Difference between revisions of "Network Time Protocol daemon"

From ArchWiki
Jump to: navigation, search
(Synchronize once per boot: the "-q" option leads to an error with systemd. oneshot is the option which controls execution in the service file.)
m (Running in a chroot: pidfile is not necessary in the current ntpd.service, move jail directory to the end and emphasize (it's *the* part related to this section))
 
(214 intermediate revisions by 35 users not shown)
Line 3: Line 3:
 
[[fr:ntp]]
 
[[fr:ntp]]
 
[[it:Network Time Protocol daemon]]
 
[[it:Network Time Protocol daemon]]
 +
[[ja:Network Time Protocol daemon]]
 
[[ru:Network Time Protocol daemon]]
 
[[ru:Network Time Protocol daemon]]
 
[[zh-CN:Network Time Protocol daemon]]
 
[[zh-CN:Network Time Protocol daemon]]
This article describes how to set up and run NTPd (Network Time Protocol daemon), the most common method to synchronize the [[Time|software clock]] of a GNU/Linux system with internet time servers using the [[Wikipedia:Network Time Protocol|Network Time Protocol]]; if set up correctly, NTPd can make your computer act as a time server itself.
+
{{Related articles start}}
 +
{{Related|Time}}
 +
{{Related|systemd-timesyncd}}
 +
{{Related|OpenNTPD}}
 +
{{Related|Chrony}}
 +
{{Related articles end}}
 +
[[Wikipedia:Network Time Protocol|Network Time Protocol]] is the most common method to synchronize the [[Time|software clock]] of a GNU/Linux system with internet time servers. It is designed to mitigate the effects of variable network latency and can usually maintain time to within tens of milliseconds over the public Internet. The accuracy on local area networks is even better, up to one millisecond.
  
==Installation==
+
[http://support.ntp.org/bin/view/Main/WebHome#The_NTP_Project The NTP Project] provides a reference implementation of the protocol called simply NTP. This article further describes how to set up and run the NTP daemon, both as a client and as a server.
[[pacman|Install]] {{pkg|ntp}}, available in the [[Official Repositories]].
+
  
==Configuration==
+
== Installation ==
{{Tip|The {{pkg|ntp}} package is installed with a default {{ic|/etc/ntp.conf}} that should make NTPd work without requiring custom configuration.}}
+
  
===Configuring connection to NTP servers===
+
[[Install]] the {{Pkg|ntp}} package. By default, ''ntpd'' works in client mode without further configuration. You can skip to [[#Usage]], if you want to use the Arch Linux default configuration file for it. For server configuration, see [[#NTP server mode]].
The first thing you define in your {{ic|/etc/ntp.conf}} is the servers your machine will synchronize to.
+
 
 +
== Configuration ==
 +
 
 +
The main daemon is ''ntpd'', which is configured in {{ic|/etc/ntp.conf}}. Refer to the manual pages for detail: {{ic|man ntp.conf}} and the related {{ic|man <nowiki>{ntpd|ntp_auth|ntp_mon|ntp_acc|ntp_clock|ntp_misc}</nowiki>}}.
 +
 
 +
=== Connection to NTP servers ===
  
 
NTP servers are classified in a hierarchical system with many levels called ''strata'': the devices which are considered independent time sources are classified as ''stratum 0'' sources; the servers directly connected to ''stratum 0'' devices are classified as ''stratum 1'' sources; servers connected to ''stratum 1'' sources are then classified as ''stratum 2'' sources and so on.
 
NTP servers are classified in a hierarchical system with many levels called ''strata'': the devices which are considered independent time sources are classified as ''stratum 0'' sources; the servers directly connected to ''stratum 0'' devices are classified as ''stratum 1'' sources; servers connected to ''stratum 1'' sources are then classified as ''stratum 2'' sources and so on.
  
It has to be understood that a server's stratum cannot be taken as an indication of its accuracy or reliability. Typically, stratum 2 servers are used for general synchronization purposes: if you do not already know the servers you are going to connect to, you should use the [http://www.pool.ntp.org/ pool.ntp.org] servers ([http://support.ntp.org/bin/view/Servers/NTPPoolServers alternate link]) and choose the server pool that is closest to your location.
+
It has to be understood that a server's stratum cannot be taken as an indication of its accuracy or reliability. Typically, stratum 2 servers are used for general synchronization purposes: if you do not already know the servers you are going to connect to, you should choose a server pool close to your location from the [http://www.pool.ntp.org/ pool.ntp.org] servers ([http://support.ntp.org/bin/view/Servers/NTPPoolServers alternative link]).
  
