[ntp:questions] Poul-Henning Kamp and re-write of NTP

Paul tik-tok at bodosom.net
Mon Dec 22 15:31:54 UTC 2014


On Sun, Dec 7, 2014 at 3:47 AM, Harlan Stenn <stenn at ntp.org> wrote:

> Not quite.  It's a leaf-node client only, and it will not include any
> built-in monitoring capablilty (that will be a separate add-on).
>

Since I like being in the cool crowd I fired up ntimed-client on a couple
of machines.  I haven't done any plots etc. of the trace file but you look
at it and go m'kay.  I don't really feel like perturbing the clock but I
might later.

Then I'm thinking how do really know.  My current zero-effort test is
ntpdate which has the offset down in the tens of microseconds -- but not
after just an hour.

I count 3.346 lines in *.c and 997 in *.h.
The resident size is about half 4.2.8 although that's probably not a fair
comparison.


More information about the questions mailing list