ntp.conf.5 revision 55895 

 $FreeBSD: head/usr.sbin/ntp/doc/ntp.conf.5 55895 2000-01-13 09:59:55Z sheldonh $

.Dd January 13, 2000
.Dt NTP.CONF 5
.Os
.Sh NAME
.Nm ntp.conf
.Nd Network Time Protocol (NTP) daemon configuration file
.Sh SYNOPSIS
.Nm /etc/ntp.conf
.Sh DESCRIPTION
The
.Nm
configuration file is read at initial startup by the
.Xr ntpd 8
daemon in order to specify the synchronization sources,
modes and other related information.
Usually, it is installed in the

a /etc
directory,
but could be installed elsewhere
(see the daemon's
.Fl c
command line option).

p
The file format is similar to other Unix configuration files.
Comments begin with a
.Qq #
character and extend to the end of the line;
blank lines are ignored.
Configuration commands consist of an initial keyword
followed by a list of arguments,
some of which may be optional, separated by whitespace.
Commands may not be continued over multiple lines.
Arguments may be host names,
host addresses written in numeric, dotted-quad form,
integers, floating point numbers (when specifying times in seconds)
and text strings.

p
The rest of this page describes the configuration and control options.
The
.Qo
Notes on Configuring NTP and Setting up a NTP Subnet
.Qc
page
(available as part of the HTML documentation
provided in

a /usr/share/doc/ntp )
contains an extended discussion of these options.
In addition to the discussion of general
.Sx Configuration Options ,
there are sections describing the following supported functionality
and the options used to control it:
l -bullet -offset indent t .Sx Authentication Support
t .Sx Monitoring Support
t .Sx Access Control Support
t .Sx Reference Clock Support
.El

p
Following these is a section describing
.Sx Miscellaneous Options .
While there is a rich set of options available,
the only required option is one or more
c server , c peer , c broadcast or
c manycastclient commands.
.Ss Configuration Options
Following is a description of the configuration commands in NTPv4.
These commands have the same basic functions as in NTPv3
and in some cases new functions and new operands.
The various modes are determined by the command keyword
and the type of the required IP address.
Addresses are classed by type as
(s) a remote server or peer (IP class A, B and C),
(b) the broadcast address of a local interface,
(m) a multicast address (IP class D),
or (r) a reference clock address (127.127.x.x).
Note that,
while autokey and burst modes are supported by these commands,
their effect in some weird mode combinationscan be meaningless
or even destructive.
l -tag -width indent t Xo Ic peer .Ar address
.Op autokey | key Ar key
.Op burst
.Op version Ar version
.Op prefer
.Op minpoll Ar minpoll
.Op maxpoll Ar maxpoll
.Xc
t Xo Ic server .Ar address
.Op autokey | key Ar key
.Op burst
.Op version Ar version
.Op prefer
.Op minpoll Ar minpoll
.Op maxpoll Ar maxpoll
.Xc
t Xo Ic broadcast .Ar address
.Op autokey | key Ar key
.Op burst
.Op version Ar version
.Op minpoll Ar minpoll
.Op maxpoll Ar maxpoll
.Op ttl Ar ttl
.Xc
t Xo Ic manycastclient .Ar address
.Op autokey | key Ar key
.Op burst
.Op version Ar version
.Op minpoll Ar minpoll
.Op maxpoll Ar maxpoll
.Op ttl Ar ttl
.Xc
These four commands specify the time server name or address
to be used and the mode in which to operate.
The address can be
either a DNS name
or an IP address in dotted-quad notation.
Additional information on association behavior can be found in
the
.Qo
Association Management
.Qc
page.
l -tag -width indent t Ic peer For type s addresses (only),
this operates as the current peer command,
which mobilizes a persistent symmetric-active mode association,
except that additional modes are available.
This command should
.Em not
be used for type b, m or r addresses.

p
The
c peer command specifies that the local server is to operate
in symmetric active mode with the remote server.
In this mode,
the local server can be synchronized to the remote server
and, in addition,
the remote server can be synchronized by the local server.
This is useful in a network of servers where,
depending on various failure scenarios,
either the local or remote server may be the better source of time.
t Ic server For type s and r addresses,
this operates as the NTPv3 server command,
which mobilizes a persistent client mode association.
The server command specifies
that the local server is to operate in client mode
with the specified remote server.
In this mode,
the local server can be synchronized to the remote server,
but the remote server can never be synchronized to the local server.
t Ic broadcast For type b and m addresses (only),
this is operates as the current NTPv3 broadcast command,
which mobilizes a persistent broadcast mode association,
except that additional modes are available.
Multiple commands can be used
to specify multiple local broadcast interfaces (subnets)
and/or multiple multicast groups.
Note that local broadcast messages go only to the interface
associated with the subnet specified,
but multicast messages go to all interfaces.
In the current implementation,
the source address used for these messages
is the Unix host default address.

p
In broadcast mode,
the local server sends periodic broadcast messages
to a client population at the address specified,
which is usually the broadcast address
on (one of) the local network(s)
or a multicast address assigned to NTP.
The IANA has assigned the multicast group address 224.0.1.1
exclusively to NTP,
but other non-conflicting addresses can be used
to contain the messages within administrative boundaries.
Ordinarily, this specification applies
only to the local server operating as a sender;
for operation as a broadcast client,
see the
c broadcastclient or
c multicastclient commands below.
t Ic manycastclient For type m addresses (only),
this mobilizes a manycast client-mode association
for the multicast address specified.
In this case a specific address must be supplied
which matches the address used on the
c manycastserver command for the designated manycast servers.
The NTP multicast address 224.0.1.1 assigned by the IANA should
.Em not
be used,
unless specific means are taken
to avoid spraying large areas of the Internet
with these messages
and causing a possibly massive implosion of replies at the sender.

