[ntp:questions] Re: ntp & adsl

Richard B. Gilbert rgilbert88 at comcast.net
Tue Jan 31 01:42:00 UTC 2006


Folkert van Heusden wrote:

>Hi,
>
>I've heard that NTP acts funny when it tries to sync over an ADSL line.
>That seems to be the case indeed:
>0 root at muur:/home/folkert/www$ ntpq -c pe
>     remote           refid      st t when poll reach   delay   offset  jitter
>==============================================================================
> GENERIC(0)      .DCFa.           0 l   23   64    3    0.000  -17.386  26.356  <- dcf77 receiver
>-keetweej.intran 130.149.17.8     2 u  810 1024  376    0.735    2.724   0.084  <- other server on my lan
><snip>
>  
>

>And there it is: 20ms offset. From what I've heard this is caused by
>ADSL being asymetric.
>Now I was wondering, can't this be fixed by telling the NTP daemon how
>much asynchronous the line is? I mean: it can be easily measured what
>the down- and upload bandwidth is.
>
>  
>
I have a similar problem.  I have a GPS reference clock.   The network 
servers I use as backup and  sanity checks each have a unique systematic 
offset from my GPS.  These offsets range from +2 to +5 milliseconds.  
There is clearly more going on here than mere asymmetry in my broadband 
cable.  There is no way to compensate for such a systematic offset on a 
server by server basis.   I would really like to see a new feature in 
ntpd that would allow me to correct for these offsets.   Without it, my 
local clock is going to get jerked around every time a new 
synchronization source is selected.

Even after allowing for the systematic offsets, it's also clear that 
some of the servers out there in netland have serious problems of their 
own!  One of the servers I use is a true jewel; it agrees with GPS quite 
closely showing only a +2 millisecond offset.  It's as solid as a rock, 
day in and day out.   Of the others, perhaps the less said, the better; 
I'm glad that I seldom need to rely on them.




More information about the questions mailing list