Strange NTP Server problem

[merlin]

My pfSense box is configured to take clock from my ISP NTP server. And this part works fine…
I saw message like
ntpdate[454]: step time server x.x.x.x offset 0.893096 sec
in log file

I enabled NTP server on pfSense 1.2 on LAN interface and tried to sync my linux pc with pfSense box but w/o sucess… Here are the debug log (command is executed on pfSense box with lan ip 192.168.100.1)

$ ntpdate -vqd 192.168.100.1
4 Jun 19:32:27 ntpdate[3603]: ntpdate 4.2.0-a Sun Feb 24 16:21:21 EST 2008 (1)
transmit(192.168.100.1)
receive(192.168.100.1)
transmit(192.168.100.1)
receive(192.168.100.1)
transmit(192.168.100.1)
receive(192.168.100.1)
transmit(192.168.100.1)
receive(192.168.100.1)
transmit(192.168.100.1)
192.168.100.1: Server dropped: Leap not in sync
server 192.168.100.1, port 123
stratum 3, precision -21, leap 11, trust 000
refid [192.168.100.1], delay 0.02609, dispersion 0.00006
transmitted 4, in filter 4
reference time:    cbf13d43.b98567ff  Wed, Jun  4 2008 19:11:47.724
originate timestamp: cbf1421b.89dd6fff  Wed, Jun  4 2008 19:32:27.538
transmit timestamp:  cbf1421b.89c630eb  Wed, Jun  4 2008 19:32:27.538
filter delay:  0.02609  0.02658  0.02612  0.02609
        0.00000  0.00000  0.00000  0.00000
filter offset: 0.000036 0.000181 0.000112 0.000110
        0.000000 0.000000 0.000000 0.000000
delay 0.02609, dispersion 0.00006
offset 0.000036
–---------------------------------------------

So... the server looks in sync (startum 3, my ISP server is startum 2) but declare that is not synchronized (leap bits set to 11)?

Any ideas how to fix this ?

[Cry Havok]

You'll find other threads on this, the OpenBSD NTP server takes a long time (tens of minutes) before it syncs.  IMO it's far, far, poorer than the stock ISC version.  On the other hand, I've not been bothered enough by it to actually look to creating a package ;)