I can highly concur with the recommendation of using IEBGENER instead of cp.  
My experience has been a much better controls, and easier comprehension and 
adoption by application coders.  There are *many* options of getting a file 
into omvs for use with those tools.  Don’t blame cp for working as it’s 
supposed to.

That being said, overlooking or ignoring Co:Z is always a great injustice.  
Pretty much everyone on this list will tout its merits.

I’m also amazed how often people think sftp is the only OpenSSH replacement for 
ftp.  Most of my users were transferring single text files with an ftp step.  
Demonstrating how to use scp instead of sftp reduces the complexity enough to 
make it a more viable option.  I suggest checking that out.  The IBM guys who 
ported OpenSSH toolset did an excellent job detecting and providing charset 
translation with scp tranfers.

Also, if it’s a text file on disk, you might want to experiment with letting 
the ssh transfer be driven by the client side.  Remotely driving an ssh shell 
to the mainframe, to do a “cat” over the shell back to the remote, allows you 
to directly access MVS cataloged datasets without the copy to the omvs 
environment, thus saving that step altogether.

Breton Imhauser
Acxiom Corp., LLC.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to