p
The
c manycastclient command specifies
that the local server is to operate in client mode
with the remote servers
that are discovered as the result of broadcast/multicast messages.
The client broadcasts a request message
to the group address associated with the specified address
and specifically enabled servers respond to these messages.
The client selects the servers providing the best time
and continues as with the
c server command.
The remaining servers are discarded as if never heard.
.El

p
The following options to these commands are available:
l -tag -width indent t autokey All packets sent to the address
are to include authentication fields
encrypted using the autokey scheme.
t burst At each poll interval,
send a burst of eight packets spaced,
instead of the usual one.
t key Ar key All packets sent to the address
are to include authentication fields
encrypted using the specified key identifier,
which is an unsigned 32-bit integer
less than 65536.
The default is to include no encryption field.
t version Ar version Specifies the version number to be used for outgoing NTP packets.
Versions 1-4 are the choices, with version 4 the default.
t prefer Marks the server as preferred.
All other things being equal,
this host will be chosen for synchronization
among a set of correctly operating hosts.
See the
.Qo
Mitigation Rules and the prefer Keyword
.Qc
page
for further information.
t ttl Ar ttl This option is used only with broadcast mode.
It specifies the time-to-live (TTL) to use
on multicast packets.
Selection of the proper value,
which defaults to 127,
is something of a black art
and must be coordinated with the network administrator.
t minpoll Ar minpoll t maxpoll Ar maxpoll These options specify the minimum
and maximum polling intervals for NTP messages,
in seconds to the power of two.
The default range is 6 (64 s) to 10 (1,024 s).
The allowable range is 4 (16 s) to 17 (36.4 h) inclusive.
.El
t Ic broadcastclient This command directs the local server to listen for and respond
to broadcast messages received on any local interface.
Upon hearing a broadcast message for the first time,
the local server measures the nominal network delay
using a brief client/server exchange with the remote server,
then enters the broadcastclient mode,
in which it listens for
and synchronizes to succeeding broadcast messages.
Note that,
in order to avoid accidental or malicious disruption in this mode,
both the local and remote servers should operate
using authentication and the same trusted key and key identifier.
t Xo Ic multicastclient .Op Ar address
.Op ...
.Xc
This command directs the local serverto listen for
multicast messages at the group address(es)
of the global network.
The default address is that assigned by the Numbers Czar
to NTP (224.0.1.1).
This command operates in the same way as the
c broadcastclient command, but uses IP multicasting.
Support for this command requires a multicast kernel.
t Ic driftfile Ar driftfile This command specifies the name of the file used
to record the frequency offset of the local clock oscillator.
If the file exists,
it is read at startup in order to set the initial frequency offset
and then updated once per hour with the current frequency offset
computed by the daemon.
If the file does not exist or this command is not given,
the initial frequency offset is assumed zero.
In this case,
it may take some hours for the frequency to stabilize
and the residual timing errors to subside.

p
The file format consists of a single line
containing a single floating point number,
which records the frequency offset
measured in parts-per-million (PPM).
The file is updated by first writing the current drift value
into a temporary file
and then renaming this file to replace the old version.
This implies that
.Xr ntpd 8
must have write permission for the directory
the drift file is located in,
and that file system links, symbolic or otherwise, should be avoided.
t Xo Ic manycastserver .Ar address
.Op ...
.Xc
This command directs the local server to listen for
and respond to broadcast messages received on any local interface,
and in addition enables the server to respond
to client mode messages to the multicast group address(es)
(type m) specified.
At least one address is required,
but the NTP multicast address 224.0.1.1
assigned by the IANA should
.Em not
be used,
unless specific means are taken to limit the span of the reply
and avoid a possibly massive implosion at the original sender.
t Xo Ic revoke .Op Ar logsec
.Xc
Specifies the interval between recomputations
of the private value used with the autokey feature,
which ordinarily requires an expensive public-key computation.
The default value is 12 (65,536 s or about 18 hours).
For poll intervals above the specified interval,
a new private value will be recomputed for every message sent.
t Xo Ic autokey .Op Ar logsec
.Xc
Specifies the interval between regenerations
of the session key list used with the autokey feature.
Note that the size of the key list for each association
depends on this interval and the current poll interval.
The default value is 12 (4096 s or about 1.1 hours).
For poll intervals above the specified interval,
a session key list with a single entry
will be regenerated for every message sent.
t Xo Ic enable .Op Ar flag
.Op ...
.Xc
t Xo Ic disable .Op Ar flag
.Op ...
.Xc
Provides a way to enable or disable various server options.
Flags not mentioned are unaffected.
Note that all of these flags can be controlled remotely
using the
.Xr ntpdc 8
utility program.
Following is a description of the flags.
l -tag -width indent t auth Enables the server to synchronize with unconfigured peers
only if the peer has been correctly authenticated
using a trusted key and key identifier.
The default for this flag is enable.
t bclient When enabled, this is identical to the broadcastclient
command.
The default for this flag is disable.
t kernel Enables the precision-time kernel support
for the
.Xr ntp_adjtime 2
system call, if implemented.
Ordinarily, support for this routine is detected automatically
when the NTP daemon is compiled,
so it is not necessary for the user to worry about this flag.
It provided primarily so that this support can be disabled
during kernel development.
t monitor Enables the monitoring facility.
See the
c monlist command of the
.Xr ntpdc 8
program
further information.
The default for this flag is enable.
t ntp Enables the server to adjust its local clock by means of NTP.
If disabled,
the local clock free-runs at its intrinsic time and frequency offset.
This flag is useful in case the local clock is controlled
by some other device or protocol and NTP is used
only to provide synchronization to other clients.
In this case,
the local clock driver can be used to provide this function
and also certain time variables for error estimates
and leap-indicators.
See the
.Qo
Reference Clock Drivers
.Qc
page
for further information.
The default for this flag is enable.
t stats Enables the statistics facility.
See the
.Sx Monitoring Options
section
for further information.
The default for this flag is enable.
.El
.El
.Ss Authentication Support
Authentication support allows the NTP client to verify
that the server is in fact known and trusted
and not an intruder intending accidentally
or on purpose to masquerade as that server.
The NTPv3 specification RFC 1305 defines a scheme
which provides cryptographic authentication of received NTP packets.
Originally, this was done using the Data Encryption Standard (DES)
operating in Cipher Block Chaining (CBC) mode,
commonly called DES-CBC.
Subsequently, this was augmented by the RSA Message Digest 5 (MD5)
using a private key, commonly called keyed-MD5.
Either algorithm computes a message digest, or one-way hash,
which can be used to verify the server has the correct private key
and key identifier.
NTPv4 retains this scheme and, in addition,
provides a new autokey scheme based on reverse hashing
and public key cryptography.
Authentication can be configured separately for each association
using the key or autokey subcommands on the
c peer Ns , c server Ns , c broadcast and
c manycastclient commands as described in the
.Sx Configuration Options
section.

