TL;DR
A prolonged utility power outage at NIST’s Boulder campus disrupted the institute’s atomic time scale and prompted an attempt to disable some Network Time Protocol (NTP) servers to avoid disseminating incorrect time. NIST later reported the site’s time services were restored and that any deviation during the event was likely no more than a few microseconds.
What happened
NIST’s atomic ensemble time scale at the Boulder campus failed after a prolonged utility power outage tied to severe weather, according to a post from a supervisory physicist responsible for the clocks. The outage affected the authority NIST provides for Network Time Protocol services; a staffer said they would attempt to disable the Boulder NTP servers to prevent spreading inaccurate time, but site access was restricted to emergency personnel due to storm conditions. Backup generators and batteries supplied some power — NIST later reported the time servers ran on battery power for about two hours — and the institute’s status page showed a clock error of under 4.8 microseconds. NIST warned users such as telecommunications and aerospace organizations to use alternate time sources during the incident. An updated statement from NIST said Boulder time services were again available and keeping good time, and that any deviation during the outage likely did not exceed approximately five microseconds.
Why it matters
- NTP is a foundational timing service many systems use to synchronize events; incorrect time can cause authentication and operational problems.
- Even tiny timing deviations matter for high-precision applications in science, telecommunications, and positioning systems.
- Best practice for NTP usage is to configure multiple, redundant time sources so clients can fail over when one source is compromised.
- Users who rely exclusively on a single NIST feed — in this case the Boulder facility — risk receiving degraded time without redundant safeguards.
Key facts
- The atomic ensemble time scale at NIST’s Boulder campus failed after a prolonged utility power outage.
- Jeffrey Sherman, a NIST supervisory physicist, reported the failure and said he intended to disable Boulder NTP servers to avoid disseminating incorrect time.
- Storm damage restricted access to the site; only emergency personnel were allowed on campus during the event.
- NIST’s status page indicated a clock error of under 4.8 microseconds during the outage.
- NIST later said servers ran on battery power for about two hours and that the disseminated UTC(NIST) signal likely did not deviate by more than about five microseconds.
- Xcel Energy, the local utility, attributed the outage to strong winds and expected most customers to have power restored within a few hours.
- NIST warned telecommunications and aerospace users to rely on other sources of time information during the incident.
- An update from NIST reported that all time services delivered from the Boulder campus were again available and keeping good time.
What to watch next
- Whether NIST publishes a formal root-cause analysis or confirmation about the reported generator failure (not confirmed in the source).
- Whether any organizations that relied solely on the Boulder NTP feeds experienced authentication or operational outages as a result (not confirmed in the source).
- NIST status pages and advisories for any further alerts or guidance about redundancy and hardening of time services.
Quick glossary
- NIST: The U.S. National Institute of Standards and Technology, a federal agency that maintains standards including official time references.
- Network Time Protocol (NTP): A protocol used to synchronize clocks of computer systems over packet-switched, variable-latency networks.
- UTC(NIST): The official Coordinated Universal Time realization provided by NIST, distributed for use in networks and systems.
- Microsecond: One millionth of a second (1 µs); a unit used to describe very small timing deviations.
Reader FAQ
Did NIST shut off the Boulder NTP servers?
A staffer said they would attempt to disable the servers to avoid disseminating incorrect time, but access to the site was limited by severe weather; final status was not provided as an explicit shutdown in the source.
How large was the timing error during the outage?
NIST’s status page reported a clock error under 4.8 microseconds, and an update assessed any deviation likely did not exceed about five microseconds.
Were services restored?
Yes. NIST later reported that all time services from the Boulder campus were again available and keeping good time.
Did the outage cause widespread system failures?
Not confirmed in the source.
SECURITY 80 NIST contemplated pulling the pin on NTP servers after blackout caused atomic clock drift Time signals shifted by a tiny amount that only very sensitive users would find…
Sources
- NIST contemplated pulling the pin on NTP servers after blackout caused atomic clock drift
- NIST warns of NTP inaccuracy after blackouts across …
- NIST warns several of its Internet Time Service servers may …
- NIST tried to pull the pin on NTP servers after blackout …
Related posts
- About 1,000 systems hit in ransomware attack on Romanian water agency
- Ex-fugitive ‘wannabe MI6’ conman Mark Acklom must repay £125,000 to victim
- Anna’s Archive says it scraped 86M Spotify tracks to build a music archive