The LANTIME/SHS/BGT takes full advantage of the integrated PZF module. The high-precision long wave receiver includes an independent antenna for DCF77, so the time string and the PPS signal are derived from a totally independent time source, DCF77. Of course, this device requires that the signal from the German long wave transmitter DCF77 can be received at the location where the device is installed.
The clock modules are assembled in a 19″ modular case (3U) which also includes a single board computer, and a power supply unit. Configurable settings can be modified via menus on the LC display and the four buttons in the front panel. A failure output can be used to generate an alarm signal if any malfunction is detected.
If it is necessary to provide redundancy against hardware failure then it´s also possible to install several LANTIME NTP servers in the same network.
LANTIME/SHS/BGT Modes Of Operation
Normal mode of operation: Both the clocks are synchronized, the difference between the clock´s timing output signals is below the configured limit. The NTP server receives the time information including the status “synchronized”, so it acts as stratum-1 server and makes the reference time available to the network.
One of the clocks falls out of sync e.g. due to an antenna failure or other reception problems: The clock changes its status to “not synchronized” and continues counting time based on its built-in high-accuracy chrystal oscillator. Depending on the configuation, it takes some days up to several weeks until the difference between the time signals exceeds the limit. Since one of the radio clocks is still synchronized, the timing information is passed to the NTP server with status “synchronized” until the limit is exceeded.
Both the clocks are not synchronized to their primary time sources although at least one of them has been synchronized before: As long as the time difference doesn´t exceed the limit, the time information is passed to the NTP server, but the status included is set to “not synchronized”. The NTP server keeps accepting the time information for a given trust time, after the trust time it discards the time information.
Both the clocks are not synchronized to their primary time sources and both have not been synchronized after the last power-up: The hybrid clock does not pass any time information to the NTP server until at least one of the independent clocks is synchronized and the time difference between the clocks is below the configured limit.
The clocks generate time information with a difference which exceeds the configured limit, or one of the radio clocks doesn´t output any time information at all:
One of the following circumstances can be the reason why:
An intended external fake
Failure or malfunction of one of the primary time transmitters
Failure or malfunction of one of the clocks
Persistent reception problems
In all the cases listed above the plausibility checks on the timing information fail, so the hybrid radio clock stops passing any timing information on to the NTP server. The NTP server´s stratum changes to a worse value to let the clients know that the server´s reference time source fails. The hybrid radio clock continues supplying time to the NTP server after all error conditions have been removed and the error has been acknowledged by an operator.
If one or more additional LANTIME NTP servers are available on the network then clients which have been configured to use all of them will automatically discard the LANTIME with the bad stratum and synchronize to another NTP server which is operating correctly at a better stratum. If no redundant LANTIME is available, however, the clients will continue to synchronize to the LANTIME with worse stratum. This way it is guaranteed that all the client devices on the network operate using the same system time.
All changes of the reception status of one of the radio clocks, and also failure of the hybrid clock´s plausibility check are logged by the local Linux system and optionally reported across the network. If the hybrid receiver passes the status “not synchronized” to the NTP server, or it has disabled time information output at all, then the alarm signal output of the LANTIME/SHS/BGT is activated.
The GNU/Linux operating system of the LANTIMEs SBC (Single Board Computer) has been optimized to ensure a high level of network security and reliability. A large display shows the state of the internal SHS receiver and the NTP subsystem.
The configuration of the system can be done by using a standard web browser to access the extensive but straightforward html interface. Alternatively a text based and menu driven setup utility can be started from the shell prompt after logging into the unit via Telnet or SSH.
The security-related features of LANTIME time servers satisfy highest demands. The time synchronization data can be reliably signed and secured by symmetric keys (MD5) and the NTP autokey procedures. This protects the clients against manipulated time and man-in-the-middle attacks and allows them to verify that the NTP packets they received were send by the LANTIME. Additionally the whole LANTIME configuration can be done by using encrypted channels (e.g. SSH, HTTPS or SNMPv3). Every unused/unneeded protocol can be disabled in order to reduce possible points of attack.
In order to support network management systems the LANTIME time servers offer an extensive SNMP interface, which can be accessed by SNMP V1, V2.c and V3. It allows the monitoring of all relevant system parameters (including operating system parameters, network interface statistics, detailed GPS and NTP status information as well as the complete system configuration) and can be used to alter the LANTIME configuration via SNMP set commands, too.
LANTIME time servers are designed to be deployed in IPv6 networks, the NTP time synchronization as well as the configuration interfaces (Web-based, SSH and SNMP) comes with IPv6 support. You can assign several IPv6 addresses and the system supports automatic configuration by IPv6 autoconf.
Because of its modular system architecture it is possible to equip a LANTIME time server with up to three additional ethernet ports and a number of different reference time sources. Optionally several additional frequency-, serial string- and pulse outputs are available and by combining two (even different) time sources and redundant power supplies, high-availability systems are no problem. Besides that a collection of oscillators, from the reliable temperature controlled base model (TCXO), the three excellent oven controlled variants (OCXO-LQ, -MQ and -HQ) to the high-end rubidium based top model, offers a wide selection of holdover characteristics (e.g. when the GPS signal is disturbed or jammed).