p
The authentication options specify the suite of keys,
select the key for each configured association
and manage the configuration operations,
as described below.
The auth flag which controls these functions
can be set or reset by the
c enable and
c disable configuration commands and also by remote configuration commands
sent by a
.Xr ntpdc 8
program running in another machine.
If this flag is set, persistent peer associations
and remote configuration commands are effective
only if cryptographically authenticated.
If this flag is disabled,
these operations are effective
even if not cryptographic authenticated.
It should be understood that operating in the latter mode
invites a significant vulnerability
where a rogue hacker can seriously disrupt client operations.

p
The auth flag affects all authentication procedures described below;
however, it operates differently
if cryptographic support is compiled in the distribution.
If this support is available and the flag is enabled,
then persistent associations are mobilized
and remote configuration commands are effective
only if successfully authenticated.
If the support is unavailable and the flag is enabled,
then it is not possible under any conditions
to mobilize persistent associations
or respond to remote configuration commands.
The auth flag normally defaults to set
if cryptographic support is available and to reset otherwise.

p
With the above vulnerabilities in mind,
it is desirable to set the auth flag in all cases.
One aspect which is often confusing
is the name resolution process
which maps server names in the configuration file to IP addresses.
In order to protect against bogus name server messages,
this process is authenticated
using an internally generated key
which is normally invisible to the user.
However, if cryptographic support is unavailable
and the auth flag is enabled,
the name resolution process will fail.
This can be avoided
either by specifying IP addresses instead of host names,
which is generally inadvisable,
or by leaving the flag disabled
and enabling it once the name resolution process is complete.

p
Following is a description
of the two available cryptographic authentication schemes.
l -tag -width indent t Private Key Scheme The original RFC 1305 specification allows any one of possibly
65,536 keys, each distinguished a 32-bit key identifier,
to authenticate an association.
The servers involved must agree on the key
and key identifier to authenticate their messages.
Keys and related information are specified in a key file,
usually called
.Xr ntp.keys 5
which should be exchanged and stored using secure procedures
beyond the scope of the NTP protocol itself.
Besides the keys used for ordinary NTP associations,
additional ones can be used as passwords for the
.Xr ntpq 8
and
.Xr ntpdc 8
utility programs.

p
When
.Xr ntpd 8
is first started,
it reads the key file and installs the keys in the key cache.
However, the keys must be activated
before they can be used with the trusted command.
This allows, for instance,
the installation of possibly several batches of keys
and then activating or inactivating each batch remotely using
.Xr ntpdc 8 .
This also provides a revocation capability
that can be used if a key becomes compromised.
The
c requestkey command selects the key used as the password for the
.Xr ntpdc 8
utility,
while the
c controlkey command selects the key used as the password for the the
.Xr ntpq 8
utility.
t Autokey Scheme The original NTPv3 authentication scheme
described in RFC 1305 continues to be supported.
In NTPv4,
an additional authentication scheme called autokey is available.
It operates much like the S-KEY scheme,
in that a session key list is constructed
and the entries used in reverse order.
A description of the scheme,
along with a comprehensive security analysis,
is contained in a technical report
available from the IETF web page
.Li http://www.ietf.org/ .

p
The autokey scheme is specifically designed for multicast modes,
where clients normally do not send messages to the server.
In these modes,
the server uses the scheme to generate a key list
by repeated hashing of a secret value.
The list is used in reverse order
to generate a unique session key for each message sent.
The client regenerates the session key
and verifies the hash matches the previous session key.
Each message contains the public values
binding the session key to the secret value,
but these values need to be verified
only when the server generates a new key list
or more than four server messages have been lost.

p
The scheme is appropriate for client/server
and symmetric-peer modes as well.
In these modes,
the client generates a session key as in multicast modes.
The server regenerates the session key
and uses it to formulates a reply using its own public values.
The client verifies
the key identifier of the reply matches the request,
verifies the public values and validates the message.
In peer mode, each peer independently generates a key list
and operates as in the multicast mode.

