The outage lasted forty-two minutes.
Short enough to dismiss.
Long enough to expose everything.
A network device failed. Redundancy didn’t engage. Traffic stalled. Phones lit up.
“What’s the failover plan?” someone asked.
Silence.
They restored service manually. Systems recovered. Users moved on.
Leadership didn’t.
“We have backups,” someone said.
“That’s not what I asked,” an executive replied.
Microsoft had been emphasizing business continuity heavily by then. Not just backups—recovery objectives. Time mattered now. Expectations were higher.
They asked the hard questions.
How long until full recovery?
What systems come back first?
Who decides?
No one liked the answers.
Recovery existed—but only technically. Not procedurally. Not decisively.
In March, disaster recovery stopped being a document and started becoming a rehearsal.
Roles assigned. Priorities ranked. Recovery timed.
The outage never happened again.
That wasn’t the point.
The point was knowing what would happen if it did.