The following lines are just an example:
+
Since ''ntp'' version 4.2.7.p465-2, Arch Linux uses its own default vendor pool of NTP servers provided by [http://www.pool.ntp.org the NTP Pool Project] (see {{Bug|41700}}). Modify those to suit your needs, e.g. if you want to use your country's servers with an option:
 
{{hc|/etc/ntp.conf|
 
{{hc|/etc/ntp.conf|
server 0.pool.ntp.org iburst
+
server 0.fr.pool.ntp.org iburst
server 1.pool.ntp.org iburst
+
server 1.fr.pool.ntp.org iburst
server 2.pool.ntp.org iburst
+
server 2.fr.pool.ntp.org iburst
server 3.pool.ntp.org iburst
+
server 3.fr.pool.ntp.org iburst
 
}}
 
}}
 
The {{ic|iburst}} option is recommended, and sends a burst of packets only if it cannot obtain a connection with the first attempt. The {{ic|burst}} option always does this, even on the first attempt, and should never be used without explicit permission and may result in blacklisting.
 
The {{ic|iburst}} option is recommended, and sends a burst of packets only if it cannot obtain a connection with the first attempt. The {{ic|burst}} option always does this, even on the first attempt, and should never be used without explicit permission and may result in blacklisting.
  
===Configuring your own NTP server===
+
=== NTP server mode ===
If setting up an NTP server, you need to add [http://www.ntp.org/ntpfaq/NTP-s-refclk.htm#Q-LOCAL-CLOCK ''local clock''] as a server, so that, in case it loses internet access, it will continue serving time to the network; add ''local clock'' as a stratum 10 server (using the ''fudge'' command) so that it will never be used unless internet access is lost:
+
  
  server 127.127.1.0
+
If setting up an NTP server, you need to add [http://www.ntp.org/ntpfaq/NTP-s-refclk.htm#Q-LOCAL-CLOCK ''local clock''] as a server, so that, in case it loses internet access, it will continue serving time to the network; add ''local clock'' as a stratum 10 server (using the ''fudge'' command) (you can set up to [http://www.ntp.org/ntpfaq/NTP-s-algo.htm#Q-ALGO-BASIC-STRATUM stratum 15]) so that it will never be used unless internet access is lost:
  fudge  127.127.1.0 stratum 10
+
 
 +
  server 127.127.1.1
 +
  fudge  127.127.1.1 stratum 12
  
 
Next, define the rules that will allow clients to connect to your service (''localhost'' is considered a client too) using the ''restrict'' command; you should already have a line like this in your file:
 
Next, define the rules that will allow clients to connect to your service (''localhost'' is considered a client too) using the ''restrict'' command; you should already have a line like this in your file:
Line 39: Line 50:
 
  restrict default nomodify nopeer noquery
 
  restrict default nomodify nopeer noquery
  
This restricts everyone from modifying anything and prevents everyone from querying the status of your time server: {{ic|nomodify}} prevents reconfiguring your ntpd (with ''ntpq'' or ''ntpdc''), and {{ic|noquery}} prevents dumping status data from your ntpd (also with ''ntpq'' or ''ntpdc'').
+
This restricts everyone from modifying anything and prevents everyone from querying the status of your time server: {{ic|nomodify}} prevents reconfiguring ''ntpd'' (with ''ntpq'' or ''ntpdc''), and {{ic|noquery}} is [https://mailman.archlinux.org/pipermail/arch-dev-public/2014-February/025872.html important to prevent] dumping status data from ''ntpd'' (also with ''ntpq'' or ''ntpdc'').
  
 
You can also add other options:
 
You can also add other options:
Line 45: Line 56:
 
  restrict default kod nomodify notrap nopeer noquery
 
  restrict default kod nomodify notrap nopeer noquery
  
{{Note|This still allows other people to query your time server. You need to add {{ic|noserve}} to stop serving time. It will also block time synchronization since it blocks all packets except ntpq and ntpdc queries.}}
+
{{Note|This still allows other people to query your time server. You need to add {{ic|noserve}} to stop serving time. It will also block time synchronization since it blocks all packets except ''ntpq'' and ''ntpdc'' queries.}}
  
Full docs for the "restrict" option are in {{ic|man ntp_acc}}. See https://support.ntp.org/bin/view/Support/AccessRestrictions for detailed instructions.
+
If you want to change any of these, see the full docs for the "restrict" option in {{ic|man ntp_acc}}, the detailed [https://support.ntp.org/bin/view/Support/AccessRestrictions ntp instructions] and [[#As a daemon]]{{Broken section link}}.
  
