Re: Why isn't OMVS command integrated with ISPF?
- From: gilmap@xxxxxxxxxxxxxxxx (Paul Gilmartin)
- Date: 30 Nov 2007 08:05:26 -0800
On Fri, 30 Nov 2007 08:43:38 -0600, McKown, John wrote:
As an experiment, I composed the EXEC on VM/CMS:
It has also been mentioned a couple times in this thread that the
z/OS operator's 327x console is free of the half-duplex constraint:
the operator never needs to press ATTN before entering a command.
I'd be delighted to see an alternative implementation of OMVS that
bypasses TSO (or is VTAM the culprit?) and passes 327x data streams
more directly to the terminal.
Can't be VTAM. Using SMCS, you can now run a true z/OS console on a VTAM
3270 LU. Works identically to a DIDOCS 3270 console. We now use them for
some Production Control people who are remote to the data center, but
need to have a console around to see messages "real time". TN3270E to
VTAM to SMCS for a console. Works well with all three of our TN3270
terminal emulators (Hummingbird, BlueZone, and Rumba).
/* Rexx */
signal on novalue
address 'COMMAND'
do forever
say date() time()
do while externals()>0
parse pull Line
say Line; end
'PIPE literal +10 | delay'; end
.... then from an ISPF session, I "TSO TELNET VMCMS" in 3270 mode.
It mostly works; I can type ahead for 10 seconds, and the input
typed ahead comes out in the pull...say loop. But it stutters;
some characters are repeated; it seems that I can type characters
after I press ENTER but before the cursor is placed at the left
end of the input area. I wonder whether TELNET plays the same
invisible polling game as OMVS? Now I must try it from a linemode
z/OS Unix session, if only to see how externals() behaves under Unix.
About SMCS being unavailable during IPL, haven't I seen some
discussion here of a remote HMC? Would that suffice? Or are
there operator commands that can't be issued at the HMC?
And more on your perplexity about my suggestion of "background
initiated foreground": I've seen some sentiment here that a
class of users prefer linemode for general Unix operations, but
OEDIT/OBROWSE in their specialty areas. Wouldn't it be at least
convenient in a linemode z/OS Unix session to issue:
DISPLAY=local.desktop export DISPLAY
oedit some.file
.... and have a new, genuine x3270 window running ISPF/PDF pop
up on one's desktop? No need to enter user ID and password;
current working directory and environment inherited from the
Unix session; and if the user has issued a "su" command, OEDIT
would run with the new user ID, not that of the original TSO
session as is the present (mis-)behavior of OEDIT under OMVS.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
.
- Follow-Ups:
- Re: Why isn't OMVS command integrated with ISPF?
- From: McKown, John
- Re: Why isn't OMVS command integrated with ISPF?
- Prev by Date: Re: Coming soon, Shark CCWS is now supported by MFNetDisk.
- Next by Date: Re: Data Center Theft
- Previous by thread: Re: Why isn't OMVS command integrated with ISPF?
- Next by thread: Re: Why isn't OMVS command integrated with ISPF?
- Index(es):
Relevant Pages
|