p
The autokey scheme requires no change to the NTP packet header format
or message authentication code (MAC), which is appended to the header;
however, if autokey is in use, an extensions field is inserted
between the header and MAC.
The extensions field contains a random public value
which is updated at intervals specified by the revoke command,
together with related cryptographic values
used in the signing algorithm.
The format of the extensions field is defined in
Internet Draft
.Li draft-NTP-auth-coexist-00.txt .
The MAC itself is constructed in the same way as NTPv3,
but using the original NTP header
and the extensions field padded to a 64-bit boundary.
Each new public value is encrypted by the host private value.
It is the intent of the design, not yet finalized,
that the public value, encrypted public value,
public key and certificate be embedded in the extensions field
where the client can decrypt as needed.
However, the relatively expensive encryption
and decryption operations are necessary
only when the public value is changed.

p
Note that both the original NTPv3 authentication scheme
and the new NTPv4 autokey scheme
operate separately for each configured association,
so there may be several session key lists
operating independently at the same time.
Since all keys, including session keys,
occupy the same key cache,
provisions have been made to avoid collisions,
where some random roll happens to collide
with another already generated.
Since something like four billion different session key identifiers
are available,
the chances are small that this might happen.
If it happens during generation,
the generator terminates the current session key list.
By the time the next list is generated,
the collided key will probably have been expired or revoked.

p
While permanent keys have lifetimes that expire
only when manually revoked,
random session keys have a lifetime
specified at the time of generation.
When generating a key list for an association,
the lifetime of each key is set to expire
one poll interval later than it is scheduled to be used.
The maximum lifetime of any key in the list
is specified by the
c autokey command.
Lifetime enforcement is a backup
to the normal procedure that revokes the last-used key
at the time the next key on the key list is used.
.El
.Ss Authentication Options
The following authentication commands are available:
l -tag -width indent t Ic keys Ar keyfile Specifies the file name containing the encryption keys and
key identifiers used by
.Xr ntpd 8 ,
.Xr ntpq 8
and
.Xr ntpdc 8
when operating in authenticated mode.
The format of this file is described in the
.Xr ntp.keys 5
page.
t Xo Ic trustedkey .Ar key
.Op ...
.Xc
Specifies the encryption key identifiers which are trusted
for the purposes of authenticating peers
suitable for synchronization, as well as keys used by the
.Xr ntpq 8
and
.Xr ntpdc 8
programs.
The authentication procedures require that
both the local and remote servers share the same key
and key identifier for this purpose,
although different keys can be used with different servers.
The
.Ar trustedkey
arguments are 32-bit unsigned integers
with values less than 65,536.
Note that NTP key 0 is used to indicate an invalid key
and/or key identifier,
so should not be used for any other purpose.
t Ic requestkey Ar key Specifies the key identifier to use with the
.Xr ntpdc 8
program,
which uses a proprietary protocol
specific to this implementation of
.Xr ntpd 8 .
This program is useful to diagnose and repair problems
that affect
.Xr ntpd 8
operation.
The
.Ar key
argument to this command is a 32-bit key identifier
for a previously defined trusted key.
If no
c requestkey command is included in
the configuration file,
or if the keys don't match,
any request to change a server variable with be denied.
t Ic controlkey Ar key Specifies the key identifier to use with the
.Xr ntpq 8
program,
which uses the standard protocol defined in RFC 1305.
This program is useful to diagnose and repair problems
that affect
.Xr ntpd 8
operation.
The
.Ar key
argument to this command is a 32-bit key identifier
for a trusted key in the key cache.
If no
c controlkey command is included in the configuration file,
or if the keys don't match,
any request to change a server variable with be denied.
.El
.Ss Monitoring Support
.Xr ntpd 8
includes a comprehensive monitoring facility
suitable for continuous, long term recording
of server and client timekeeping performance.
See the
c statistics command below for a listing
and example of each type of statistics currently supported.
Statistic files are managed using file generation sets
and scripts in the

a ./scripts
directory of the source distribution.
Using these facilities and Unix
.Xr cron 8
jobs,
the data can be automatically summarized and archived
for retrospective analysis.
.Ss Monitoring Options
The following monitoring commands are available:
l -tag -width indent t Xo Ic statistics .Ar name
.Op ...
.Xc
Enables writing of statistics records.
Currently, four kinds of
.Ar name
statistics are supported.
l -tag -width indent t loopstats Enables recording of loop filter statistics information.
Each update of the local clock outputs
a line of the following form
to the file generation set named loopstats:

p
.Dl 50935 75440.031 0.000006019 13.778190 0.000351733 0.013380 6

p
The first two fields show the date (Modified Julian Day)
and time (seconds and fraction past UTC midnight).
The next five fields show time offset (seconds),
frequency offset (parts per million - PPM), RMS jitter (seconds),
Allan deviation (PPM) and clock discipline time constant.
t peerstats Enables recording of peer statistics information.
This includes statistics records of all peers of a NTP server
and of special signals, where present and configured.
Each valid update appends a line of the following form to
the current element of a file generation set named peerstats:

p
.Dl 48773 10847.650 127.127.4.1 9714 -0.001605 0.00000 0.00142

p
The first two fields show the date (Modified Julian Day)
and time (seconds and fraction past UTC midnight).
The next two fields show the peer address in dotted-quad notation
and status, respectively.
The status field is encoded in hex in the format
described in Appendix A of the NTP specification RFC 1305.
The final three fields show the offset, delay and RMS jitter,
all in seconds.
t clockstats Enables recording of clock driver statistics information.
Each update received from a clock driver appends a line
of the following form to the file generation set named clockstats:

p
.Dl 49213 525.624 127.127.4.1 93 226 00:08:29.606 D

p
The first two fields show the date (Modified Julian Day)
and time (seconds and fraction past UTC midnight).
The next field shows the clock address in dotted-quad notation.
The final field shows the last timecode received from the clock
in decoded ASCII format, where meaningful.
In some clock drivers
a good deal of additional information can be gathered and displayed
as well.
See information specific to each clock for further details.
t rawstats Enables recording of raw-timestamp statistics information.
This includes statistics records of all peers of a NTP server
and of special signals, where present and configured.
Each NTP message received from a peer or clock driver
appends a line of the following form
to the file generation set named rawstats:

p
d -ragged -offset indent .Li 50928
.Li 2132.543
.Li 128.4.1.1

 XXX The next field is unaccounted for in the descriptive text
 that follows.

