Meinberg LANTIME, Руководство по эксплуатации

Страница: 17 / 153

The NTP daemon is started with "ntpd" or, using "rc.local", while booting the system.
Status messages during operation are saved in /var/adm/messages and /
var/adm/syslog (corresponding to the syslog configuration).
e.g.: tail /var/log/messages
shows the last lines from the file "messages"
The status messages can also be redirected in a log file by using the following option:
ntpd -llogfile
The command "ntpq" in the directory "ntpq" requests the actual status of the NTP
daemon (see also doc/ntpq.8).
e.g.: ntpq/ntpq
An interpreter appears; Type "?" for a list of all available commands. The command
"peer" is used to list all active reference clocks:
remote refid st t when poll reach delay offset jitter
================================================================================
LOCAL(0)
LOCAL(0) 3 l 36 64 3 0.00 0.000 7885
lantime .GPS. 0 l 36 64 1 0.00 60.1 15875
with the following meaning:
- remote:
list of all valid time servers (ntp.conf)
- refid: reference number
- st: actual stratum value (hierarchy level)
- when: last request (seconds)
- poll: period of requesting the time server (seconds)
- reach: octal notation of the successful requests, shifted left
- delay: delay of the network transmission (milliseconds)
- offset: difference between system time and reference time
(milliseconds)
- jitter: variance of the offsets (milliseconds)
Repeatedly "peer" commands lets the user observe the accuracy of the NTP daemon.
Every 64 seconds (value of -poll) a new time string is red in from the radio clock. The
NTP daemon needs approx. 3...5 minutes for initialisation. This is indicated by a
wildcard (*) on the left side of the remote name.
The NTP daemon terminates itself if the system time differs from the UTC time by
more than 1024 seconds. This often happens when the time zone is not correctly set
(see also system manual "zic" or "man zic").
17