At the Technion in Haifa I mentioned the Multics Cookie Monster to some CS 
students, they asked whether it could be done on TSO and I explained how.

There was a local tradition of celebrating April 1, so I deployed the code. 
Most people took it in the context of the day, but one professor was not amused.

--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
עַם יִשְׂרָאֵל חַי
נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר



________________________________________
From: IBM Mainframe Discussion List <[email protected]> on behalf of 
Gabe Goldberg <[email protected]>
Sent: Sunday, March 9, 2025 1:56 AM
To: [email protected]
Subject: Re: Stupid outages you caused

External Message: Use Caution


I've resisted chiming in with these VM stories. But resistance is futile.

First one wasn't directly my fault, but I wrote the service machine that
someone misused to cause an outage. Semi-early days of VM, I wrote an
automatic operator -- cleverly named AUTOOP. It executed time-of-day
tasks, allowed access to project media (tapes, disks) off hours, did a
few other things including (via an undocumented feature) sharing what
the evening's take-out food would be, fetched by the second shift
operator while I watched the system console. But I digress. Another
function was executing privileged system commands on behalf of
designated -- but themselves non-privileged -- users. In use at a small
VM-oriented software firm (VM Systems Group, not that other, much larger
and similarly named place), the CEO (former data center manager where
I'd developed AUTOOP) for no sensible reason sent a SHUTDOWN command to
AUTOOP. Being good software, it obeyed. When the system was brought up
again, the system operator's Profile Exec ran, including an AUTOLOG
command to start AUTOOP. Which found, in its command queue, the
processed but undeleted SHUTDOWN command. There being no way to
interrupt that elegant loop, the system was cold started, losing the
SHUTDOWN command, and everything else in system Spool including plenty
good stuff. But it was a toy computer running in a toy company, so not
much real damage.

Next was my fault. From early days, VM commemorated each two user CPU
seconds used with the "blip character". Initially a twitch of Selectric
type ball, it later became a repeating word on 3270 screens, marching
down from top. Wanting to entertain a colleague, I used privileged CP
command STCP (Store into CP real memory)  to change his blip character,
so it would be like a Burmashave sign on his terminal. I got it wrong,
severely annoying CP. When system logo appeared on my screen, I told the
operator I'd fill out the outage report.

Next, also mine. Very early time when VM had gone production, after a
live test period, I was at home accessing the system using Silent 700
terminal. Maybe the operator annoyed me or maybe I played a joke -- I
shut the system down, then remembered -- we're live with users. It was
early enough with VM for us -- and evening -- so I'm not sure anyone cared.

Not an outage, but a practical joke -- which surely deserves and will
likely get its own thread here. Again, early in my site's VM usage, I
rigged my manager's CMS Profile Exec to log her off every other time she
logged on. She was a good sport so much merriment ensued as she tried to
diagnose and demonstrate the problem.

--
Gabriel Goldberg, Computers and Publishing, Inc.  [email protected]
3401 Silver Maple Place, Falls Church, VA 22042      (703) 204-0433
LinkedIn: http://www.linkedin.com/in/gabegold

----------------------------------------------------------------------
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