[ntp:questions] Re: requestor's time; transmit timestamp?

Brian Utterback brian.utterback at sun.removeme.com
Mon Nov 28 16:49:40 UTC 2005

roy wrote:

>> I should have also read RFC2030, which says:
>>    To calculate the roundtrip delay d and local clock offset t relative
>>    to the server, the client sets the **transmit timestamp** in the request
>>    to the time of day according to the client clock in NTP timestamp
>>    format. The server copies this field to the originate timestamp in
>>    the reply and sets the receive timestamp and transmit timestamp to
>>    the time of day according to the server clock in NTP timestamp
>>    format.
> There is at least one implementation that does not conform to this
> convention.  Instead of sending the local time in the transmitt
> timestamp, OpenNTPD sets it to a random 64-bit cookie.  See
> http://unduli.bsws.de/papers/21c3/mgp00052.html
> So, if the client is OpenNTPD, the NTP server (and anything else
> looking at the packets) is unable to determine the time of the client
> system.

It is not a convention. It is a requirement of the protocol. Contrary
to the statement made at the URL, it does nothing to improve the
security while at the same time breaking the protocol. Using this
"feature" destroys the symmetric peer mode, which requires that the
time flow both ways.


"Having them stolen may become our distribution model..."
Nicolas Negroponte on the Hundred Dollar Laptop.
Brian Utterback - OP/N1 RPE, Sun Microsystems, Inc.
Ph:877-259-7345, Em:brian.utterback-at-ess-you-enn-dot-kom

More information about the questions mailing list