xntpd : time sync status : unreach

[thomas gourdin]

hi,

ntpdate works fine but xntpd doesn't synchronise clock with the
time server :

client ip : 10.60.10.8
server ip : 10.60.20.145
(not behind a firewall)

ntpdate -dv 10.60.20.145 :

ntpdate 4.0.98a Wed Oct 16 17:25:25 EDT 2002
transmit(10.60.20.145)
receive(10.60.20.145)
transmit(10.60.20.145)
receive(10.60.20.145)
transmit(10.60.20.145)
receive(10.60.20.145)
transmit(10.60.20.145)
receive(10.60.20.145)
transmit(10.60.20.145)
server 10.60.20.145, port 123
stratum 15, precision -6, leap 00, trust 000
refid [0.0.0.0], delay 0.04124, dispersion 0.00031
transmitted 4, in filter 4
reference time: c34a3c15.87ef9d15 Wed, Oct 29 2003 14:16:05.531
originate timestamp: c34a3c2f.2fdf3b2d Wed, Oct 29 2003 14:16:31.187
transmit timestamp: c34a3c2f.1c17a893 Wed, Oct 29 2003 14:16:31.109
filter delay: 0.04124 0.04124 0.04221 0.04124
0.00000 0.00000 0.00000 0.00000
filter offset: 0.078240 0.078240 0.077752 0.077264
0.000000 0.000000 0.000000 0.000000
delay 0.04124, dispersion 0.00031
offset 0.078240

adjust time server 10.60.20.145 offset 0.078240 sec

/etc/services :

ntp 123/udp

ntp.conf :

server 10.60.20.145 burst
driftfile /cluster/members/member/etc/ntp.drift
enable stats
logfile /var/log/ntpd
logconfig =syncall +clockall +sysall

/etc/ntp.drift :

0.000

/var/log/ntpd :

system event 'event_restart' (0x01) status 'sync_alarm, sync_unspec, 1 event, event_unspec' (0xc010)

daemon.log :

ntpd 4.0.98a Wed Oct 16 17:36:24 EDT 2002
precision = 976 usec
kern_enable is 1
using kernel phase-lock loop 0000
frequency initialized 0.000 from /cluster/members/member/etc/ntp.drift
using kernel phase-lock loop 0000

ntpq -c as :

ind assID status conf reach auth condition last_event cnt
===========================================================
1 924 8000 yes no

ntpq -c rv :

status=c011 sync_alarm, sync_unspec, 1 event, event_restart,
processor="alpha", system="OSF1V5.1", leap=11, stratum=16,
precision=-11, rootdelay=0.000, rootdispersion=2.385, peer=0,
refid=0.0.0.0, reftime=00000000.00000000 Thu, Feb 7 2036 7:28:16.000,
poll=6, clock=c34a3ade.46d04e61 Wed, Oct 29 2003 14:10:54.276, state=1,
phase=0.000, frequency=0.000, jitter=0.000, stability=0.000

ntptrace -dnv 10.60.20.145 :

[athena.info.latelec.fr-gourdin]/var/adm/syslog.dated/current>ntptrace -dnv 10.60.20.145
DoTransmit(10.60.20.145)
DoTransmit to 10.60.20.145
ReceiveBuf(10.60.20.145, 10.60.20.145)
server 10.60.20.145, port 123
stratum 15, precision -6, leap 00
refid [0.0.0.0] delay 0.00096, dispersion 0.00000 offset 0.000760
rootdelay 0.00000, rootdispersion 0.00000, synch dist 0.00000
reference time: c34a3822.e51eb749 Wed, Oct 29 2003 13:59:14.895
originate timestamp: c34a3afe.cd0e5517 Wed, Oct 29 2003 14:11:26.801
transmit timestamp: c34a3afe.ccbc8c0c Wed, Oct 29 2003 14:11:26.799

DoTransmit(0.0.0.0)
DoTransmit to 0.0.0.0
ReceiveBuf(0.0.0.0, 10.60.10.8)
receive: wrong server
DoTransmit(0.0.0.0)
DoTransmit to 0.0.0.0
ReceiveBuf(0.0.0.0, 10.60.10.8)
receive: wrong server
DoTransmit(0.0.0.0)
DoTransmit to 0.0.0.0
ReceiveBuf(0.0.0.0, 10.60.10.8)
receive: wrong server
DoTransmit(0.0.0.0)
DoTransmit to 0.0.0.0
ReceiveBuf(0.0.0.0, 10.60.10.8)
receive: wrong server
DoTransmit(0.0.0.0)
DoTransmit to 0.0.0.0
ReceiveBuf(0.0.0.0, 10.60.10.8)
receive: wrong server
0.0.0.0: *Timeout*

why timeout ?

ping 10.60.20.145 :

PING 10.60.20.145 (10.60.20.145): 56 data bytes
64 bytes from 10.60.20.145: icmp_seq=0 ttl=128 time=0 ms
64 bytes from 10.60.20.145: icmp_seq=1 ttl=128 time=0 ms
64 bytes from 10.60.20.145: icmp_seq=2 ttl=128 time=0 ms
64 bytes from 10.60.20.145: icmp_seq=3 ttl=128 time=0 ms


----10.60.20.145 PING Statistics----
4 packets transmitted, 4 packets received, 0% packet loss
round-trip (ms) min/avg/max = 0/0/0 ms

Can you help me ?
thanks,

Thomas GOURDIN
System & database administrator
LATelec - Latecoere Group

[thomas gourdin]

no, the timeserver is not a cluster. it's
a simple computer with 2000 pro.
My client is a ds20 with tru64 v5.1 OSF1.
The others clients (hpux 11i,RH 7,8) works
fine with the timeserver.

[Joris Denayer]

Thomas,

What is the OS of the Timeserver ?
If Microsoft Windows
then
check if it uses NTP or SNTP
if SNTP
then
This will not work
fi
fi

SNTP (Simple Network Time Protocol) has a precision of 1 second and this is not precise enough to keep NTP-clients synchronised.

Joris

[thomas gourdin]

yes, the timeserver is a windows 2000 pro
and it uses sntp, so i'll try to configure
another timeserver with linux and ntp.

Thanks.