If you use the "server" directive, the first response back from the resolver 
library will be used. Most dual-stack-enabled OSs will prefer the IPv6 one 
among that group if the host has a valid IPv6 configuration. If you don't reach 
it, you're out of luck. The application has no intrinsic knowledge of the 
"state" of the resolver's configuration. 

If you have an IPv4-only host(i.e. v6 disabled), your resolver library will not 
ask for or give give ntpd the ipv6 address, but rather the first ipv4 that it 
gets. 

If you use the "pool" directive, however, it'll load ALL of the hosts into 
runtime and attempt to connect to each of them, after which it decides which is 
be best ticker and uses it (either keeping the others in reserve or marking 
them unreachable depending on the reachability).

Dan

Dan Geist dan(@)polter.net
(33.942962,-84.312118)
http://www.polter.net


----- Original Message -----
> From: "Franck Martin" <[email protected]>
> To: [email protected]
> Sent: Thursday, June 23, 2016 4:27:27 PM
> Subject: [Pool] NTP failover to IPv4 when IPv6 is enabled

> If I have
> 
> /etc/ntp.conf
> server ntp1.example.com
> 
> and in DNS
> ntp1.example.com A 10.0.0.1
> ntp1.example.com A  10.0.0.2
> ntp1.example.com AAAA 10::1
> 
> ntp will pick up the AAAA record and if not successful to communicate with
> it, it will not try the other records (ie A)?
> _______________________________________________
> pool mailing list
> [email protected]
> http://lists.ntp.org/listinfo/pool
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to