Internet-Draft | Timestamps | February 2021 |
Sharma | Expires 25 August 2021 | [Page] |
This document defines a date and time format for use in Internet protocols that is a profile of the ISO 8601 standard for representation of dates and times using the proleptic Gregorian calendar.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 25 August 2021.¶
Copyright (c) 2021 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.¶
Date and time formats cause a lot of confusion and interoperability problems on the Internet. This document addresses many of the problems encountered and makes recommendations to improve consistency and interoperability when representing and using date and time in Internet protocols.¶
This document includes an Internet profile of the [ISO8601] standard for representation of dates and times using the proleptic Gregorian calendar.¶
There are many ways in which date and time values might appear in Internet protocols: this document focuses on just one common usage, viz. timestamps for Internet protocol events. This limited consideration has the following consequences:¶
All dates and times are assumed to be in the "current era", somewhere between 0000AD and 9999AD.¶
All times expressed have a stated relationship (offset) to Coordinated Universal Time (UTC). (This is distinct from some usage in scheduling applications where a local time and location may be known, but the actual relationship to UTC may be dependent on the unknown or unknowable actions of politicians or administrators. The UTC time corresponding to 17:00 on 23rd March 2005 in New York may depend on administrative decisions about daylight savings time. This specification steers well clear of such considerations.)¶
Timestamps can express times that occurred before the introduction of UTC. Such timestamps are expressed relative to universal time, using the best available practice at the stated time.¶
Date and time expressions indicate an instant in time. Description of time periods, or intervals, is not covered here.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119].¶
Coordinated Universal Time as maintained since 1988 by the Bureau International des Poids et Mesures (BIPM) in conjunction with leap seconds as announced by the International Earth Rotation and Reference Frames Service [IERS]. From 1972 through 1987 UTC was maintained entirely by Bureau International de l'Heure (BIH). Before 1972 UTC was not generally recognized and civil time was determined by individual jurisdictions using different techniques for attempting to follow Universal Time based on measuring the rotation of the earth.¶
The unit of time in the International System of Units. Since Resolution 1 of the 13th CGPM on 1967-10-13 [CGPM] the second is defined as the duration of 9,192,631,770 cycles of microwave radiation absorbed or emitted by the hyperfine transition of cesium-133 atoms in their ground state undisturbed by external fields, but this definition was not in practical use for civil time until 1972-01-01. Prior to 1972-01-01 civil time was based on Universal Time which was measured by observations of the rotation of the earth, and the practical definition of the second was 1/86400 of the mean solar day.¶
A period of time of 60 seconds. However, see also the restrictions in section Section 5.7 and Appendix C for how leap seconds are denoted within minutes.¶
A period of time of 60 minutes.¶
Starting 1972-01-01 a duration of 86400 SI seconds for the UTC time scale. In other contexts the duration of one mean solar day as agreed internationally by the 1884 International Meridian Conference and measured using Universal Time.¶
In the proleptic Gregorian calendar, a year which has 366 days. A leap year is a year whose number is divisible by four an integral number of times, except that if it is a centennial year (i.e. divisible by one hundred) it shall also be divisible by four hundred an integral number of times.¶
The date/time format defined in section 5 of this document.¶
This term is used in this document to refer to an unambiguous representation of some instant in time.¶
A suffix which, when applied to a time, denotes a UTC offset of 00:00; often spoken "Zulu" from the ICAO phonetic alphabet representation of the letter "Z".¶
The use of 2 (and 3) digit years was allowed but deprecated in [RFC3339], the predecessor of this document.¶
The use of such a format is no longer allowed, and implementations should use either a standard 4-digit year or the extended 6-digit value with a sign.¶
Because the daylight saving rules for local time zones are so convoluted and can change based on local law at unpredictable times, true interoperability is best achieved by using Coordinated Universal Time (UTC). This specification does not cater to local time zone rules.¶
The offset between local time and UTC is often useful information. For example, in electronic mail ([RFC2822]) the local offset provides a useful heuristic to determine the probability of a prompt response. Attempts to label local offsets with alphabetic strings have resulted in poor interoperability in the past [RFC1123]. As a result, [RFC2822] has made numeric offsets mandatory.¶
Numeric offsets are calculated as "local time minus UTC". So the
equivalent time in UTC can be determined by subtracting the offset
from the local time. For example, 18:50:00-04:00
is the same time as
22:50:00Z
. (This example shows negative offsets handled by adding
the absolute value of the offset.)¶
Numeric offsets may differ from UTC by any number of seconds, or even a
fraction of seconds. This can be easily represented by including an
optional seconds value in the offset, which may further optionally include
a fraction of seconds behind a decimal point, for example +12:34:56.789
.
