Not my booboo, but one that was too easy to commit at my current employer: We use ODDS for console automation and making our lives simpler. ODDS lets you define commands that you simply enter at the console to trigger scripts. We have one named SYSDRAIN that does what you'd expect: it shuts down everything cleanly in preparation for an IPL.
SYSDRAIN now, as its first action, pops a WTO with the SMFID of the system and asks you to enter it for verification. Before that was added, it was far too easy to shut down the wrong LPAR... On Thu, Mar 6, 2025 at 7:52 AM roscoe5 < [email protected]> wrote: > I brought down almost our entire network with a vacuum cleaner. IBM had > DROP-shipped a remote 3174 for our DR exercise. It came off the truck on > its side, busted, but still worked later that day. As I was vacuuming > styrofoam out of the diskette drive, I tripped a circuit breaker. Now, this > was in a solid data center w redundant power grid and UPS DPUs. No power > issues were likely! But miscellaneous items, like workstations, monitors, > and the single Token Ring connection box going on to the 3720 and the > mainframe had just a single power cord. > > Needless to say, we built some redundancy into that “weakest link”. 😅 > > Sent from [Proton Mail](https://proton.me/mail/home) for iOS > > On Thu, Mar 6, 2025 at 8:24 AM, [email protected] <[ > [email protected]](mailto:On Thu, Mar 6, > 2025 at 8:24 AM, [email protected] <<a href=)> wrote: > > > After more than 20 years working as a RACF administrator, a couple of > years ago I achieved my own horror story:I deleted the active RACF > databases (BOTH, PRIMARY and BACKUP), in a peak hour of the day...Details: > > We did recently migrate to a new z/OS version, and my intention was to > delete the old RACF databases residing in the old volumes (their names were > the same as the active ones). I listed them in 3.4, specifying VOLUME > (because they were NOT cataloged), and then issued a DEL line command. As a > result, I deleted the active ones!To be honest, I wasn't aware that a > cataloged version on another volume may be deleted. A beginner mistake. > Carefully looking again at what happened, I noticed that the DEL command > issues the following caution alert (I didn't pay attention to it):<<< > > CAUTION: > > If TSO delete command was issued against an uncataloged data set, a > > cataloged version on a volume other than the one listed here may be > > deleted. > >>>>>> > > It was a tough day... > > Juan G. Mautalen > > > > ---------------------------------------------------------------------- > > 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 > -- Jay Maynard ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