.Li 128.4.1.20
.Li 3102453281.584327000
.Li 3102453281.58622800031
.Li 02453332.540806000
.Li 3102453332.541458000
.Ed

p
The first two fields show the date (Modified Julian Day)
and time (seconds and fraction past UTC midnight).
The next field shows the peer or clock address
in dotted-quad notation.
The final four fields show the originate,
receive, transmit and final NTP timestamps in order.
The timestamp values are as received and before processing
by the various data smoothing and mitigation algorithms.
.El
t Ic statsdir Ar directory_path Indicates the full path of a directory
where statistics files should be created (see below).
This keyword allows the
(otherwise constant) filegen filename prefix to be modified
for file generation sets,
which is useful for handling statistics logs.
t Xo Ic filegen .Ar name
.Op file Ar filename
.Op type Ar typename
.Op link | nolink
.Op enable | disable
.Xc
Configures setting of generation file set name.
Generation file sets provide a means for handling files
that are continuously growing during the lifetime of a server.
Server statistics are a typical example for such files.
Generation file sets provide
access to a set of files used to store the actual data.
At any time at most one element of the set is being written to.
The type given specifies when and how data will be directed
to a new element of the set.
This way, information stored in elements of a file set
that are currently unused are available for administrative operations
without the risk of disturbing the operation of
.Xr ntpd 8 .
Most importantly,
they can be removed to free space for new data produced.

p
Note that this command can be sent from the
.Xr ntpdc 8
program running at a remote location.
l -tag -width indent t name This is the type of the statistics records,
as shown in the
c statistics command.
t file Ar filename This is the file name for the statistics records.
Filenames of set members are built from three elements:
l -tag -width indent t prefix This is a constant filename path.
It is not subject to modifications via the
c filegen option.
It is defined by the server,
usually specified as a compile-time constant.
It may, however, be configurable for individual file generation sets
via other commands.
For example, the prefix used with loopstats and peerstats generation
can be configured using the
c statsdir option explained above.
.Ar filename
This string is directly concatenated to the prefix mentioned above
(no intervening
.Qq /
(slash)) .
This can be modified using the
.Ar filename
argument to the
c filegen statement.
No
.Qq ..
elements are allowed in this component
to prevent filenames referring to parts
outside the filesystem hierarchy denoted by prefix.
c suffix This part is reflects individual elements of a file set.
It is generated according to the type of a file set.
.El
t type Ar typename A file generation set is characterized by its type.
The following types are supported:
l -tag -width indent t none The file set is actually a single plain file.
t pid One element of file set is used per incarnation of a
.Xr ntpd 8
server.
This type does not perform any changes
to file set members during runtime,
however it provides an easy way of separating files
belonging to different
.Xr ntpd 8
server incarnations.
The set member filename is built by appending a
.Qq .
(dot) to concatenated prefix and
.Ar filename
strings,
and appending the decimal representation
of the process ID of the
.Xr ntpd 8
server process.
t day One file generation set element is created per day.
A day is defined as the period between 00:00 and 24:00 UTC.
The file set member suffix consists of a
.Qq .
(dot) and a day specification in the form YYYYMMDD.
YYYY is a 4-digit year number (e.g. 1992).
MM is a two digit month number.
DD is a two digit day number.
Thus, all information written at 10 December 1992
would end up in a file named

a <prefix><filename>.19921210 .
t week Any file set member contains data
related to a certain week of a year.
The term week is defined by computing the day of the year modulo 7.
Elements of such a file generation set are distinguished
by appending the following suffix to the file set
.Ar filename
base:
A dot, a 4-digit year number, the letter W,
and a 2-digit week number.
For example, information from January, 10th 1992
would end up in a file with suffix .1992W1.
t month One generation file set element is generated per month.
The file name suffix consists of a dot, a 4-digit year number,
and a 2-digit month.
t year One generation file element is generated per year.
The filename suffix consists of a dot and a 4 digit year number.
t age This type of file generation sets changes to a new element
of the file set every 24 hours of server operation.
The filename suffix consists of a dot, the letter a,
and an 8-digit number.
This number is taken to be the number of seconds
the server has been running
at the start of the corresponding 24-hour period.
Information is only written to a file generation
by specifying enable; output is prevented by specifying disable.
t link | nolink It is convenient to be able to access the current element
of a file generation set by a fixed name.
This feature is enabled by specifying link
and disabled using nolink.
If link is specified,
a hard link from the current file set element
to a file without suffix is created.
When there is already a file with this name
and the number of links of this file is one,
it is renamed appending a dot, the letter C,
and the pid of the
.Xr ntpd
server process.
When the number of links is greater than one,
the file is unlinked.
This allows the current file to be accessed by a constant name.
t enable | disable Enables or disables the recording function.
.El
.El
.El
.Ss Access Control Support
.Xr ntpd 8
implements a general purpose
address-and-mask based restriction list.
The list is sorted by address and by mask,
and the list is searched in this order for matches,
with the last match found
defining the restriction flags associated with the incoming packets.
The source address of incoming packets is used for the match,
with the 32-bit address being AND'ed with the mask
associated with the restriction entry
and then compared with the entry's address
(which has also been AND'ed with the mask)
to look for a match.
Additional information and examples can be found in the
.Qo
Notes on Configuring NTP and Setting up a NTP Subnet
.Qc
page.

