Having been designed on Linux, NTP (Network Time Protocol) is relatively simple to configure on a Linux machine. By using NTP (available free to download via NTP.org) any Linux machine can be easily set up to run as an NTP server.

Once downloaded the NTP distribution should contain the NTP daemon and also a number of utilities and configuration scripts. These aid the installation process and provide debugging facilities. The NTP daemon is configured using the file ‘ntp.conf’. A list of commands can be specified in the ‘ntp.conf’ file to indicate which servers to synchronise to and to specify various authentication and access options.

The NTP daemon synchronises to an external reference clock. The internet can be used as a time source but these can’t be authenticated and being the wrong side of the firewall could leave the system compromised. It’s much better to use an external source such as a GPS clock or radio clock that receive time from long wave transmissions (broadcast by such institutions as NIST or NPL).

Multiple external time servers can be specified in the configuration file, which allows NTP server to select the most appropriate time server and to use an average of the most reliable sources ensuing a higher level of accuracy.

The NTP daemon is controlled by a series of scripts such as ‘ntpd start’, ‘ntp stop’ or ‘ntpd restart’. Debugging and querying can be done by using the ‘ntpq’ utility. This utility provides information relating to the synchronisation status of the NTP daemon.

Many network administrators are aware of the NTP server (Network Time Protocol) and how it synchronises a computer network to UTC time (Coordinated Universal Time). These devices have revolutionised the way we work and trade in the global marketplace ensuring that computer networks from across the world are synchronised to the same time.

But computer networks are not the only part of an organisation that requires to know the time, people too are constantly relying on the time and if a computer network is running reliable and accurate time then it makes sense that the workforce is too.

However, in many organisations it is common for the office wall clock to be several minutes behind or ahead of the NTP server which is why many office managers now insist that digital wall clocks are used that connect to a NTP server.

Digital wall clocks can be used to display the correct UTC time in any location. These devices ensure that everybody in your organisation can be aware of the exact UTC time and that entire organisations can be synchronised to the exact same time.

Displaying hours, minute sand seconds these devices are easy to install and can be powered by Ethernet (POE).

The NTP Pool of time servers has announced that they now have a collection of over 1,000 time servers in Europe alone. The NTP pool was set up to prevent NTP server abuse and to insure there is not too much strain on the stratum 1 time servers.

The pool is now so popular over one million users choose top get their timing sources from there as most of the stratum 2 NTP servers on the NTP pool website are free to use and access over the Internet.

The NTP Pool is an ideal location for those running single machines or smaller networks were security and accuracy is not really an issue, however, for those charged with running a secure network or where accuracy is really important then internet time servers should not be seriously considered.

Unfortunately Internet time services cannot be authenticated (NTP’s security measure) which can leave a machine vulnerable, furthermore, Internet time sources are outside of a network’s firewall so a port needs to be left open to allow the time server can network to communicate.

For those serious about synchronisation there is no alternative other than to invest in a dedicated NTP server that receives its time securely from either radio signals or the GPS network.

When we take a glance at our watches or the office clock we often take for granted that the time we are given is correct. We may notice if our watches are ten minutes fast or slow but take little heed if they are a second or two out.

Yet for thousands of years mankind has strode to get ever increasingly accurate clocks the benefits of which are plentiful today in our age of satellite navigation, NTP servers, the Internet and global communications.

To understand how accurate time can be measured it is first important to understand the concept of time itself. Time as it has been measured on Earth for millennia is a different concept to time itself which as Einstein informed us was part of the fabric of the universe itself in what he described as a four dimensional space-time.

Yet we have historically measured time based not on the passing of time itself but the rotation of our planet in relation to the Sun and the Moon. A day is divided into 24 equal parts (hours) each of which is divided into 60 minutes and the minute is divided into 60 seconds.

However, it has now been realised that measuring time this way can not be considered accurate as the Earth’s rotation varies from day to day. All sorts of variable such as tidal forces, hurricanes, solar winds and even the amount of snow at the poles effects the speed of the Earth’s rotation. In fact when the dinosaurs first started roaming the Earth, the length of a day as we measure it now would have only been 22 hours.

We now base our timekeeping on the transition of atoms with a second based on 9,192,631,770 periods of the radiation emitted by the hyperfine transition of a unionized caesium atom in the ground state. Whilst this may sound complicated it really is just an atomic ‘tick’ that never alters and therefore can provide a highly accurate reference to base our time on.

Atomic clocks use this atomic resonance and can keep time that is so accurate a second isn’t lost in even a billion years. Modern technologies all take advantage of this precision enabling many of the communications and global trade we benefit from today with the utilisation of satellite navigation, NTP servers and air traffic control changing the way we live our lives.

Be careful when counting down to the New Year this week as an extra second is to be added to the last minute of 2008. Leap Seconds are added to UTC (Coordinated Universal Time) the time used by most NTP Server systems and has been going on since its inception in 1972 with a total of 33 seconds having already been added to UTC since then.

However there are calls to make this leap second the last ever. The International Telecommunications Union (ITU) is proposing to abolish them in favour of adding a “leap hour” every 600 years.

They claim that the Leap Second creates confusion and can cause software crashes. They cite 1996 as an example, when computers at Associated Press Radio crashed causing them to broadcast the wrong programmes, and a 2003 bug caused some GPS units to show the time as 62:30.

However, there is stiff opposition to abolishing the Leap Second, among them astronomers, Robert Massey of the Royal Astronomical Society told the London Times: “It would be a change with profound cultural implications.  We’d be decoupling our clocks from what the Sun is telling us.”

The implications for astronomers would require expensive changes to the software of astronomical telescopes and it would become almost impossible for sailors to navigate by sextant. It would also mean the Britain would lose its role as the World’s timekeepers as the Greenwich Meridian (the position of the sun at midday) would gradually move south to France until the proposed leap hour would return it to its original position after 600 years.

Network Time Protocol is a hierarchical protocol and is divided into stratum which define the distance from the reference clock. A reference clock source that relays UTC (Coordinated Universal Time) time and has little or no delay is known as a stratum-0 device. Stratum-0 servers cannot be used on the network, instead, they are directly connected to computers which then operate as primary time servers.

A primary server that receives a time signal from a stratum 0 device either through the GPS network or national time and frequency transmission is known as a stratum-1 device. On a network a stratum 1 time server supplies the time to other devices on the network which are known as stratum-2 devices. These also can be used as a time source and equipment that connects to a stratum-2 device to receive it become stratum-3 and so on.

NTP can handle up to 16 different stratum levels, although the lower down the hierarchy you go the less accurate the devices become. However, to make the system more reliable, each client can receive a time source from multiple servers. Stratum 2 devices and below can also synchronise with each other. The NTP software monitors continuously the figures of stability and accuracy of all the servers and always chooses a server with the best figures.

Multiple stratum are used an in larger networks because to bombard a single stratum-1 time server with NTP requests from thousands of machines could cause it to overload or block the network itself with workstations/routers etc repeatedly waiting for their time synchronisation requests to go through.

An example of a typical NTP network   (Image courtesy of Wikipedia)

Stratum 1 time servers connected to a stratum 0 source, stratum 2 servers connected to stratum 1 and stratum 3 connected to stratum 2.

We have recently started to develop a new NTP Server resource. Please take a moment to visit this resource and see for yourself.