 
Following this line, you need to tell ''ntpd'' what to allow through into your server; the following line is enough if you are not configuring an NTP server:
 
Following this line, you need to tell ''ntpd'' what to allow through into your server; the following line is enough if you are not configuring an NTP server:
Line 63: Line 74:
 
  logfile /var/log/ntp.log
 
  logfile /var/log/ntp.log
  
A very basic configuration file will look like this ('''all comments have been stripped out for clarity'''):
+
A very basic configuration file will look like this:
  
 
{{hc|/etc/ntp.conf|
 
{{hc|/etc/ntp.conf|
Line 83: Line 94:
 
{{Note|Defining the log file is not mandatory, but it is always a good idea to have feedback for ''ntpd'' operations.}}
 
{{Note|Defining the log file is not mandatory, but it is always a good idea to have feedback for ''ntpd'' operations.}}
  
===Other resources about NTP configuration===
+
== Usage ==
In conclusion, never forget man pages: {{ic|man ntp.conf}} is likely to answer any doubts you could still have (see also the related man pages: {{ic|man <nowiki>{ntpd|ntp_auth|ntp_mon|ntp_acc|ntp_clock|ntp_misc}</nowiki>}}).
+
  
==Using without daemon==
+
The package has a default client-mode configuration and its own user and group to drop root privileges after starting. If you start it from the console, you should always do so with the {{ic|-u}} option:
To synchronize your system clock just once, without starting the NTP daemon, run:
+
# ntpd -qg
+
  
{{Note|This has the same effect as the [http://support.ntp.org/bin/view/Dev/DeprecatingNtpdate now deprecated] {{ic|ntpdate}}.}}
+
# ntpd -u ntp:ntp
  
The {{ic|-g}} option allows shifting the clock further than the panic threshold (15 min by default) without a warning. Note that such offset is abnormal and might indicate either wrong timezone setting, clock chip failure, or simply a very long period of neglect. If in these cases you would rather not set the clock and print an error to syslog, remove {{ic|-g}}.
+
The {{ic|-u}} option is employed by the two included systemd services. These services also use the {{ic|-g}} option, which disables a threshold (so-called ''panic-gate''). Hence, they will synchonize time even in case the ntp-server's time exceeds the threshold deviation from the system clock.
  
After updating the system clock, store the time to the hardware clock so that it is preserved when rebooting:
+
{{Warning|One reason the panic-gate was introduced is that background jobs/services may be susceptible to time-jumps. If the system's clock was never synchronized before, consider stopping them before running ''ntpd'' for the first time.}}
# hwclock -w
+
  
===Synchronize once per boot===
+
Both services are tied to the system's resolver, and will start synchronizing when an active network connection is detected.
{{Warning|Using this method is discouraged on servers and in general on machines that need to run continuously for more than 2 or 3 days, as the system clock will be updated only once at boot time.}}
+
  
Write a ''oneshot'' [[systemd]] unit:
+
=== Start ntpd at boot ===
{{hc|/etc/systemd/system/ntp-once.service|2=
+
[Unit]
+
Description=Network Time Service (once)
+
After=network.target nss-lookup.target
+
  
 +
[[Enable]] the daemon with {{ic|ntpd.service}}. See also [[#Running in a chroot]].
 +
 +
{{Note|The systemd command ''timedatectl'' can only be used to control [[systemd-timesyncd]], executing {{ic|timedatectl set-ntp 1}} as root will inadvertedly stop a running {{ic|ntpd.service}}.[http://lists.freedesktop.org/archives/systemd-devel/2015-April/030277.html]}}
 +
 +
Use ''ntpq'' to see the list of configured peers and status of synchronization:
 +
 +
$ ntpq -p
 +
 +
The delay, offset and jitter columns should be non-zero. The servers ''ntpd'' is synchronizing with are prefixed by an asterisk. It can take several minutes before ''ntpd'' selects a server to synchronize with; try checking after 17 minutes (1024 seconds).
 +
 +
=== Synchronize time once per boot ===
 +
 +
Alternatively, [[enable]] {{ic|ntpdate.service}} to synchronize time once (option {{ic|-q}}) and non-forking (option {{ic|-n}}) per boot, instead of running the daemon in the background. This method is discouraged on servers, and in general on machines that run without rebooting for more than a few days.
 +
 +
If the synchronized time should be written to the hardware clock as well, configure the provided unit as described in [[systemd#Editing provided units]] before starting it:
 +
 +
{{hc|/etc/systemd/system/ntpdate.service.d/hwclock.conf|2=
 
