[ntp:questions] Re: Ntpdate - socket in use

Paul Croome Paul.Croome at softwareag.com
Wed Jan 14 07:55:21 UTC 2004


nope at invalid.mynetwork (George Kaplan) wrote in message news:<40037e62.2604725 at news.centennialpr.net>...
(...)
> Now for another problem.  My WinXP client won't update its time from my ntpd
> server, claiming:  "An error occurred while Windows was synchronizing with
> gibson.  The time sample was rejected because: The peer's stratum is less than
> the host's stratum."
> 
> My ntpq output is:
> 
> [root at gibson /etc]# ntpq -p -n
>      remote           refid      st t when poll reach   delay   offset  jitter
> ==============================================================================
>  209.244.17.41   .INIT.          16 u    -   64    0    0.000    0.000 4000.00
>  209.244.0.5     10.64.126.39     2 u  114  128    7  149.877   12.397   5.856
>  209.244.0.6     10.64.126.39     2 u  110  128    7  149.828   10.382   3.148
>  209.247.10.162  .INIT.          16 u    -   64    0    0.000    0.000 4000.00
>  128.101.101.101 160.94.54.248    2 u  109  128    7  160.268    5.620  10.301
>  134.84.84.84    160.94.54.248    2 u  109  128    7  160.274    7.290   6.507
>  192.168.100.1   .INIT.          16 u    -   64    0    0.000    0.000 4000.00
> [root at gibson /etc]# 

Isaac,

The 7's in the 'reach' column imply that ntp has only been running for a
couple of minutes, so it's not synced yet. Typically, you should let an
ntp server run for at least 2 hours before you try to sync a client to it;
it can take that long for the phase-lock/frequency-lock loops to settle down.
Standard practice is to run an ntp server 24*7.

Regards,
Paul



More information about the questions mailing list