> -----Ursprungligt meddelande-----
> Från: IBM Mainframe Discussion List [mailto:[email protected]] För
> Steve Comstock
> Skickat: den 27 februari 2012 15:13
> Till: [email protected]
> Ämne: Re: SV: Duplicating SYSOUT output to another DD/DSN in realtime ?
> (JCL)
> 
> On 2/27/2012 7:01 AM, Thomas Berg wrote:
> >> -----Ursprungligt meddelande-----
> >> Från: IBM Mainframe Discussion List [mailto:[email protected]] För
> >> Steve Comstock
> >> Skickat: den 27 februari 2012 14:58
> >> Till: [email protected]
> >> Ämne: Re: Duplicating SYSOUT output to another DD/DSN in realtime ?
> (JCL)
> >>
> >> On 2/27/2012 4:18 AM, Thomas Berg wrote:
> >>> Is there any possibility to duplicate the output to SYSOUT to another
> >> Ddname/DSname in realtime ?
> >>> I want to follow the execution by inspecting the output but at the
> same
> >> time save it for processing in a following step.
> >>> Preferably by JCL means.
> >>>
> >>>
> >>>
> >>> Regards,
> >>> Thomas Berg
> >>> _________________________________________
> >>> Thomas Berg   Specialist   A M   SWEDBANK
> >>>
> >>
> >> If you want to 'follow the execution by inspecting the output'
> >> and 'in realtime', I gather you don't want an automated
> >> process.
> >>
> >> Why not just go to SDSF (or (E)JES or IOF or Flasher or ... )
> >> and select the sysout file? It will show you what's been
> >> output so far, but does not disturb the final distribution.
> >>
> >> You can set up an automated, timed refresh of the screen
> >> (say, every 5 seconds) to watch the SYSOUT data grow, or
> >> just hit enter to see the latest entries.
> >>
> >> Or maybe I'm not clear on what you are after.
> >
> > The output needs to be processed in a later step. But if the output is
> directed to SYSOUT, are there any way for a later step to read that output
> ?
> >
> >
> > Regards,
> > Thomas Berg
> > _________________________________________
> > Thomas Berg   Specialist   A M   SWEDBANK
> >
> 
> Ah! So the output is not originally SYSOUT? Can you direct
> the output to a z/OS UNIX file? Then from a UNIX session
> (omvs or telnet) you could use 'tail' commands to watch
> the file grow.
> 
> If things are not going well, cancel the job. If all goes
> well, copy the UNIX file back to an MVS file (or just
> process the UNIX file if that would be supported).

Well, I suppose that's a good idea. But here I'm out on unchartered 
territories. :)  Although *NIX land is not totally unknown for me I haven't 
used the z/OS variant much. 
Maybe it's a time for me to explore it...

But as this routine will go into automated production in future I prefer MVS 
proper (as I don't want to change the job to much between the "manual" period 
and the automated production. 


 
Regards, 
Thomas Berg 
_________________________________________ 
Thomas Berg   Specialist   A M   SWEDBANK 

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

Reply via email to