Re: NTP

Hi Andy,

  Thanks for the response.....I foreget to mention in the earlier email,I
did tried using prefer keyword on the BB statement on R3 and found the same
characteristic.Do you think the prefer statement should have overlooked the
offset issue ?

Thanks

Regards
Anantha Subramanian Natarajan
On Mon, Jul 6, 2009 at 1:03 PM, andy Chisholm <andy_at_chisholms.info> wrote:

> Have you tried appending the bb statements with the 'prefer' keyword?
>
> 2009/7/6 Anantha Subramanian Natarajan <anantha.natarajan_at_gravitant.com>
>
>> Hi All,
>>
>> I was working on one of the lab requirements based on NTP and have few
>> questions on the same.The lab requirement ask R3 to take clock from BB3
>> and
>> R6 to take clock from BB1.Also ask to R3 and R6 peer each other to get
>> clock
>> if there respective BB clock goesaway.I used the below configuration on R3
>> and R6
>>
>> R3
>>
>> ntp server <bb3 ip address>
>> ntp peer <r6 ip address>
>>
>> R6
>>
>> ntp server <bb1 ip address>
>> ntp peer <r3 ip address>
>>
>> BBs are giving Stratum 5 clock and so I as assuming both R3 and R6 should
>> see BBs as the preferred server.Is this understanding right ? The actual
>> result which I obtained was on R3,it sees the R6 as preferred clock server
>> and even it doesn't keep the BB3 server as selected backup.So I removed
>> the
>> ntp peer statement on R3 to R6 and found the R3 takes the BB3 clock as
>> server and gets synchronised.So I added back again the ntp peer statement
>> on
>> R3 and found again R3 choosing the R6 and ignoring BB3.I ran debug based
>> on
>> NTP and found the below messages
>>
>> Mar 1 04:18:37.119: NTP: nlist 3, allow 0, found 3, low -7.964615, high
>> 7.962570
>> Mar 1 04:18:37.119: *NTP: 204.12.1.254 discarded, outlier*
>> Mar 1 04:18:37.119: NTP: candidate 183.1.46.6 cdist 87.963593 error
>> 7.893570
>> Mar 1 04:18:37.119: NTP: candidate 150.1.6.6 cdist 95.960403 error
>> 15.891159
>> Mar 1 04:18:37.119: NTP: survivor 183.1.46.6 offset -0.001007, cdist
>> 87.96359
>> Mar 1 04:18:37.119: NTP: survivor 150.1.6.6 offset -0.001109, cdist
>> 95.96040
>> Mar 1 04:18:37.123: NTP: synced to new peer 183.1.46.6
>>
>> It gave message as outlier and tried to understand what is that mean ,my
>> understanding was that if the clock offset is too high when compared to
>> other peers,it may ignore the one which have high offset.I am not sure
>> whether I understood right .Kindly correct me if I am wrong in
>> that.Assuming
>> my understanding on that is right,is not always the high stratum clock
>> wins
>> ?.Does this means it compars the offset before comparing the startum
>> clock.I
>> am sure missing some basic on NTP.
>>
>> How to make in this scenario,R3 prefering BB clock and not the R6 even
>> though the offset for BB clock is comparitvely higher.
>>
>> Thanks for the assistance.
>>
>> Regards
>> Anantha Subramanian Natarajan
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Mon Jul 06 2009 - 13:14:11 ART