reader.ipv6.xsnews.nl time-out

[ngjvjRbYM]

time-out response:
Ping6 reader.ipv6.xsnews.nl
tracert6 reader.ipv6.xsnews.nl
tracert6 2a02:10:103:9::10
server status : http://www.xsnews.com/ipv6/
This happend for the first time around 12 and 13  april 2009.


good ping6 + tracert6 response from ipv6 adresses:
ipv6.google.com
news.ipv6.eweka.nl
newszilla6.xs4all.nl

My tunnel is using the core1.lon1.he.net.

A tracert6 from 3 HE Routers gives 1 good tracert and 2 bad tracert.

It looks like there is something not right. Can this be fixed?

[broquea]

Thanks for bringing this to our attention, we're contacting networks after seeing some routing oddities coming from them. Hopefully they get this cleared up in a timely manner. I'll post if we get any resolution from them.

[broquea]

Well they changed something, that's for sure. However their return path must be horrible and not going through the peering that is set up, as the latency is pretty bad.. However traces show it reaching now, so try accessing their service.

[ngjvjRbYM]

The ping6 and tracert6 are working again by using core1.lon1.he.net.

Thank you for the support. :)

edit: You are right, the latency is bad. Perhaps this can be fixed in the future.
At least the address is reachable.

[ngjvjRbYM]

I have tested all the "routers" with the "Looking Glass" of your site and found two who can't reach 2a02:10:103:9::10  /  reader.ipv6.xsnews.nl. All the other "routers" are working.

core1.ash1.he.net
Equinix Ashburn (United States)
2001:504:0:2::3549:1

and

core1.pao1.he.net
PAIX Palo Alto (United States)
2001:504:d::37

Perhaps you can fix this also for the users using this route.

[broquea]

We'll ping the peer again and see what they come up with this time. Last time it was announcing a specific from the /32 to a location that didn't have a path back, so,...yeah. Maybe they'll stop breaking themselves.

[ngjvjRbYM]

The latency is good now for reader.ipv6.xsnews.nl.

core1.ash1.he.net and core1.pao1.he.net are working now.

Thanks for the support.