[Service]
 
[Service]
Type=oneshot
+
ExecStart=/usr/bin/hwclock -w
ExecStart=/usr/bin/ntpd -g -u ntp:ntp ; /sbin/hwclock -w
+
}}
  
[Install]
+
== Tips and tricks ==
WantedBy=multi-user.target}}
+
  
and enable it: {{bc|# systemctl enable ntp-once}}
+
=== Start ntpd on network connection ===
Note that a [[systemd]] unit of the type ''oneshot'' executes once only. Hence the {{ic|ntpd -q}} option should not be used in this case.
+
  
==Running as a daemon==
+
''ntpd'' can be started by your network manager, so that the daemon only runs when the computer is online.
  
To start ntpd:
+
;Netctl
{{bc|# systemctl start ntpd}}
+
  
To enable ntpd at startup:
+
{{Style|add {{ic|-u}} and optionally refer to [[#Usage]], or use systemctl if possible}}
{{bc|# systemctl enable ntpd}}
+
  
Or alternatively with the command:  
+
Append the following lines to your [[netctl]] profile:
{{bc|# timedatectl set-ntp 1}}
+
  
===Check whether the daemon is synchronizing correctly===
+
ExecUpPost='/usr/bin/ntpd || true'
Before you can use the ntpq command you'll need to use pacman and install the {{pkg|libedit}} package.  Then use ntpq to see the list of configured peers:
+
ExecDownPre='killall ntpd || true'
{{bc|$ ntpq -np}}
+
The delay, offset and jitter columns should be non-zero. The servers ntpd is synchronizing with are prefixed by an asterisk. It can take several minutes before ntpd selects a server to synchronize with; try checking after 17 minutes (1024 seconds).
+
  
===NetworkManager===
+
;NetworkManager
{{Note|ntpd should still be running when the network is down if the hwclock daemon is disabled, so you should not use this.}}
+
''ntpd'' can be brought up/down along with a network connection through the use of [[NetworkManager#Network services with NetworkManager dispatcher|NetworkManager's dispatcher scripts]]. You will need to install {{Pkg|networkmanager-dispatcher-ntpd}} from the [[official repositories]].
+
  
===Running in a chroot===
+
The ''ntpd'' daemon can be brought up/down along with a network connection through the use of NetworkManager's [[NetworkManager#Network services with NetworkManager dispatcher|dispatcher]] scripts. The {{Pkg|networkmanager-dispatcher-ntpd}} package installs one, pre-configured to start and stop the [[#Start ntpd at boot|ntpd service]] with a connection.
{{Note|ntpd should be run as non-root before attempting to jail it in a chroot (default in the vanilla Arch Linux package), since chroots are relatively useless at securing processes running as root.}}
+
  
Edit {{ic|/etc/conf.d/ntpd.conf}} and change
+
;Wicd
NTPD_ARGS="-g -u ntp:ntp"
+
  
to
+
For [[Wicd]], create a start script in the {{ic|postconnect}} directory and a stop script in the {{ic|predisconnect}} directory. Remember to make them executable:
  NTPD_ARGS="-g -i /var/lib/ntp -u ntp:ntp"
+
 
 +
{{hc|/etc/wicd/scripts/postconnect/ntpd|
 +
#!/bin/bash
 +
systemctl start ntpd &
 +
}}
 +
 
 +
{{hc|/etc/wicd/scripts/predisconnect/ntpd|
 +
#!/bin/bash
 +
systemctl stop ntpd &
 +
}}
 +
 
 +
{{Note|You are advised to customize the options for the ''ntpd'' command as explained in [[#Usage]].}}
 +
 
 +
See also [[Wicd#Scripts]].
 +
 
 +
;KDE
 +
 
 +
KDE can use NTP (ntp must be installed) by right clicking the clock and selecting ''Adjust date/time''. However, this requires the ntp daemon to be [[disable]]d before configuring KDE to use NTP. [https://bugs.kde.org/show_bug.cgi?id=178968]
 +
 
 +
=== Using ntpd with GPS ===
 +
 
 +
Most of the articles online about configuring ''ntpd'' to receive time from a GPS suggest to use the SHM (shared memory) method. However, at least since ''ntpd'' version 4.2.8 a ''much better'' method is available. It connects directly to ''gpsd'', so {{Pkg|gpsd}} needs to be installed.
 +
 