p
The restriction facility was implemented
in conformance with the access policies
for the original NSFnet backbone time servers.
While this facility may be otherwise useful
for keeping unwanted or broken remote time servers
from affecting your own,
it should not be considered an alternative
to the standard NTP authentication facility.
Source address based restrictions are easily circumvented
by a determined cracker.
.Ss Access Control Options
The following access control commands are available:
l -tag -width indent t Xo Ic restrict .Ar numeric_address
.Op mask Ar numeric_mask
.Op Ar flag
.Op ...
.Xc
The
.Ar numeric_address
argument, expressed in dotted-quad form,
is the address of an host or network.
The
.Ar numeric_mask
argument, also expressed in dotted-quad form,
defaults to 255.255.255.255,
meaning that the
.Ar numeric_address
is treated as the address of an individual host.
A default entry
(address 0.0.0.0, mask 0.0.0.0)
is always included and, given the sort algorithm,
is always the first entry in the list.
Note that, while
.Ar numeric_address
is normally given in dotted-quad format,
the text string default, with no mask option,
may be used to indicate the default entry.

p
In the current implementation, flag always restricts access,
i.e. an entry with no flags indicates
that free access to the server is to be given.
The flags are not orthogonal, in that more restrictive flags
will often make less restrictive ones redundant.
The flags can generally be classed into two catagories,
those which restrict time service
and those which restrict informational queries
and attempts to do run-time reconfiguration of the server.
One or more of the following flags may be specified:
l -tag -width indent t ignore Ignore all packets from hosts which match this entry.
If this flag is specified neither queries
nor time server polls will be responded to.
t noquery Ignore all NTP mode 6 and 7 packets
(i.e. information queries and configuration requests)
from the source.
Time service is not affected.
t nomodify Ignore all NTP mode 6 and 7 packets
which attempt to modify the state of the server
(i.e. run time reconfiguration).
Queries which return information are permitted.
t notrap Decline to provide mode 6 control message trap service
to matching hosts.
The trap service is a subsystem
of the mode 6 control message protocol
which is intended for use by remote event logging programs.
t lowpriotrap Declare traps set by matching hosts to be low priority.
The number of traps a server can maintain is limited
(the current limit is 3).
Traps are usually assigned on a first come,
first served basis,
with later trap requestors being denied service.
This flag modifies the assignment algorithm
by allowing low priority traps to be overridden
by later requests for normal priority traps.
t noserve Ignore NTP packets whose mode is other than 6 or 7.
In effect,
time service is denied,
though queries may still be permitted.
t nopeer Provide stateless time service to polling hosts,
but do not allocate peer memory resources to these hosts
even if they otherwise might be considered useful
as future synchronization partners.
t notrust Treat these hosts normally in other respects,
but never use them as synchronization sources.
t limited These hosts are subject to limitation
of number of clients from the same net.
Net in this context refers to the IP notion of net
(class A, class B, class C, etc.).
Only the first
.Va client_limit
hosts (see below) that have shown up at the server
and that have been active during the last
.Va client_limit_period
seconds (see below) are accepted.
Requests from other clients from the same net are rejected.
Only time request packets are taken into account.
Query packets sent by the
.Xr ntpq 8
and
.Xr ntpdc 8
programs are not subject to these limits.
A history of clients is kept using the monitoring capability of
.Xr ntpd 8 .
Thus, monitoring is always active
as long as there is a restriction entry with the limited flag.
t ntpport This is actually a match algorithm modifier,
rather than a restriction flag.
Its presence causes the restriction entry to be matched
only if the source port in the packet
is the standard NTP UDP port (123).
Both ntpport and non-ntpport may be specified.
The ntpport is considered more specific
and is sorted later in the list.
.El

p
Default restriction list entries,
with the flags ignore and ntpport,
for each of the local host's interface addresses
are inserted into the table at startup
to prevent the server from attempting to synchronize
to its own time.
A default entry is also always present,
unless if it is otherwise unconfigured;
no flags are associated with the default entry
(i.e. everything besides your own NTP server is unrestricted).
t clientlimit Ar limit Set the
.Va client_limit
variable,
which limits the number of simultaneous access-controlled clients.
The default value for this variable is 3.
t clientperiod Ar period Set the
.Va client_limit_period
variable,
which specifies the number of seconds
after which a client is considered inactive
and thus no longer is counted for client limit restriction.
The default value for this variable is 3600 seconds.
.El
.Ss Reference Clock Support
The NTP Version 4 daemon supports many different radio,
satellite and modem reference clocks
plus a special pseudo-clock used for backup
or when no other clock source is available.
Detailed descriptions of individual device drivers
and options can be found in the
.Qo
Reference Clock Drivers
.Qc
page.
Additional information can be found in the pages referenced there,
including the
.Qo
Debugging Hints for Reference Clock Drivers
.Qc
and
.Qo
How To Write a Reference Clock Driver
.Qc
pages.
In many drivers,
support for a PPS signal is available as described in the
.Qo
Pulse-per-second (PPS) Signal Interfacing
.Qc
page.
Many drivers support special line discipline/streams modules
which can significantly improve the accuracy using the driver.
These are described in the
.Qo
Line Disciplines and Streams Drivers
.Qc
page.