This is especially useful in the case of certain historical time zones.¶
If the time in UTC is known, but the offset to local time is unknown, this can be represented with an offset of "-00:00". This differs semantically from an offset of "Z" or "+00:00", which imply that UTC is the preferred reference point for the specified time. RFC2822 [RFC2822] describes a similar convention for email.¶
A number of devices currently connected to the Internet run their internal clocks in local time and are unaware of UTC. While the Internet does have a tradition of accepting reality when creating specifications, this should not be done at the expense of interoperability. Since interpretation of an unqualified local time zone will fail in approximately 23/24 of the globe, the interoperability problems of unqualified local time are deemed unacceptable for the Internet. Systems that are configured with a local time, are unaware of the corresponding UTC offset, and depend on time synchronization with other Internet systems, MUST use a mechanism that ensures correct synchronization with UTC. Some suitable mechanisms are:¶
This section discusses desirable qualities of date and time formats and defines a profile of ISO 8601 for use in Internet protocols.¶
If date and time components are ordered from least precise to most
precise, then a useful property is achieved. Assuming that the time
zones of the dates and times are the same (e.g., all in UTC),
expressed using the same string (e.g., all "Z" or all "+00:00"), and all
times have the same number of fractional second digits then the date and
time strings may be sorted as strings (e.g., using the strcmp()
function
in C) and a time-ordered sequence will result. The presence of optional
punctuation would violate this characteristic.¶
Human readability has proved to be a valuable feature of Internet
protocols. Human readable protocols greatly reduce the costs of
debugging since telnet often suffices as a test client and network
analyzers need not be modified with knowledge of the protocol. On
the other hand, human readability sometimes results in
interoperability problems. For example, the date format "10/11/1996"
is completely unsuitable for global interchange because it is
interpreted differently in different countries. In addition, the
date format in (RFC822) has resulted in interoperability problems when
people assumed any text string was permitted and translated the three
letter abbreviations to other languages or substituted date formats
which were easier to generate (e.g. the format used by the C function
ctime
). For this reason, a balance must be struck between human
readability and interoperability.¶
Because no date and time format is readable according to the conventions of all countries, Internet clients SHOULD be prepared to transform dates into a display format suitable for the locality. This may include translating UTC to local time.¶
A format which includes rarely used options is likely to cause interoperability problems. This is because rarely used options are less likely to be used in alpha or beta testing, so bugs in parsing are less likely to be discovered. Rarely used options should be made mandatory or omitted for the sake of interoperability whenever possible.¶
If a date/time format includes redundant information, that introduces the possibility that the redundant information will not correlate. For example, including the day of the week in a date/time format introduces the possibility that the day of week is incorrect but the date is correct, or vice versa. Since it is not difficult to compute the day of week from a date (see Appendix A), the day of week should not be included in a date/time format.¶
The complete set of date and time formats specified in ISO 8601 [ISO8601] is quite complex in an attempt to provide multiple representations and partial representations. Internet protocols have somewhat different requirements and simplicity has proved to be an important characteristic. In addition, Internet protocols usually need complete specification of data in order to achieve true interoperability. Therefore, the complete grammar for ISO 8601 is deemed too complex for most Internet protocols.¶
The following section defines a profile of ISO 8601 for use on the Internet. It is a conformant subset of the ISO 8601 extended format. Simplicity is achieved by making most fields and punctuation mandatory.¶
The following profile of [ISO8601] dates SHOULD be used in new protocols on the Internet. This is specified using the syntax description notation defined in [RFC2234].¶
This date/time format may be used in some environments or contexts that distinguish between the upper- and lower-case letters 'A'-'Z' and 'a'-'z' (e.g. XML). Specifications that use this format in such environments MAY further limit the date/time syntax so that the letters 'T' and 'Z' used in the date/time syntax must always be upper case. Applications that generate this format SHOULD use upper case letters.¶