 +
Add these lines to your {{ic|/etc/ntp.conf}}:
 +
 
 +
{{hc|head=/etc/ntp.conf|output=
 +
#=========================================================
 +
# GPSD native ntpd driver
 +
#=========================================================
 +
# This driver exists from at least ntp version 4.2.8
 +
# Details at
 +
#  https://www.eecis.udel.edu/~mills/ntp/html/drivers/driver46.html
 +
server 127.127.46.0
 +
fudge 127.127.46.0 time1 0.0 time2 0.0 refid GPS
 +
}}
 +
 
 +
This will work as long as you have ''gpsd'' working. It connects to ''gpsd'' via the local socket and queries the "gpsd_json" object that is returned.
 +
 
 +
To test the setup, first ensure that ''gpsd'' is working by running:
 +
 
 +
  $ cgps -s
 +
 
 +
Then wait a few minutes and run {{ic|ntpq -p}}. This will show if ''ntpd'' is talking to ''gpsd'':
 +
 
 +
{{hc|$ ntpq -p|output=
 +
    remote          refid            st t when poll reach  delay  offset  jitter
 +
==================================================================================
 +
*GPSD_JSON(0)    .GPS.            0 l  55  64  377    0.000    2.556  14.109
 +
}}
 +
 
 +
{{Tip|If the ''reach'' column is 0, it means ''ntpd'' has not been able to talk to ''gpsd''. Wait a few minutes and try again. Sometimes it takes ''ntpd'' a while.}}
 +
 
 +
=== Running in a chroot ===
 +
 
 +
{{Note|''ntpd'' should be started as non-root (default in the Arch Linux package) before attempting to jail it in a chroot, since chroots are relatively useless at securing processes running as root.}}
 +
 
 +
Create a new directory {{ic|/etc/systemd/system/ntpd.service.d/}} if it does not exist and a file named {{ic|customexec.conf}} inside with the following content:
 +
 
 +
[Service]
 +
ExecStart=
 +
ExecStart=/usr/bin/ntpd -g -u ntp:ntp '''-i /var/lib/ntp'''
  
 
Then, edit {{ic|/etc/ntp.conf}} to change the driftfile path such that it is relative to the chroot directory, rather than to the real system root. Change:
 
Then, edit {{ic|/etc/ntp.conf}} to change the driftfile path such that it is relative to the chroot directory, rather than to the real system root. Change:
Line 152: Line 220:
  
 
Create a suitable chroot environment so that getaddrinfo() will work by creating pertinent directories and files (as root):
 
Create a suitable chroot environment so that getaddrinfo() will work by creating pertinent directories and files (as root):
 +
 
  # mkdir /var/lib/ntp/etc /var/lib/ntp/lib /var/lib/ntp/proc
 
  # mkdir /var/lib/ntp/etc /var/lib/ntp/lib /var/lib/ntp/proc
 
  # touch /var/lib/ntp/etc/resolv.conf /var/lib/ntp/etc/services
 
  # touch /var/lib/ntp/etc/resolv.conf /var/lib/ntp/etc/services
Line 168: Line 237:
 
  # mount -a
 
  # mount -a
  
Finally, restart the daemon again:
+
Finally, restart {{ic|ntpd}} daemon again. Once it restarted you can verify that the daemon process is chrooted by checking where {{ic|/proc/{PID}/root}} symlinks to:  
  # systemctl restart ntpd
+
 
 +
  # ps -C ntpd | awk '{print $1}' | sed 1d | while read -r PID; do ls -l /proc/$PID/root; done
 +
 
 +
should now link to {{ic|/var/lib/ntp}} instead of {{ic|/}}.
 +
 
 +
It is relatively difficult to be sure that your driftfile configuration is actually working without waiting a while, as ''ntpd'' does not read or write it very often. If you get it wrong, it will log an error; if you get it right, it will update the timestamp. If you do not see any errors about it after a full day of running, and the timestamp is updated, you should be confident of success.
 +
 
 +
== Troubleshooting ==
 +
 
 +
=== Cannot assign requested address ===
 +
 
 +
{{Accuracy|This is not a solution for people who want/need to use ipv6 and have the same problem.}}
 +
 
