[ntp:hackers] ntp-dev ntp_proto.c receive AM_NEWPASS

Danny Mayer mayer at ntp.org
Sun Jul 5 02:52:26 UTC 2009


Ronan Flood wrote:
>> in case of a wrong configuration for a real NTP node (i.e. using peer 
>> instead of server) I'd say the configuration should be fixed rather than a 
>> workaround added in ntpd.
> 
> It's not a case of adding a workaround, rather that an existing
> workaround has been removed.
> 
> Was there a decision made to break the existing pragmatic behaviour
> ("peer ..." works like "server ..." for an unauthenticated client ntpd),
> or is it an avoidable (even unpredicted) side-effect of the w32time
> change?  If avoidable, I suggest the previous behaviour be reinstated.
> If decided upon, it should be documented.

I'm not sure what you are asking. If you configure a server as a peer
that server also needs to be similarly configured as a peer of this
server. Peers are meant to be backups of each other and operate in
push-pull mode. See Symmetric Active/Passive Mode in Association
Management (assoc.html) for details. Is the documentation there
insufficient? A system like USNO should always be dropping peer requests
and they would not appreciate increased traffic from Windows SNTP
clients (which is what ws32time is), they are heavily loaded enough.

Danny

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



More information about the hackers mailing list