p
A reference clock will generally (though not always)
be a radio timecode receiver
which is synchronized to a source of standard time
such as the services offered by the NRC in Canada
and NIST and USNO in the United States.
The interface between the computer and the timecode receiver
is device dependent, but is usually a serial port.
A device driver specific to each reference clock
must be selected and compiled in the distribution;
however, most common radio, satellite and modem clocks
are included by default.
Note that an attempt to configure a reference clock
when the driver has not been included
or the hardware port has not been appropriately configured
results in a scalding remark to the system log file,
but is not otherwise hazardous.

p
For the purposes of configuration,
.Xr ntpd 8
treats reference clocks in a manner
analogous to normal NTP peers as much as possible.
Reference clocks are identified by a syntactically correct
but invalid IP address,
in order to distinguish them from normal NTP peers.
Reference clock addresses are of the form 127.127.t.u,
where
.Ar t
is an integer denoting the clock type and
.Ar u
indicates the unit number.
While it may seem overkill,
it is in fact sometimes useful
to configure multiple reference clocks of the same type,
in which case the unit numbers must be unique.

p
The
c server command is used to configure a reference clock,
where the address argument in that command is the clock address.
The key,
version and ttl options are not used for reference clock support.
The mode option is added for reference clock support,
as described below.
The prefer option can be useful
to persuade the server to cherish a reference clock
with somewhat more enthusiasm than other reference clocks or peers.
Further information on this option can be found in the
.Qo
Mitigation Rules and the prefer Keyword
.Qc
page.
The minpoll and maxpoll options have meaning
only for selected clock drivers.
See the individual clock driver document pages
for additional information.