 +
If you get the message ''Cannot assign requested address'' as shown below:
 +
 
 +
{{hc|$ journalctl -u ntpd|2=
 +
ntpd[2130]: bind(21) AF_INET6 fe80::6ef0:49ff:fe51:4946%2#123 flags 0x11 failed: Cannot assign requested address
 +
ntpd[2130]: unable to create socket on eth0 (5) for fe80::6ef0:49ff:fe51:4946%2#123
 +
ntpd[2130]: failed to init interface for address fe80::6ef0:49ff:fe51:4946%2
 +
}}
  
It is relatively difficult to be sure that your driftfile configuration is actually working without waiting a while, as ntpd does not read or write it very often. If you get it wrong, it will log an error; if you get it right, it will update the timestamp. If you do not see any errors about it after a full day of running, and the timestamp is updated, you should be confident of success.
+
you can get rid of it by disabling IP6. To do so, [[Systemd#Editing_provided_units|edit]] {{ic|ntpd.service}} and add {{ic|-4}}:
  
==Alternatives==
+
[Service]
An alternative to NTPd is [[Chrony]], a dial-up friendly and specifically designed for systems that are not online all the time.
+
ExecStart=
 +
ExecStart=/usr/bin/ntpd -g -u ntp:ntp '''-4'''
  
==See also==
+
== See also ==
* [[Time]] (for more information on computer timekeeping)
+
  
==External links==
 
 
* http://www.ntp.org/
 
* http://www.ntp.org/
* http://support.ntp.org/
+
* https://support.ntp.org/
 
* http://www.pool.ntp.org/
 
* http://www.pool.ntp.org/
* http://www.eecis.udel.edu/~mills/ntp/html/index.html
+
* https://www.eecis.udel.edu/~mills/ntp/html/index.html
 
* http://www.akadia.com/services/ntp_synchronize.html
 
* http://www.akadia.com/services/ntp_synchronize.html

Latest revision as of 18:56, 13 August 2016

Network Time Protocol is the most common method to synchronize the software clock of a GNU/Linux system with internet time servers. It is designed to mitigate the effects of variable network latency and can usually maintain time to within tens of milliseconds over the public Internet. The accuracy on local area networks is even better, up to one millisecond.

The NTP Project provides a reference implementation of the protocol called simply NTP. This article further describes how to set up and run the NTP daemon, both as a client and as a server.

Installation

Install the ntp package. By default, ntpd works in client mode without further configuration. You can skip to #Usage, if you want to use the Arch Linux default configuration file for it. For server configuration, see #NTP server mode.

Configuration

The main daemon is ntpd, which is configured in /etc/ntp.conf. Refer to the manual pages for detail: man ntp.conf and the related man {ntpd|ntp_auth|ntp_mon|ntp_acc|ntp_clock|ntp_misc}.

Connection to NTP servers

NTP servers are classified in a hierarchical system with many levels called strata: the devices which are considered independent time sources are classified as stratum 0 sources; the servers directly connected to stratum 0 devices are classified as stratum 1 sources; servers connected to stratum 1 sources are then classified as stratum 2 sources and so on.

It has to be understood that a server's stratum cannot be taken as an indication of its accuracy or reliability. Typically, stratum 2 servers are used for general synchronization purposes: if you do not already know the servers you are going to connect to, you should choose a server pool close to your location from the pool.ntp.org servers (alternative link).

Since ntp version 4.2.7.p465-2, Arch Linux uses its own default vendor pool of NTP servers provided by the NTP Pool Project (see FS#41700). Modify those to suit your needs, e.g. if you want to use your country's servers with an option:

/etc/ntp.conf
server 0.fr.pool.ntp.org iburst
server 1.fr.pool.ntp.org iburst
server 2.fr.pool.ntp.org iburst
server 3.fr.pool.ntp.org iburst

The iburst option is recommended, and sends a burst of packets only if it cannot obtain a connection with the first attempt. The burst option always does this, even on the first attempt, and should never be used without explicit permission and may result in blacklisting.

NTP server mode

If setting up an NTP server, you need to add local clock as a server, so that, in case it loses internet access, it will continue serving time to the network; add local clock as a stratum 10 server (using the fudge command) (you can set up to stratum 15) so that it will never be used unless internet access is lost:

server 127.127.1.1
fudge  127.127.1.1 stratum 12

Next, define the rules that will allow clients to connect to your service (localhost is considered a client too) using the restrict command; you should already have a line like this in your file:

restrict default nomodify nopeer noquery

This restricts everyone from modifying anything and prevents everyone from querying the status of your time server: nomodify prevents reconfiguring ntpd (with ntpq or ntpdc), and noquery is important to prevent dumping status data from ntpd (also with ntpq or ntpdc).

