It is acting more and more like one all the time :-) That is the point
of all of this - to not get phony error messages from the devices when
performing normal functions (such as RUN during DETACH or Logoff). We
have made a lot of progress from streaming unsolicited interrupts, to
reflecting a state that caused 11 lines of error message, to one that
only triggered 6 lines, to this single message. The best would be no
message at logoff and only a successful detach message if the user
detached the device. Currently, the user who detaches the device does
get the successful detach message AND the operator gets the HCPERP2215A.
So we are halfway there.

What manual do you use to determine there is a special Unit Check? Is it
online where I can see it?

Regards, 
Richard Schuh 

 

> -----Original Message-----
> From: The IBM z/VM Operating System 
> [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark
> Sent: Thursday, July 24, 2008 1:22 PM
> To: IBMVM@LISTSERV.UARK.EDU
> Subject: Re: CP Message
> 
> On Thursday, 07/24/2008 at 04:05 EDT, "Schuh, Richard" 
> <[EMAIL PROTECTED]>
> wrote:
> > The vendor of the pseudo tape unit has changed several things in the
> response 
> > during a Detach so that CP now issues only a single message,
> "HCPERP2215A TAPE  
> > 0670 REQUIRES OPERATOR INTERVENTION; MAKE DRIVE READY", 
> which brings 
> > up
> a 
> > question. Why does CP care that the drive is not ready when it is
> detached? 
> > What will get mounted the next time the drive is attached 
> is unknown, 
> > so
> he 
> > Ready/Not Ready state is not too important at the time of 
> the detach. 
> > At
> the 
> > very least, this is not an item that requires immediate action.
> 
> When you DETACH, CP will issue Rewind-Unload to the drive.  
> There is a special UNIT CHECK that a tape drive is supposed 
> to generate if it is in a NOT READY state when a 
> Rewind-Unload is issued AND there is a cartridge in the 
> drive.  (What?!?  You didn't know that!?!)  :-D
> 
> For a device that isn't really a tape drive, it sure does act 
> like one.
> 
> Alan Altmark
> z/VM Development
> IBM Endicott
> 

Reply via email to