349ae4118b
contributions from Peter Jeremy, Poul-Henning Kamp, Ian Lepore, Xin LI, and Warner Losh. Requested by Bartek Rutkowski.
118 lines
4.8 KiB
XML
118 lines
4.8 KiB
XML
<?xml version="1.0" encoding="iso-8859-1"?>
|
|
<!DOCTYPE article PUBLIC "-//FreeBSD//DTD DocBook XML V5.0-Based Extension//EN"
|
|
"http://www.FreeBSD.org/XML/share/xml/freebsd50.dtd">
|
|
|
|
<article xmlns="http://docbook.org/ns/docbook"
|
|
xmlns:xlink="http://www.w3.org/1999/xlink" version="5.0"
|
|
xml:id="leapseconds">
|
|
|
|
<info>
|
|
<title>&os; Support for Leap Seconds</title>
|
|
|
|
<pubdate>$FreeBSD$</pubdate>
|
|
</info>
|
|
|
|
<sect1 xml:id="leapseconds-definition">
|
|
<title>Introduction</title>
|
|
|
|
<para>A <emphasis>leap second</emphasis> is an ad-hoc one-second
|
|
correction to synchronize atomic timescales with Earth rotation.
|
|
This article describes how &os; interacts with leap
|
|
seconds.</para>
|
|
|
|
<para>As of this writing, the next leap second will occur at
|
|
2015-Jun-30 23:59:60. This particular leap second is unusual,
|
|
being in the middle of a business day for North and South
|
|
America and the Asia/Pacific region.</para>
|
|
|
|
<para>Leap seconds are announced by <link
|
|
xlink:href="http://datacenter.iers.org/"><acronym>IERS</acronym></link>
|
|
on <link
|
|
xlink:href="http://datacenter.iers.org/web/guest/bulletins/-/somos/5Rgv/product/16">Bulletin
|
|
C</link>.</para>
|
|
|
|
<para>Standard leap second behavior is described in <link
|
|
xlink:href="https://tools.ietf.org/html/rfc7164#section-3">RFC
|
|
7164</link>. Also see &man.time2posix.3;.</para>
|
|
</sect1>
|
|
|
|
<sect1 xml:id="leapseconds-posix">
|
|
<title>Default Leap Second Handling on &os;</title>
|
|
|
|
<para>The easiest way to handle leap seconds is with the
|
|
<acronym>POSIX</acronym> time rules &os; uses by default,
|
|
combined with <link
|
|
xlink:href="&url.books.handbook;/network-ntp.html"><acronym>NTP</acronym></link>.
|
|
When &man.ntpd.8; is running and the time is synchronized with
|
|
upstream <acronym>NTP</acronym> servers that handle leap seconds
|
|
correctly, the leap second will cause the system time to
|
|
automatically repeat the last second of the day. No other
|
|
adjustments are necessary.</para>
|
|
|
|
<para>If the upstream <acronym>NTP</acronym> servers do not handle
|
|
leap seconds correctly, &man.ntpd.8; will step the time by one
|
|
second after the errant upstream server has noticed and stepped
|
|
itself.</para>
|
|
|
|
<para>If <acronym>NTP</acronym> is not being used, manual
|
|
adjustment of the system clock will be required after the leap
|
|
second has passed.</para>
|
|
</sect1>
|
|
|
|
<sect1 xml:id="leapseconds-cautions">
|
|
<title>Cautions</title>
|
|
|
|
<para>Leap seconds are inserted at the same instant all over the
|
|
world: <acronym>UTC</acronym> midnight. In Japan that is
|
|
mid-morning, in the Pacific mid-day, in the Americas late
|
|
afternoon, and in Europe at night.</para>
|
|
|
|
<para>We believe and expect that &os;, if provided correct and
|
|
stable <acronym>NTP</acronym> service, will work as designed
|
|
during this leap second, as it did during the previous
|
|
ones.</para>
|
|
|
|
<para>However, we caution that practically no applications have
|
|
ever asked the kernel about leap seconds. Our experience is
|
|
that, as designed, leap seconds are essentially a replay of the
|
|
second before the leap second, and this is a surprise to most
|
|
application programmers.</para>
|
|
|
|
<para>Other operating systems and other computers may or may not
|
|
handle the leap-second the same way as &os;, and systems without
|
|
correct and stable <acronym>NTP</acronym> service will not know
|
|
anything about leap seconds at all.</para>
|
|
|
|
<para>It is not unheard of for computers to crash because of
|
|
leap seconds, and experience has shown that a large fraction
|
|
of all public <acronym>NTP</acronym> servers might handle and
|
|
announce the leap second incorrectly.</para>
|
|
|
|
<para>Please try to make sure nothing horrible happens because of
|
|
the leap second.</para>
|
|
</sect1>
|
|
|
|
<sect1 xml:id="leapseconds-testing">
|
|
<title>Testing</title>
|
|
|
|
<para>It is possible to test whether a leap second will be used.
|
|
Due to the nature of <acronym>NTP</acronym>, the test might work
|
|
up to 24 hours before the leap second. Some major reference
|
|
clock sources only announce leap seconds one hour ahead of the
|
|
event. Query the <acronym>NTP</acronym> daemon:</para>
|
|
|
|
<screen>&prompt.user; <userinput>ntpq -c 'rv 0 leap'</userinput></screen>
|
|
|
|
<para>Output that includes <literal>leap_add_sec</literal>
|
|
indicates proper support of the leap second. Before the 24 hour
|
|
period before the leap second, or after the leap second has
|
|
passed, <literal>leap_none</literal> will be reported.</para>
|
|
</sect1>
|
|
|
|
<sect1 xml:id="leapseconds-conclusion">
|
|
<title>Conclusion</title>
|
|
|
|
<para>We hope this discussion of leap seconds on &os; has been of
|
|
use. Now let us never speak of it again.</para>
|
|
</sect1>
|
|
</article>
|