You can also add other options:

restrict default kod nomodify notrap nopeer noquery
Note: This still allows other people to query your time server. You need to add noserve to stop serving time. It will also block time synchronization since it blocks all packets except ntpq and ntpdc queries.

If you want to change any of these, see the full docs for the "restrict" option in man ntp_acc, the detailed ntp instructions and #As a daemon[broken link: invalid section].

Following this line, you need to tell ntpd what to allow through into your server; the following line is enough if you are not configuring an NTP server:

restrict 127.0.0.1

If you want to force DNS resolution to the IPv6 namespace, write -6 before the IP address or host name (-4 forces IPv4 instead), for example:

restrict -6 default kod nomodify notrap nopeer noquery
restrict -6 ::1    # ::1 is the IPv6 equivalent for 127.0.0.1

Lastly, specify the drift file (which keeps track of your clock's time deviation) and optionally the log file location:

driftfile /var/lib/ntp/ntp.drift
logfile /var/log/ntp.log

A very basic configuration file will look like this:

/etc/ntp.conf
server 0.pool.ntp.org iburst
server 1.pool.ntp.org iburst
server 2.pool.ntp.org iburst
server 3.pool.ntp.org iburst

restrict default kod nomodify notrap nopeer noquery
restrict -6 default kod nomodify notrap nopeer noquery

restrict 127.0.0.1
restrict -6 ::1  

driftfile /var/lib/ntp/ntp.drift
logfile /var/log/ntp.log
Note: Defining the log file is not mandatory, but it is always a good idea to have feedback for ntpd operations.

Usage

The package has a default client-mode configuration and its own user and group to drop root privileges after starting. If you start it from the console, you should always do so with the -u option:

# ntpd -u ntp:ntp

The -u option is employed by the two included systemd services. These services also use the -g option, which disables a threshold (so-called panic-gate). Hence, they will synchonize time even in case the ntp-server's time exceeds the threshold deviation from the system clock.

Warning: One reason the panic-gate was introduced is that background jobs/services may be susceptible to time-jumps. If the system's clock was never synchronized before, consider stopping them before running ntpd for the first time.

Both services are tied to the system's resolver, and will start synchronizing when an active network connection is detected.

Start ntpd at boot

Enable the daemon with ntpd.service. See also #Running in a chroot.

Note: The systemd command timedatectl can only be used to control systemd-timesyncd, executing timedatectl set-ntp 1 as root will inadvertedly stop a running ntpd.service.[1]

Use ntpq to see the list of configured peers and status of synchronization:

$ ntpq -p

The delay, offset and jitter columns should be non-zero. The servers ntpd is synchronizing with are prefixed by an asterisk. It can take several minutes before ntpd selects a server to synchronize with; try checking after 17 minutes (1024 seconds).

Synchronize time once per boot

Alternatively, enable ntpdate.service to synchronize time once (option -q) and non-forking (option -n) per boot, instead of running the daemon in the background. This method is discouraged on servers, and in general on machines that run without rebooting for more than a few days.

If the synchronized time should be written to the hardware clock as well, configure the provided unit as described in systemd#Editing provided units before starting it:

/etc/systemd/system/ntpdate.service.d/hwclock.conf
[Service]
ExecStart=/usr/bin/hwclock -w

Tips and tricks

Start ntpd on network connection

ntpd can be started by your network manager, so that the daemon only runs when the computer is online.

Netctl

Tango-edit-clear.pngThis article or section needs language, wiki syntax or style improvements.Tango-edit-clear.png

Reason: add -u and optionally refer to #Usage, or use systemctl if possible (Discuss in Talk:Network Time Protocol daemon#)

Append the following lines to your netctl profile:

ExecUpPost='/usr/bin/ntpd || true'
ExecDownPre='killall ntpd || true'
NetworkManager

The ntpd daemon can be brought up/down along with a network connection through the use of NetworkManager's dispatcher scripts. The networkmanager-dispatcher-ntpd package installs one, pre-configured to start and stop the ntpd service with a connection.

Wicd

For Wicd, create a start script in the postconnect directory and a stop script in the predisconnect directory. Remember to make them executable:

/etc/wicd/scripts/postconnect/ntpd
#!/bin/bash
systemctl start ntpd &
/etc/wicd/scripts/predisconnect/ntpd
#!/bin/bash
systemctl stop ntpd &
Note: You are advised to customize the options for the ntpd command as explained in #Usage.

