This is the point where the back office scene looks a bit like this

Boss 1 - Conference call what do we do guys?,
Tech 1 - I think its network issue,
Net 1 - No its an apps issue,
Apps 1 - Its a hardware issue,
Hard 1 - Nope all systems are up and the monitors show no errors.

Minutes and loads of speculation later,

Boss 1 - Can we roll back to previous backup (Drive silo),
Hard 1 - Yes but not everything gets backed up cus of file locks,
Boss 2 - Okay whats the risk,
Arch 1 - If its a corrupted file it may have been backed up that way, or not backup at all because it was perma file locked.
Boss 1 - Other options,
Boss 2 - Restore from tape and roll back, stunned scilence from Tech 1, Net 1, Apps 1, Hard 1 and Arch 1.
Boss 1 - decision, we will restore from tape.

Minutes pass, Hard 1 recovers from tape and gives Boss 1 and 2, 15 minute updates.

Restore completes (with errors)

Tech 1 - Best we can get.
Hard 2 - Emerges from comms room all back up.
Boss 1 - What did you do, I rebooted the server.

Sigh! of relief all round, post incident review schedule, Hard 2 gets a $10 bonus certficate.

3 months later Post incident review dries up due to lack of interest. And incident is doomed to repeat itself.

And that ladies and gentlemen is how it works and its called best practice.

Just once in a while cant we just check the wheel nuts before we drive off.