATE.FILE? That is for creating one in an "account", i.e. a project. As noted, a pathed hashed file is created using mkdbfile as your log entry showed and can be created anywhere.

No, I just grabbed one of the first links that Google returned that explained it was used by MKS Toolkit. It was just meant to be a starting point. I doubt it is any kind of error, but it would take more digging (or a support call) to find out exactly what service it does provide.

Google is your friend.

http://www.file.net/process/nutsrv4.exe.html

The VOC record depends on how you create it. For a pathed hashed file, use the mkdbfile command you'll find in any job's log that has created one for you.

IBM won't care and there's no advantage, one over the other. Choose whichever you are familiar with and have the ability to support.

Is your environment setup correctly? Meaning, did you make sure the dsenv from the 8.5 installation was sourced as part of this script?

Why not renew your membership and read the entire message? By "simply clear it" I meant uncheck the other two options you listed.

chulett wrote:Also make sure you use an Array Size of 1.

What is 5000x2000000000? Stick with 1.

You do know they are not supported, yes? However, that doesn't mean they can't be used although 2GB might be a wee bit on the 'too big' side. Can you leave the length empty? Seem to recall that as one trick. Also make sure you use an Array Size of 1.

An exact search for "deadlock detected" returned 27 matches, did none of them help?

... and does it work if you run the job on a single node?

The stage only creates 32bit hashed files for some reason. You'll need to create it one time as 64bit from the command line and then have the stage simply clear it. A before-routine could emulate the settings you are using in the stage - delete and recreate - and then the job could just use it. To c...

I think you're going to be pretty much on your own with this one. We did something like this many years ago for a Compaq Tru64 cluster with shared EMC storage but even then there was no "continuing". Everything died and then was restarted on the failover node. And that was simplified by be...