See also Wicd#Scripts.

KDE

KDE can use NTP (ntp must be installed) by right clicking the clock and selecting Adjust date/time. However, this requires the ntp daemon to be disabled before configuring KDE to use NTP. [2]

Using ntpd with GPS

Most of the articles online about configuring ntpd to receive time from a GPS suggest to use the SHM (shared memory) method. However, at least since ntpd version 4.2.8 a much better method is available. It connects directly to gpsd, so gpsd needs to be installed.

Add these lines to your /etc/ntp.conf:

/etc/ntp.conf
#=========================================================
#  GPSD native ntpd driver
#=========================================================
# This driver exists from at least ntp version 4.2.8
# Details at
#   https://www.eecis.udel.edu/~mills/ntp/html/drivers/driver46.html
server 127.127.46.0 
fudge 127.127.46.0 time1 0.0 time2 0.0 refid GPS

This will work as long as you have gpsd working. It connects to gpsd via the local socket and queries the "gpsd_json" object that is returned.

To test the setup, first ensure that gpsd is working by running:

 $ cgps -s 

Then wait a few minutes and run ntpq -p. This will show if ntpd is talking to gpsd:

$ ntpq -p
remote           refid            st t when poll reach   delay   offset  jitter
 ==================================================================================
*GPSD_JSON(0)    .GPS.            0 l   55   64  377    0.000    2.556  14.109
Tip: If the reach column is 0, it means ntpd has not been able to talk to gpsd. Wait a few minutes and try again. Sometimes it takes ntpd a while.

Running in a chroot

Note: ntpd should be started as non-root (default in the Arch Linux package) before attempting to jail it in a chroot, since chroots are relatively useless at securing processes running as root.

Create a new directory /etc/systemd/system/ntpd.service.d/ if it does not exist and a file named customexec.conf inside with the following content:

[Service]
ExecStart=
ExecStart=/usr/bin/ntpd -g -u ntp:ntp -i /var/lib/ntp

Then, edit /etc/ntp.conf to change the driftfile path such that it is relative to the chroot directory, rather than to the real system root. Change:

driftfile       /var/lib/ntp/ntp.drift

to

driftfile       /ntp.drift

Create a suitable chroot environment so that getaddrinfo() will work by creating pertinent directories and files (as root):

# mkdir /var/lib/ntp/etc /var/lib/ntp/lib /var/lib/ntp/proc
# touch /var/lib/ntp/etc/resolv.conf /var/lib/ntp/etc/services

and by bind-mounting the aformentioned files:

/etc/fstab
...
#ntpd chroot mounts
/etc/resolv.conf  /var/lib/ntp/etc/resolv.conf none bind 0 0
/etc/services	  /var/lib/ntp/etc/services none bind 0 0
/lib		  /var/lib/ntp/lib none bind 0 0
/proc		  /var/lib/ntp/proc none bind 0 0
# mount -a

Finally, restart ntpd daemon again. Once it restarted you can verify that the daemon process is chrooted by checking where /proc/{PID}/root symlinks to:

# ps -C ntpd | awk '{print $1}' | sed 1d | while read -r PID; do ls -l /proc/$PID/root; done

should now link to /var/lib/ntp instead of /.

It is relatively difficult to be sure that your driftfile configuration is actually working without waiting a while, as ntpd does not read or write it very often. If you get it wrong, it will log an error; if you get it right, it will update the timestamp. If you do not see any errors about it after a full day of running, and the timestamp is updated, you should be confident of success.

Troubleshooting

Cannot assign requested address

Tango-inaccurate.pngThe factual accuracy of this article or section is disputed.Tango-inaccurate.png

Reason: This is not a solution for people who want/need to use ipv6 and have the same problem. (Discuss in Talk:Network Time Protocol daemon#)

If you get the message Cannot assign requested address as shown below:

$ journalctl -u ntpd
ntpd[2130]: bind(21) AF_INET6 fe80::6ef0:49ff:fe51:4946%2#123 flags 0x11 failed: Cannot assign requested address
ntpd[2130]: unable to create socket on eth0 (5) for fe80::6ef0:49ff:fe51:4946%2#123
ntpd[2130]: failed to init interface for address fe80::6ef0:49ff:fe51:4946%2

you can get rid of it by disabling IP6. To do so, edit ntpd.service and add -4:

[Service]
ExecStart=
ExecStart=/usr/bin/ntpd -g -u ntp:ntp -4

See also