p
The stratum number of a reference clock is by default zero.
Since the
.Xr ntpd 8
daemon adds one to the stratum of each peer,
a primary server ordinarily displays stratum one.
In order to provide engineered backups,
it is often useful to specify the reference clock stratum
as greater than zero.
The stratum option is used for this purpose.
Also, in cases involving both a reference clock
and a pulse-per-second (PPS) discipline signal,
it is useful to specify the reference clock identifier
as other than the default, depending on the driver.
The refid option is used for this purpose.
Except where noted,
these options apply to all clock drivers.
.Ss Reference Clock Options
l -tag -width indent t Xo Ic server No 127.127. Ns Xo .Ar t Ns No . Ns Xo
.Ar u
.Op prefer
.Op mode Ar int
.Op minpoll Ar int
.Op maxpoll Ar int
.Xc
.Xc
.Xc
This command can be used to configure reference clocks
in special ways.
The options are interpreted as follows:
l -tag -width indent t prefer Marks the reference clock as preferred.
All other things being equal,
this host will be chosen for synchronization
among a set of correctly operating hosts.
See the
.Qo
Mitigation Rules and the prefer Keyword
.Qc
page
for further information.
t mode Ar int Specifies a mode number
which is interpreted in a device-specific fashion.
For instance, it selects a dialing protocol in the ACTS driver
and a device subtype in the parse drivers.
t minpoll Ar int t maxpoll Ar int These options specify the minimum and maximum polling interval
for reference clock messages, in seconds to the power of two.
For most directly connected reference clocks,
both minpoll and maxpoll default to 6 (64 s).
For modem reference clocks,
minpoll defaults to 10 (17.1 m)
and maxpoll defaults to 14 (4.5 h).
The allowable range is 4 (16 s) to 17 (36.4 h) inclusive.
.El
t Xo Ic fudge No 127.127. Ns Xo .Ar t Ns No . Ns Xo
.Ar u
.Op time1 Ar sec
.Op time2 Ar sec
.Op stratum Ar int
.Op refid Ar string
.Op mode Ar int
.Op flag1 Ar 0 Ns | Ns Ar 1
.Op flag2 Ar 0 Ns | Ns Ar 1
.Op flag3 Ar 0 Ns | Ns Ar 1
.Op flag4 Ar 0 Ns | Ns Ar 1
.Xc
.Xc
.Xc
This command can be used to configure reference clocks
in special ways.
It must immediately follow the
c server command which configures the driver.
Note that the same capability is possible at run time
using the
.Xr ntpdc 8
program.
The options are interpreted as follows:
l -tag -width indent t time1 Ar sec Specifies a constant to be added to the time offset produced
by the driver, a fixed-point decimal number in seconds.
This is used as a calibration constant
to adjust the nominal time offset of a particular clock
to agree with an external standard,
such as a precision PPS signal.
It also provides a way to correct a systematic error
or bias due to serial port latencies,
different cable lengths or receiver internal delay.
The specified offset is in addition to the propagation delay
provided by other means, such as internal DIPswitches.
Where a calibration for an individual system
and driver is available,
an approximate correction is noted
in the driver documentation pages.
t time2 Ar secs Specifies a fixed-point decimal number in seconds,
which is interpreted in a driver-dependent way.
See the descriptions of specific drivers in the
.Qo
Reference Clock Drivers
.Qc
page.
t stratum Ar int Specifies the stratum number assigned to the driver,
an integer between 0 and 15.
This number overrides the default stratum number
ordinarily assigned by the driver itself, usually zero.
t refid Ar string Specifies an ASCII string from one to four characters
which defines the reference identifier used by the driver.
This string overrides the default identifier
ordinarily assigned by the driver itself.
t mode Ar int Specifies a mode number which is interpreted
in a device-specific fashion.
For instance,
it selects a dialing protocol in the ACTS driver
and a device subtype in the parse drivers.
t flag1 flag2 flag3 flag4 These four flags are used for customizing the clock driver.
The interpretation of these values,
and whether they are used at all,
is a function of the particular clock driver.
However, by convention
flag4 is used to enable recording monitoring data
to the clockstats file configured with the
c filegen command.
When a PPS signal is available,
a special automatic calibration facility is provided.
If the flag1 switch is set
and the PPS signal is actively disciplining the system time,
the calibration value is automatically adjusted
to maintain a residual offset of zero.
Further information on the
c filegen command can be found in the
.Sx Monitoring Options
section.
.El
t Ic pps device [assert|clear] [hardpps] Specifies the name and options for the serial port device
to which the PPS signal is connected.
Note, this command replaces use of fudge flag3,
which was used for the same purpose in NTPv3.
Note that this command should preceed the
c server and
c fudge commands for the same device.
Note also that the assert,
clear and hardpps options are only available
if the ppsapi standard PPS interface is available.
l -tag -width indent t device Specify the device name associated with the PPS signal.
The name must match exactly the link name specified
in the driver documentation page.
c assert c clear Using assert or clear specifies
if the high going or low going edge
of the signal must be used.
The default is assert.
c hardpps This flag is used to tell the kernel that the signal
from this device must be used to drive hardpps().
The assert, clear and hardpps options are only available
if the PPSAPI is used.
.El
.El
.Ss Miscellaneous Options
The following miscellaneous configuration options are available:
l -tag -width indent t Ic broadcastdelay Ar seconds The broadcast and multicast modes require a special calibration
to determine the network delay between the local and remote
servers.
Ordinarily, this is done automatically
by the initial protocol exchanges
between the local and remote servers.
In some cases, the calibration procedure may fail
due to network or server access controls, for example.
This command specifies
the default delay to be used under these circumstances.
Typically (for Ethernet),
a number between 0.003 and 0.007 seconds is appropriate.
The default when this command is not used is 0.004 seconds.
t Xo Ic trap .Ar host_address
.Op port Ar port_number
.Op interface Ar interface_address
.Xc
This command configures a trap receiver
at the given host address and port number
for sending messages with the specified local interface address.
If the port number is unspecified, a value of 18447 is used.
If the interface address is not specified,
the message is sent with a source address of the local interface
the message is sent through.
Note that on a multihomed host
the interface used may vary from time to time
with routing changes.
The trap receiver will generally log event messages
and other information from the server in a log file.
While such monitor programs
may also request their own trap dynamically,
configuring a trap receiver
will ensure that no messages are lost when the server is started.
t Ic setvar Ar variable Op default This command adds an additional system variable.
These variables can be used
to distribute additional information such as the access policy.
If the variable of the form
.Va name
=
.Ar value
is followed by the default keyword,
the variable will be listed
as part of the default system variables
(see the
.Xr ntpq 8
c rv command).
These additional variables serve informational purposes only.
They are not related to the protocol
other that they can be listed.
The known protocol variables will always override any variables
defined via the
c setvar mechanism.
There are three special variables
that contain the names of all variables of the same group.
The
.Va sys_var_list
holds the names of all system variables.
The
.Va peer_var_list
holds the names of all peer variables and the
.Va clock_var_list
holds the names of the reference clock variables.
t Ic logfile Ar logfile This command specifies the location of an alternate log file
to be used instead of the default system
.Xr syslog 3
facility.
t Ic logconfig Ar configkeyword This command controls the amount and type of output
written to the system
.Xr syslog 3
facility or the alternate
c logfile log file.
By default, all output is turned on.
All
.Ar configkeyword
keywords can be prefixed with =, + and -,
where = sets the syslogmask,
+ adds and - removes messages.
.Xr syslog 3
messages can be controlled
in four classes (clock, peer, sys and sync).
Within these classes
four types of messages can be controlled.
Informational messages (info) control configuration information.
Event messages (events) control logging of events
(reachability, synchronization, alarm conditions).
Statistical output is controlled with the
c statistics keyword.
The final message group is the status messages.
This describes mainly the synchronizations status.

p
Configuration keywords are formed
by concatenating the message class with the event class.
The all prefix can be used instead of a message class.
A message class may also be followed by the all keyword
to enable/disable all messages of the respective message class.
Thus, a minimal log configuration could look like this:

p
.Dl logconfig = syncstatus +sysevents

p
This would just list the synchronizations state of
.Xr ntpd 8
and the major system events.
For a simple reference server,
the following minimum message configuration could be useful:

p
.Dl logconfig = syncall +clockall

p
This configuration will list all clock information
and synchronization information.
All other events and messages about peers,
system events and so on is suppressed.
.El
.Sh FILES
l -tag -width /etc/ntp.drift -compact t Pa /etc/ntp.conf the default name of the configuration file
.El
.Sh SEE ALSO
.Xr ntpd 8 ,
.Xr ntpdc 8 ,
.Xr ntpq 8

p
In addition to the manual pages provided,
comprehensive documentation is available on the world wide web
at
.Li http://www.ntp.org/ .
A snapshot of this documentation is available in HTML format in

a /usr/share/doc/ntp .
.Rs
.%A David L. Mills
.%T Network Time Protocol (Version 3)
.%O RFC1305
.Re
.Sh HISTORY
Written by
.An David Mills
at the University of Delaware.
.Sh BUGS
.Xr ntpd 8
has gotten rather fat.
While not huge, it has gotten larger than might
be desireable for an elevated-priority daemon running on a workstation,
particularly since many of the fancy features which consume the space
were designed more with a busy primary server, rather than a high
stratum workstation, in mind.