Do you have a reference for what you mention about the NTP source being
specified in the "ntp server" statement? I just labbed this up and didn't
see anything like what you describe. What I got looked exactly like I'd
expect.

Topology doesn't matter, suffice to say that R3 has a loopback 3.3.3.3 and
R4 has a loopback 4.4.4.4, with some other networks in the middle.

R3 (server):
ntp master 5

R4 (client):
ntp server 3.3.3.3
ntp source Loopback0


R3 IP Packet Debug:
Jan 12 20:27:25.123: IP: tableid=0, *s=4.4.4.4* (Serial1/3),
*d=3.3.3.3*(Loopback0), routed via RIB
Jan 12 20:27:25.127: IP: s=4.4.4.4 (Serial1/3), d=3.3.3.3, len 76, rcvd 4
Jan 12 20:27:25.131:     UDP src=123, dst=123

R3 is seeing the NTP requests sources from R4 Lo0. If I remove the "ntp
source" statement from R4, it is reflected int the debug. 10.1.124.4 is
R4's Fa0/0 interface, the interface nearest to R3.

R3 IP Packet Debug:
Jan 12 20:28:36.099: IP: tableid=0, *s=10.1.124.4* (Serial1/3),
*d=3.3.3.3*(Loopback0), routed via RIB
Jan 12 20:28:36.103: IP: s=10.1.124.4 (Serial1/3), d=3.3.3.3, len 76, rcvd 4
Jan 12 20:28:36.107:     UDP src=123, dst=123


If I specify the source in the "ntp server" statement, then I can use
distinct sources for different servers:

R4:
ntp server 3.3.3.3
ntp server 1.1.1.1 source Loopback0

R3 Debug:
Jan 12 20:28:36.099: IP: tableid=0, s=*10.1.124.4* (Serial1/3),
d=*3.3.3.3*(Loopback0), routed via RIB
Jan 12 20:28:36.103: IP: s=10.1.124.4 (Serial1/3), d=3.3.3.3, len 76, rcvd 4
Jan 12 20:28:36.107:     UDP src=123, dst=123

R1 Debug:
Jan 12 20:35:43.047: IP: tableid=0, s=*4.4.4.4* (FastEthernet0/0), d=*
1.1.1.1* (Loopback0), routed via RIB
Jan 12 20:35:43.051: IP: s=4.4.4.4 (FastEthernet0/0), d=1.1.1.1, len 76,
rcvd 4
Jan 12 20:35:43.055:     UDP src=123, dst=123


I tried the same with authentication, and it behaved the same. In other
words, the "ntp source" command will apply by default, and the source used
for a specific server can be overridden with the "ntp server x.x.x.x source
lo0" command.

If you've found documentation to the contrary, can you link to it? My lab
is on 12.4(15)T14.


On Fri, Jan 11, 2013 at 12:15 PM, Ryan Krcelic <[email protected]>wrote:

> I have problems remembering the ntp commands, can someone help me validate:
>
> Server with authentication:
> Ntp authenticate ­ only needed on server?
> Ntp trusted-key x
> Ntp authentication-key 1 md5 X
> Ntp source lo0 (sets the source of the ntp updates)
>
>
> Client:
> ***No need for ntp authenticate?
> Ntp authentication-key 1 md5 X
> Ntp trusted-key 1
> ***From what I read, if you want to source ntp updates from a loopback, you
> cannot use the ntp source lo0 command...it has to be done like this***
> Ntp server x.x.x.x source lo0 key 1
>
> So I guess my question is, do you need ntp authenticate on the client and
> will ntp source lo0 work on the client or should it be tied to the ntp
> server command?
>
> V/r,
>
> Ryan Krcelic
>
> _______________________________________________
> For more information regarding industry leading CCIE Lab training, please
> visit www.ipexpert.com
>
> Are you a CCNP or CCIE and looking for a job? Check out
> www.PlatinumPlacement.com
>
> http://onlinestudylist.com/mailman/listinfo/ccie_rs
>
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Are you a CCNP or CCIE and looking for a job? Check out 
www.PlatinumPlacement.com

http://onlinestudylist.com/mailman/listinfo/ccie_rs

Reply via email to