Leap Seconds: GPS, Mobile Networks, and Internet Time
Leap seconds do affect GPS, mobile networks, and internet time, but the impact is managed through various correction mechanisms so that most users never notice. These systems rely on precise timekeeping, and a leap second—an extra second added to Coordinated Universal Time (UTC) to keep it aligned with Earth’s rotation—can cause brief disruptions or require careful handling. This article explains how leap seconds work, why they matter for modern technology, and what happens when one is inserted.
What Is a Leap Second?
A leap second is a one-second adjustment applied to UTC to keep it within 0.9 seconds of astronomical time (UT1), which is based on Earth’s rotation. Earth’s rotation is gradually slowing due to tidal friction, so atomic clocks—which are extremely stable—run slightly faster than the rotating Earth. Without leap seconds, UTC would drift away from solar time, eventually causing noon to occur in the middle of the night over centuries.
The International Earth Rotation and Reference Systems Service (IERS) decides when to add or remove a leap second, typically on June 30 or December 31. Since 1972, 27 leap seconds have been added, the most recent on December 31, 2016. All have been positive (adding a second), because Earth’s rotation has consistently been slower than atomic time.
How GPS Handles Leap Seconds
GPS satellites broadcast their own time scale, called GPS time, which is not adjusted for leap seconds. GPS time was set to match UTC in 1980 and has since diverged by 18 seconds (as of 2024). The GPS system does not insert leap seconds; instead, it broadcasts a parameter called the UTC offset, which tells receivers how many seconds to add to GPS time to obtain UTC.
This design means GPS receivers must know the current offset. The offset is transmitted in the navigation message, but it changes only when a leap second occurs. If a receiver has outdated almanac data, it could report the wrong time until it updates. Most receivers handle this automatically, but older devices or those with poor satellite visibility might show a one-second error temporarily.
Leap Seconds and Mobile Networks
Mobile networks rely on precise synchronization for call handoffs, data transmission, and billing. Many cellular systems use network time protocol (NTP) or precision time protocol (PTP) to align with UTC, which means they must handle leap seconds. When a leap second is inserted, a network that naively adds the second can cause timestamps to repeat, leading to confusion in logs, billing records, or even call drops.
To avoid issues, many network operators use a technique called “smearing,” where the leap second is spread out over a longer period (e.g., 24 hours) by slightly slowing down the clock. This way, the clock never jumps backward or repeats a second, and applications that are not leap-second-aware continue to work. However, smearing means that during the smear period, the network’s time is not exactly UTC, which can affect applications that require strict synchronization with external time sources.
Internet Time and the Leap Second Problem
Internet time synchronization is primarily handled by NTP, which distributes UTC to millions of devices. When a leap second occurs, NTP servers must decide how to handle it. The traditional approach is to step the clock backward by one second at the moment of the leap, causing a repeat of the last second of the day. This can confuse applications that assume time always moves forward, leading to errors in logging, transaction ordering, and even security protocols that rely on timestamps.
Major internet companies have developed their own solutions. For example, Google uses “leap smear” where they adjust the clock gradually over a 24-hour window, avoiding the sudden jump. Other companies, like Amazon and Microsoft, have adopted similar approaches. However, not all systems are smeared, so during a leap second, there can be a discrepancy between smeared and non-smeared servers, which can cause issues for applications that compare timestamps across services.
Common Problems and Misconceptions
One common misconception is that leap seconds cause widespread internet outages. While there have been notable incidents—such as the 2012 Reddit outage and the 2015 Cloudflare DNS glitch—these are relatively rare and often due to software that was not designed to handle the extra second. Most systems now have mitigations in place, but the risk remains for any software that assumes time is monotonic.
Another issue is that leap seconds are unpredictable. The Earth’s rotation is not perfectly regular, so leap seconds are announced only about six months in advance. This makes it difficult for systems to plan for them, especially in distributed environments where different components may handle the leap second differently.
Alternatives to Leap Seconds
Because of the technical challenges, there have been proposals to abolish leap seconds. In 2022, the General Conference on Weights and Measures (CGPM) decided to eliminate leap seconds by 2035, allowing UTC to drift from astronomical time by up to a minute over the next century. This would simplify timekeeping for technology, but it would mean that solar time and UTC would gradually diverge, which could affect astronomy and navigation.
Until then, leap seconds remain a reality. For most users, the impact is negligible, but for system administrators and network engineers, handling leap seconds correctly is an ongoing concern. The key is to ensure that time synchronization software is configured to handle leap seconds properly, either by smearing or by stepping, depending on the application’s needs.
Frequently Asked Questions
Do leap seconds affect my phone’s clock?
Your phone typically gets time from the network or GPS, which are designed to handle leap seconds. Most phones will automatically adjust, but you might notice a one-second discrepancy if your phone’s time source is not updated promptly. In practice, this is rarely noticeable.
Why do we need leap seconds at all?
Leap seconds keep UTC aligned with solar time, which is based on Earth’s rotation. Without them, noon would eventually drift to midnight over many centuries, which would be inconvenient for activities that rely on the sun’s position, such as agriculture and navigation.
What is leap second smearing?
Leap second smearing is a technique where the leap second is distributed over a longer period, typically 24 hours, by slightly slowing down the clock. This avoids the sudden jump and prevents errors in systems that assume time is monotonic.
Can a leap second cause a computer crash?
Yes, if software is not designed to handle a repeated second, it can cause errors, including crashes. This happened with some Linux systems in 2012 and with some network equipment in 2016. Most modern software has been updated, but the risk remains for older or poorly maintained systems.
Will leap seconds be removed in the future?
The CGPM has decided to stop adding leap seconds by 2035, but the exact mechanism is still being worked out. Until then, leap seconds will continue, and systems must remain prepared.
How can I protect my systems from leap second issues?
Use time synchronization software that supports leap second smearing, such as certain NTP implementations. Also, ensure that your applications use monotonic clocks for measuring intervals, and avoid relying on wall-clock time for ordering events.
Related
Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.