NTP wouldn't work

karim.benhabyles · ‎09-22-2008

Hi,

I've got two Cisco 2811 routers on my lab with IOS 12.4(20)T. I try to do a simple NTP configuration R5 as NTP server and R4 as NTP client.

R4 and R5 are connected by frame-relay.

here are the configurations:

R4:

interface Serial0/0/0

ip address 155.1.0.4 255.255.255.0

encapsulation frame-relay

ntp broadcast client

frame-relay map ip 155.1.0.5 405 broadcast

no frame-relay inverse-arp

frame-relay lmi-type cisco

!

R5:

interface Serial0/0/0

ip address 155.1.0.5 255.255.255.0

encapsulation frame-relay

ntp broadcast

frame-relay map ip 155.1.0.4 504 broadcast

no frame-relay inverse-arp

!

ntp master 1

!

here are "show ntp status" and "show ntp associations" and "show clock" outputs on both routers:

R4:

Clock is unsynchronized, stratum 16, reference is 155.1.0.5

nominal freq is 250.0000 Hz, actual freq is 249.9997 Hz, precision is 2**24

reference time is 00000000.00000000 (00:00:00.000 UTC Mon Jan 1 1900)

clock offset is 0.6109 msec, root delay is 0.00 msec

root dispersion is 0.01 msec, peer dispersion is 0.00 msec

loopfilter state is 'SPIK' (Spike), drift is 0.000000983 s/s

system poll interval is 64, never updated.

!

address ref clock st when poll reach delay offset disp

* 155.1.0.5 127.127.1.1 8 1 64 377 0.000 574.391 0.867

* sys.peer, # selected, + candidate, - outlyer, x falseticker, ~ configured

!

07:53:19.659 UTC Tue Sep 23 2008

R5:

Clock is synchronized, stratum 1, reference is .LOCL.

nominal freq is 250.0000 Hz, actual freq is 250.0000 Hz, precision is 2**24

reference time is CC831EB1.D3665396 (07:52:17.825 UTC Tue Sep 23 2008)

clock offset is 0.0000 msec, root delay is 0.00 msec

root dispersion is 0.93 msec, peer dispersion is 0.93 msec

loopfilter state is 'CTRL' (Normal Controlled Loop), drift is 0.000000000 s/s

system poll interval is 16, last update was 7 sec ago.

!

address ref clock st when poll reach delay offset disp

*~127.127.1.1 .LOCL. 0 7 16 37 0.000 0.000 437.69

* sys.peer, # selected, + candidate, - outlyer, x falseticker, ~ configured

!

07:53:35.921 UTC Tue Sep 23 2008

Please notice that NTP is configured in broadcast mode. R5 send broadcast and R4 is an ntp broadcast client.

Thanks in advanced. This make me crazy :)

Best Regards,

Karim

Calin C. · ‎09-23-2008

Hello!

I'm not an expert with this, but shouldn't you had a

ntp server "R5_IP"

on your client R4?

Good luck!

Cheers,

Calin

Giuseppe Larosa · ‎09-23-2008

Hello Karim,

the NTP broadcast option should be used out a LAN interface.

So here no practical advantages in using it on a point-to-point link:

only one possible peer is present so no advantage on normal unicast NTP.

I would configure on R5 an NTP source:

ntp source se0/0/0

use this link :

http://www.cisco.com/en/US/docs/ios/12_2/configfun/configuration/guide/fcf012.html#wp1001170

you can also try to debug using the debug ntp commands to see what happens :

if NTP broadcasts are really sent out the se0/0/0 of R5 and if they are received on R4.

Hope to help

Giuseppe