At the Chicago SHARE 43 meeting in August, 1974, IBM had invited
attendees to a demonstration of the new MVS operating system on the 145
at the Chicago Ed center.  At 8 p.m. Tuesday, three attendees found the
IBM demonstrator on duty was enthralling an attractive lady with his
expertise.  Not wanting to be interrupted by three males, he said, "Go
play with the system on that user TSO terminal over there -- you can
find out how good the security of an MVS system is for a typical TSO
user”.  The challenge was accepted, and in short order, "Tim W." observed
that SYS1.NUCLEUS was not protected, so he scratched it.  "Tim W." knew
that once the system is up, the dataset SYS1.NUCLEUS is not read again,
so the SHARE demonstration continued without flaw.  It was later heard
that IBM took the SHARE MVS demonstration down at 11 p.m. to IPL for a
customer benchmark; it took until 3 a.m. for IBM to find a CE who could
correctly decipher the wait state code and explain that the IPLs kept
failing because there was no SYS1.NUCLEUS on the IPL volume.


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Tony Harminc
Sent: Thursday, November 12, 2015 9:43 AM
To: [email protected]
Subject: Re: IPL wait state

On 12 November 2015 at 05:09, Nathan Astle <[email protected]> wrote:
> i got the below message in the HMC,
>
> Central processor (CP) 0 is looping due to switching between program 
> status words (PSWs) that are not valid. The program status word (PSW) 
> is 0000000000000000.
>
> The IPLTEXT was not written to the newly cloned RES Volume. Could that 
> be a reason ?

Yes. But there are many other possible causes; anything that overlays the 
program new PSW in low storage can provoke it. If another processor is 
available to z/OS it should be able to stop the loop, but not always.

> I am not getting a correct explanation about x'000' on my search about 
> the above PSW in MVS manual codes.

If it's not a wait-state PSW, there will be no wait state code present. More 
exactly, if it's not a disabled wait-state PSW, there is no wait state code 
documented in the MVS System Codes because this is not an error situation; it's 
just the system with no work to do. In this case of an enabled wait, the PSW 
address will usually be 0, though there is no architectural reason it couldn't 
be anything at all that the system wants to leave there..

In this case almost certainly the system is in that strange condition described 
in the Principles of Operation where the program new PSW is invalid (and 
subject to early detection). When this PSW is loaded as the result of a program 
interrupt, another (specification) program interrupt immediately becomes 
pending. This interrupt has a higher priority than almost anything else, and so 
hitting STOP or even RESTART will not stop the loop.

I imagine that on modern machines the HMC warns when this situation occurs, 
just as VM has for decades.

Tony H.

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

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

Reply via email to