Today I royally fucked up at work. Commiserate with me

[QUOTE=Typo Knig]
I crashed a mainframe. That is not supposed to be possible, but I did. Um, acheivement unlocked??
[/QUOTE]

We still have some Thou Shalt Not commands.

TSS LIST(ALL) DATA(ALL) WAIT(Y) comes to mind. For the mainframe-unaware, that would list every detail of every ID on every machine, region and environment. Basically, this will thrash the DASD (disks) and tape silos pretty close to death, and keep the processors very, very busy looking up hundreds of thousands of IDs, applications and resources. :eek:

You can also get into a lot of trouble with a command called UNDERCUT. Its use is restricted to certain overnight hours, just in case something goes wrong. I recently needed to use it, and it was a week and a half long process just to get its use approved and scheduled.

This didn’t happen to me, but I was in the vicinity. Early January, and a mainframe implementation has gone horribly wrong and one of the main production databases is having incorrect transactions applied. Disaster. OK, so we take offline most of the production systems, halt all the scheduled batch updates, fix the dodgy code (it was super obvious what was wrong), roll back the database a couple of hours to a correct state, reapply the transactions and then bring production back up.

This is all standard stuff, a bit stressful with a couple of thousand folk unable to do their job, but hey-ho. The problem here was that the most senior developer available (a total guru type) committed the database rollback with a timestamp of “a couple of hours and one year” in the past.

That was fun.

Reported.