[ntp:questions] ntptime says that clock is unsynchronized

BJörn Lindqvist bjourne at gmail.com
Mon Nov 10 16:43:26 UTC 2008


2008/11/8 David Woolley <david at ex.djwhome.demon.co.uk.invalid>:
> BJörn Lindqvist wrote:
>
>>     # ntptime
>>     ntp_gettime() returns code 5 (ERROR)
>>       time ccbd930d.7cfa6000  Thu, Nov  6 2008 16:00:13.488, (.488195),
>>       maximum error 16384000 us, estimated error 16 us
>
> What ntp.conf and command line options are you using.  Some options,
> such as used to disable stepping, are incompatible with the use of the
> kernel time discipline.

ntp.conf on the server:

server 127.127.1.0 burst iburst minpoll 4 maxpoll 4
fudge 127.127.1.0 stratum 10
trap 127.0.0.1
restrict default noquery nomodify
restrict 127.0.0.1
driftfile /var/lib/ntp/drift/ntp.drift
logfile /var/log/ntp
disable kernel

and on the client:

server 127.127.1.0 burst iburst minpoll 4 maxpoll 4
fudge 127.127.1.0 stratum 12
server 192.168.1.4
restrict default noquery nomodify
restrict 127.0.0.1
driftfile /var/lib/ntp/drift/ntp.drift
logfile /var/log/ntp
disable kernel

And standard options I think:

/usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp

The thing is that the ntptime problems are intermittent. Sometimes
ntptime returns ERROR and other times OK. I can't see any pattern to
it.

> PS I know of no valid reason for configuring a LOCAL clock driver on a
> pure client.  It can only cause confusion.

Actually the client may serve as a server in case its upstream
reference disappears. So that other pure clients can still get time
from someplace.


-- 
mvh Björn



More information about the questions mailing list