NSA’s classified network goes down for hours and the explanation arrives after the leak
Six anonymous sources told CNN that NSANet was offline on Tuesday; the agency’s on-the-record account of “scheduled routine maintenance” followed the story rather than preceded it.
NSANet, the National Security Agency’s internal classified communications and information-sharing network, went offline on Tuesday 17 November. Bloomberg reported it had been down “since at least early Tuesday morning”, citing a US official and one other person familiar with the matter. The agency staggered staff arrivals at Fort Meade; some personnel could not reach workplace systems. CNN broke the story citing six people familiar with the matter and reported no evidence of a cyberattack. CBS News spoke to multiple US officials, two of whom said there was no evidence of an attack and two of whom attributed the failure to routine system patching; the Wall Street Journal, per the Jerusalem Post’s write-up, linked it to a software update. On the record, an NSA spokesperson said the agency experienced a “temporary degradation during scheduled routine maintenance,” and that “while some internal systems were affected, mission/critical capabilities remain fully operational and our ability to defend National Security continues.” A current and a former US official told CNN an NSANet outage would not affect the ability to pass urgent intelligence to the President.
NSANet is the fourth entry in an eight-day cluster of American infrastructure failures, each officially non-malicious: the 13 November FAA nationwide ground stop, a Microsoft outage, a commercial bank’s online banking, and now the NSA’s core classified network. The FAA has published the most specific technical account of any of them. Its 15 November newsroom Q&A states that “the FAA’s legacy equipment became unstable when data began flooding the FAA FTI network, at a rate of 1,000 connection attempts every 4 to 5 seconds,” disrupting aircraft communications and radar feeds. Initial investigation identified two candidate sources — a radar site in Amarillo, Texas, and a radio communications site in El Dorado, Arkansas — and the broadcasting function was disabled on all new routers as a precaution. Root cause was still listed as under investigation as of 17 November. Transportation Secretary Sean Duffy blamed old air traffic control technology and overloaded routers; CNN’s live file characterised some controllers as not buying the explanation, without naming one.
Assessment: Al Jazeera’s sequencing is the detail that matters for a reader outside Washington: the maintenance explanation was issued after the leak, not before it. That does not make it false — patching is the most mundane explanation available and two CBS sources support it — but an agency that explains itself only once reported on has surrendered the ability to be believed on the next one. The real exposure is the cluster. Four separate non-malicious failures inside eight days is exactly the pattern that invites retroactive attribution, and every party with an interest — hawks in Washington, hacktivist branders in Tehran and Baghdad — gains by linking them. Watch who starts, and with what evidence. Meanwhile a pre-existing viral hoax predicting an internet shutdown on 17 November collided with a genuine outage week, generating enough traffic to spawn dedicated debunk pages. That is the mechanism: a fabricated calendar finding real-world corroboration by coincidence.