On a Centos 5.0 box I was having an issue with the tunnel timing out and not routing. Besides needing to fix that, it was easier to create traffic on the tunnel. I needed to create traffic from the server that went outwards - not just a web site being accessed.
ntp6.space.net
ntp.ipv6.linux.ee
noc.sixxs.net
nlams01.sixxs.net
ntp1.bit.nl
ntp2.bit.nl
ntp.doubleukay.com
ntp1.ipv6.teleport-iabg.de
ntp2.ipv6.teleport-iabg.de
I choose to update my NTP config by adding the following Stratum-2 servers. I found the list at http://wiki.go6.net/index.php?title=IPv6_NTP_servers (http://wiki.go6.net/index.php?title=IPv6_NTP_servers) . Don't use the Stratum-1 servers for all the normal reasons.
egrep '^server' /etc/ntp.conf
server 0.rhel.pool.ntp.org
server 1.rhel.pool.ntp.org
server 2.rhel.pool.ntp.org
server ntp6.space.net
server ntp.ipv6.linux.ee
server noc.sixxs.net
server nlams01.sixxs.net
server ntp1.bit.nl
server ntp2.bit.nl
server ntp.doubleukay.com
server ntp1.ipv6.teleport-iabg.de
server ntp2.ipv6.teleport-iabg.de
Note that I left in the standard IPv4 entries from Centos/Redhat. Restarting NTP via:
service ntpd reload
Followed by a simple check of the NTP status:
# ntpq -n -c peers
remote refid st t when poll reach delay offset jitter
==============================================================================
...
-2001:608::1000: 32.246.249.54 2 u 38 64 377 236.486 -5.295 3.735
2001:7d0:0:5:20 .INIT. 16 u - 64 0 0.000 0.000 0.000
*2001:838:1:1:21 81.41.136.168 2 u 42 64 377 228.125 -1.242 0.845
+2001:838:2:1:2a 193.0.0.228 2 u 36 64 377 227.806 -1.244 1.632
+2001:7b8:3:2c:: 193.0.0.228 2 u 41 64 377 227.755 -1.413 2.430
-2001:7b8:3:2d:: 193.0.0.228 2 u 37 64 377 226.925 0.206 0.819
2001:c18:405::1 .INIT. 16 u - 64 0 0.000 0.000 0.000
+2001:1b10:100:3 124.226.83.28 2 u 34 64 377 240.815 -1.495 1.501
-2001:1b10:100:3 73.120.242.92 2 u 35 64 377 240.646 -2.914 3.044
#
Now I have traffic, a tunnel that stays up and the correct time - v6 time!
Enjoy!
i understand this thread posted a while back, but on the off chance the author (or anyone else) is still reading this:
i'm having a problem, pointing my ntp client (Ubuntu) to ipv6 stratum 2 ntp servers, over an HE tunnel. i can even ping6 the target server, but the ntp daemon never gets out of the INIT state for this target.
it syncs, instead, to a v4 server (much farther away).
i've tried restricting the ntp daemon to a subset of interfaces (using the -I eth0 invocation), and verified that the tunnel (virtual) interface is excluded, but still, no joy.
any ideas?
sff