The NSA did not decide to go public. Six people decided for it.
The NSA’s classified internal network went down on November 17, and CNN reported it based on six inside sources before the agency said anything. The post argues the NSA never had the option of keeping it quiet until it was fixed. Once tens of thousands of staff had their arrival times staggered and the campus was cut to essential personnel, a leak was going to happen within hours. The agency was right to confirm the outage, because that ended the cyberattack speculation and avoided a denial that
PressContact Blog · 3 minute read

The NSA did not decide to go public. Six people decided for it.
When an outage changes how tens of thousands of people work, the secret is gone before the fix is in. What matters is what you say next.
On November 17, CNN reported that NSA Net, the classified network the National Security Agency uses for internal communications, was down. Six people familiar with the matter were the source. The agency had said nothing.
My first reaction was that this should have stayed quiet until it was fixed. An agency this important should not advertise that its own network is broken while it is still broken.
That reaction assumes the NSA had a choice. It did not.
The choice the NSA did not have
NSA Net connects Fort Meade to the cryptologic centers in Colorado, Georgia, Hawaii and Texas, to the Utah Data Center, and to personnel at combatant commands worldwide. To keep tens of thousands of employees from logging in at once, the agency staggered arrival times. By Tuesday night it had closed the campus to everyone but essential personnel. The New York Times reported the network had been unstable for weeks, and that the workforce was frustrated.
A secret held by that many frustrated people does not last. Once the agency changed when people came to work, a leak was a matter of hours.
The call it got right
So the real decision was narrower: stay silent after the leak, or confirm a little. The NSA confirmed a little. Its statement said it "experienced temporary degradation during scheduled routine maintenance," and that mission-critical capabilities remained fully operational.
That was the right call. It ended the attack theory first. The story broke in the same week as the Microsoft 365 outage, the FAA ground stop and the Chase banking failure, and silence from the NSA that week would have read as a breach. The line about mission systems was aimed at adversaries who already had the news. And denying something six sources had confirmed would have cost more credibility than the outage did.
Where it went wrong
Where the agency went wrong is the wording. "Routine maintenance" does not describe a campus cut to essential staff after weeks of instability. Every new leak makes the phrase look worse, and the leaks kept coming. One source told the Washington Post the agency was in the middle of, or had recently finished, a Microsoft software upgrade. That is the kind of detail that should have stayed inside the building.
The limit on my read: no one outside the NSA knows the cause or how long this will run. The agency may have reasons for its wording that it cannot share.
The lesson
The lesson reaches well past Fort Meade. If an outage changes how your people work, plan as if it is already public. Draft the holding statement before the maintenance window opens. Confirm the disruption, say whether it was an attack, say what is still running, and stop. Keep the details secret until it is fixed. The outage itself will not stay secret.
Sources
1. Highly classified U.S. national security communications network goes down (CNN via WRAL)
2. NSA Net outage and the sites it connects (ClearanceJobs)
3. NSA's highly classified communications network disrupted (Fox News)
4. NSA closes campus to essential personnel (CNN via CTV News)
5. NSANet unstable for weeks (New York Times via Boston Globe)
6. Classified communications network at the NSA goes down (AP via US News)
7. Communications system at top-secret NSA goes down (Washington Post)
8. Microsoft says months-old bug caused global Microsoft 365 outage (BleepingComputer)
9. Nationwide Ground Stop: Questions and Answers (FAA)
10. A technical issue at JPMorgan Chase triggers online banking delays (AP via ABC11)