[ntp:questions] Best practice: re-resolve server IP address before each poll/query

Jan Ceuleers janspam.ceuleers at skynet.be
Wed Jun 20 10:36:54 UTC 2007


Jan Ceuleers wrote:
> [SNTP implementations] "SHOULD resolve the server IP address via
 > DNS before each poll/query, so  that the pertinent DNS entries'
 > Time-To-Live values are respected."
> 
> I was wondering whether the readership of this forum feels that this 
> should also apply to NTP implementations.

Thanks for a stimulating discussion. I note the following (do shout if I 
got it wrong):

- re-resolving server addresses serves the purpose of making clients 
either follow the server if it has moved, or making them stop querying 
the server if it has ceased to be an ntp server.

- the ntpd development team don't think it is a good idea to re-resolve 
before _every_ poll/query, but do think it is a good idea to do this 
regularly. Such behaviour is likely to be included in a future version 
of ntpd.

I was wondering what the impact of such re-resolving behaviour would be 
on the pool. Both cases are relevant, I think: best-practice recommended 
behaviour for sntp implementations as quoted above, and the behaviour 
that is likely to be included in ntpd in the future.

In particular: given the relatively short TTL values used by the pool, 
would this client behaviour not impede the pool's efforts to distribute 
load between pool members within the same "bucket" (e.g. 
europe.pool.ntp.org)?

Thanks, Jan




More information about the questions mailing list