Server Time Synchronisation

Owned by Charlotte Dakin-Norris, created
Last updated: Nov 08, 2021
3 min read

Introduction

The Witness™ software is a real-time, distributed network application and therefore is sensitive to time differences between server components. Many disparate components within the system can generate data with timestamps; it is therefore essential that the servers on which the components run are synchronised to a single time source.

It is not essential that the hardware systems are configured to use the correct time, it is just critical that all the systems are using the same time, right or wrong. However if the times are not accurate be aware that this will reflect in the reporting of all data that contains dates and times, including alarms, logs and track sightings. We strongly recommend that all systems are synchronised with a reliable and accurate time source so the time is correct as well as being the same on all systems.

Contents

Time Synchronisation Alarms

The Witness™ software is continually monitoring all the software components to ensure they are reporting the same time. If there is a problem with time differences between the software services on different servers, a system alarm will be raised. This alarm will persist whilst the time difference remains.

If one or more systems are not synchronised then the behaviour of the software can be become unpredictable and therefore this must be addressed as a matter of urgency.

Be aware that time synchronisation errors can be caused by excessive load on one of the systems or a network delay. This can result in a time sync check being delayed which will cause an alarm. These alarms are normally temporary and clear once the load drops. The first action must always be to check that the system times are synchronised. Assuming they are, then you should investigate the load and network performance on the systems in question and ensure they are sufficiently resourced to handle their expected workload.

Time Synchronisation Options

The two most common ways of synchronising time are NTP and the Windows Time Sync Service.

NTP

NTP is an internet protocol that’s used to synchronise the clocks on computer networks to within a few milliseconds of universal coordinated time (UTC). It enables devices to request and receive UTC from a server that, in turn, receives precise time from an atomic clock. NTP also makes allowances for network latency and time drift thereby maintaining accurate synchronisation across a distributed system.

For organisations of any size, NTP is a cost-effective, reliable and user-friendly method of distributing precise time throughout a network, allowing users to boost productivity, enhance security and more. Most operating systems support NTP, either a client, server or more typically both. This is true of Windows and most Linux distributions. Windows NTP setup is not that straightforward, however there are some very good third party applications that can be installed to provide a time source and also act as an NTP client. If you decide this is the most appropriate route, we strongly recommend the Meinberg driver. This offers the ability to use both a network or internet time source as well as the local hardware clock. The latter option means the software can be used to synchronise all the systems on the network even if an internet connection is not available. The driver must be installed on all systems and at least one of the systems must be configured as the primary time source.

Windows Time Sync Service

This is a built in feature of the Windows operating system and can be configured to work in a very similar way to NTP. Within an Active Directory, a client PC will automatically synchronise its time with the Domain Controllers. Outside an Active Directory the client PCs can synchronise their time directly with an internet based time source. By default this is public Windows Time server but can be changed to another NTP source if required.

Related Information

Safety is everything.