[ntp:questions] Peering and synching over multiple interfaces and subnets.

Steve Kostecke kostecke at ntp.isc.org
Fri Jun 29 14:49:11 UTC 2007

On 2007-06-29, ulf.norberg at banverket.se wrote:

>Danny Meyer wrote:
>>I'm not sure where you think you are getting redundancy since the
>>redundancy comes from picking a sufficient number of servers to
>>provide time service to the SCADA systems. You can skip the routers.
>>They don't need to be involved here.
> I can't skip the routers because they are a part of the NTP solution.

... keep in mind that routers are primarily designed to route packets
in one interface and out another, and they usually have lots of custom
silicon chips to help them perform this role very well and very
quickly. They are not typically well-suited to the role of providing
general-purpose services.

In many cases, these kinds of functions are handed off to an internal
shared CPU which is asked to perform all sorts of less common tasks on
the router, and doing excessive amounts of work with NTP may cause it to
be less able to do 'real work' as a router, or may cause it to perform
poorly as an NTP server.

(from http://support.ntp.org/bin/view/Support/DesigningYourNTPNetwork)

> Because of all of the different subnets in this SCADA network (not
> just around these servers) it is not possible to have the Stratum-1
> servers reachable on each subnet. The security guys won't allow it.


Steve Kostecke <kostecke at ntp.org>
NTP Public Services Project - http://support.ntp.org/

More information about the questions mailing list