-----Original Message-----
From: IBM Mainframe Discussion List <[email protected]> On
Behalf Of Seymour J Metz
Sent: Thursday, March 6, 2025 4:38 AM
To: [email protected]
Subject: Re: Stupid outages you caused (was: Cost of an outage)
Is it fair to say that you cause the outage? I would attribute it to the bad
update.
--
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 Dave Gibney <000006fb76de82cb-dmarc-
[email protected]>
Sent: Wednesday, March 5, 2025 8:29 PM
To: [email protected]
Subject: Re: Stupid outages you caused (was: Cost of an outage)
External Message: Use Caution
Very early in my career, I had a Wylbur Exec that dropped a TYPRUN=PRINT
jobcard as line 0.0 and submitted it. At that time, a line printer was in the
same
room as my desk.
Apparently, they updated Wylbur, or the JES2 mods, I never learned which.
One day, I did my print command and Wylbur crashed. It came back up, and I
resumed my work, issued the command again. I probably shouldn't have done
it the third time.
They left it down until they came to my "office" and asked me not to do that.
Apparently, the update did not account for a Wylbur file with line number
zero.
-----Original Message-----
From: IBM Mainframe Discussion List <[email protected]> On
Behalf Of Seymour J Metz
Sent: Wednesday, March 5, 2025 3:42 PM
To: [email protected]
Subject: Re: Stupid outages you caused (was: Cost of an outage)
Welll, this may seem penny ante and not nearly dramatic enough, but I once
type EXEC CMS ERASE when I meant to type ERASE CMS EXEC. It was the
fastest PA1 in the West and a very red face. No permanent damage, and
nobody pointing at me laughing, but I was still embarrassed.
--
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 Phil Smith III <[email protected]>
Sent: Wednesday, March 5, 2025 6:01 PM
To: [email protected]
Subject: Stupid outages you caused (was: Cost of an outage)
External Message: Use Caution
Rupert Reynolds wrote about taking down a system by compressing a PDS.
What stories can y'all share about times you or someone you worked with
took down a system in a way that made you SMH afterward?
I'll start with a couple of VM stories:
Back at University of Waterloo, we had four systems running VM/SP in an
SSI
configuration (think "Sysplex", only less so) with 20,000 students using the
system (among other things). We had a service virtual machine (an SVM;
think
"STC") named PRIV that would accept commands via SMSG (think "TELL"),
validate the issuer and command against a table, and issue the command (or
not) depending on whether they were authorized. This was nice, and had
granularity so, for example, BOB could recycle some SVMs but not others, or
could force off specific users.
I was doing some enhancements to PRIV and logged onto it. Hmm, how to
take it down? I know: SMSG * SHUTDOWN
Then I waited. And waited. And all of a sudden an operator came barreling
out
of the Red Room yelling, "System A just shut itself down?!"
Oops. Nothing I've written since has accepted SHUTDOWN as a command,
so
as not to tempt anyone.
Years later, at my first vendor, I was testing a product for possible
acquisition.
This was in the early days of VM/XA SP, which was notoriously unreliable at
that stage in its development (at one point the service for it overflowed a
tape,
necessitating some quick work on IBM's part because nobody had ever
considered that a possibility).
Because the possible acquisition was a Big Secret, I went down to our
(unstaffed) toy data center to work. I fired up the product and the system
crashed; not unusual for VM/XA SP, so I went over and started bringing it
back
up. About halfway through, the other two developers came down to see if
they needed to do anything. I let them finish the process, and as soon as I
got
a logo on my terminal, I logged back on and fired up the product again. And
it
crashed again instantly. They both turned around and said, "What did you
do?" and I had to come clean! Turned out the product was mucking with low
core, ick.
Last one isn't my fault, from 15 years later. I was at Linuxcare, where we
were
doing Linux provisioning under z/VM. One of our guys was onsite at a bank
doing a trial install and needed some disk space. He was really a Linux guy,
not
a VM guy, but had mucked around on our MP3000, so he [thought he]
knew
what to do: he found a free volume, attached it, and formatted it. Oops:
z/OS
had had plans for that data, and folks were NOT happy when they realized
what he'd done. Of course it was at least partly their fault for having left him
alone on a production system on a privileged ID.
This was on a Friday and I was off that day because I was having knee
surgery.
I got a call late that evening from our CEO saying, "You need to be in Chicago
first thing Monday morning". So early Monday I flew to ORD and took a cab
to
an Embassy Suites and spent the day there working, waiting for a call to go
do...something. Finally I got one late in the day saying "Nevermind, go
home".
I guess they found enough of a backup and didn't want to have to discuss
who
screwed up worse.
What have YOU done that you wouldn't want on your resume?
----------------------------------------------------------------------
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
----------------------------------------------------------------------
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