The grammar element date-mday represents the day number within the current month. The maximum value varies based on the month and year as follows:¶
Month Number | Month/Year | Maximum value of date-mday |
---|---|---|
01 | January | 31 |
02 | February, normal | 28 |
02 | February, leap year | 29 |
03 | March | 31 |
04 | April | 30 |
05 | May | 31 |
06 | June | 30 |
07 | July | 31 |
08 | August | 31 |
09 | September | 30 |
10 | October | 31 |
11 | November | 30 |
12 | December | 31 |
Appendix B contains sample C code to determine if a year is a leap year.¶
The grammar element time-second may have the value "60" at the end of months in which a leap second occurs - to date: June (XXXX-06- 30T23:59:60Z) or December (XXXX-12-31T23:59:60Z); see Appendix C for a table of leap seconds. It is also possible for a leap second to be subtracted, at which times the maximum value of time-second is "58". At all other times the maximum value of time-second is "59". Further, in time zones other than "Z", the leap second point is shifted by the zone offset (so it happens at the same instant around the globe).¶
Leap seconds cannot be predicted far into the future. The International Earth Rotation Service publishes bulletins [IERS] that announce leap seconds with a few weeks' warning. Applications should not generate timestamps involving inserted leap seconds until after the leap seconds are announced.¶
Although ISO 8601 permits the hour to be "24", this profile of ISO 8601 only allows values between "00" and "23" for the hour in order to reduce confusion.¶
Here are some examples of Internet date/time format.¶
This represents 20 minutes and 50.52 seconds after the 23rd hour of April 12th, 1985 in UTC.¶
This represents the same instant as the previous example but with the expanded 6-digit year format.¶
This represents 39 minutes and 57 seconds after the 16th hour of December 19th, 1996 with an offset of -08:00 from UTC (Pacific Standard Time). Note that this is equivalent to 1996-12-20T00:39:57Z in UTC.¶
This represents the leap second inserted at the end of 1990.¶
This represents the same leap second in Pacific Standard Time, 8 hours behind UTC.¶
This represents the same instant of time as noon, January 1, 1937, Netherlands time. Standard time in the Netherlands was exactly 19 minutes and 32.13 seconds ahead of UTC by law from 1909-05-01 through 1937-06-30.¶
Since the local time zone of a site may be useful for determining a time when systems are less likely to be monitored and might be more susceptible to a security probe, some sites may wish to emit times in UTC only. Others might consider this to be loss of useful functionality at the hands of paranoia.¶
The following is a sample C subroutine loosely based on Zeller's Congruence [ZELLER] which may be used to obtain the day of the week for dates on or after 0000-03-01:¶
Here is a sample C subroutine to calculate if a year is a leap year:¶
In 1970 CCIR Recommendation 460 produced international agreement that starting on 1972-01-01 radio broadcast time signals should provide SI seconds with occasional leaps of 1 SI second as necessary to agree with Universal Time. The time scale in radio broadcasts became known as UTC, and the current version of that recommendation is [ITU-R-TF]. Since 1988 IERS has the responsibility for announcing when leap seconds will be introduced into UTC. Further information about leap seconds can be found at the US Navy Oceanography Portal. In particular, it notes that:¶
The decision to introduce a leap second in UTC is the responsibility of the International Earth Rotation Service [IERS]. According to the CCIR Recommendation, first preference is given to the opportunities at the end of December and June, and second preference to those at the end of March and September.¶
When required, insertion of a leap second occurs as an extra second at the end of a day in UTC, represented by a timestamp of the form YYYY-MM-DDT23:59:60Z. A leap second occurs simultaneously in all time zones, so that time zone relationships are not affected. See section Section 5.8 for some examples of leap second times.¶
The following table is an excerpt from the table maintained by the IERS. The source data are located at the Earth Orientation Parameters Product Centre at Observatoire de Paris.¶
For dates after the initial adjustment on 1972-01-01 this table shows the date of the leap second, and the difference between the time scale TAI (which is not adjusted by leap seconds) and UTC after that leap second.¶
UTC Date | TAI - UTC After Leap Second |
---|---|
1972-06-30 | 11 |
1972-12-31 | 12 |
1973-12-31 | 13 |
1974-12-31 | 14 |
1975-12-31 | 15 |
1976-12-31 | 16 |
1977-12-31 | 17 |
1978-12-31 | 18 |
1979-12-31 | 19 |
1981-06-30 | 20 |
1982-06-30 | 21 |
1983-06-30 | 22 |
1985-06-30 | 23 |
1987-12-31 | 24 |
1989-12-31 | 25 |
1990-12-31 | 26 |
1992-06-30 | 27 |
1993-06-30 | 28 |
1994-06-30 | 29 |
1995-12-31 | 30 |
1997-06-30 | 31 |
1998-12-31 | 32 |
2005-12-31 | 33 |
2008-12-31 | 34 |
2012-06-30 | 35 |
2015-06-30 | 36 |
2016-12-31 | 37 |