Re: News./Tcpreset./Net Open for peering

On Sat, 05 Apr 2025 14:27:27 +0200, Gabx wrote:

noel wrote:
 

2a01:4f8:c0c:2f94::1  296.838 ms !<4-1>  296.597 ms !<4-1>  296.303 ms
!
<4-1>
 
 
You have a filter dropping packets, or more likely hetzner has.
 
 

It's not a filter, but it's hetzner:
 
root@news:~# cat /etc/netplan/50-cloud-init.yaml network:
     version: 2 ethernets:
         eth0:
             addresses:
             - 2a01:4f8:c0c:2f94::1/64 dhcp4: true match:
                 macaddress: 52:54:a2:02:2c:94
             nameservers:
                 addresses:
                 - 2a01:4ff:ff00::add:2 - 2a01:4ff:ff00::add:1
             routes:
             -   on-link: true
                 to: default via: fe80::1
             set-name: eth0
 
Is dhcp6 and/or gateway6 missing?

I dont and wont use that format, even on the one single debian machine we
have (freepbx requires debian now) (all other servers are slackware and
much simpler to configure networking)

But our pbx uses in /etc/network/interfaces  after the ipv4 stuff

iface eth0 inet6 static
   address your:servers:six::ip
   netmask 64
   gateway your:servers:six:range::1
   post-up ip -6 route add gwIP::1 dev eth0  (if using fe80::1 not needed)
   dns-nameservers ::1
   accept_ra 0



but fe80::1 is link-local address used for default gateways, as you are
using, and is acceptable as default route.

#  ip -6 a

should see similar to

inet6 your:six::ip/64 scope global

(scope global is important)

#  ip -6 route

should see similar to

::1 dev lo proto kernel metric 256 pref medium
you::six:range::/64 dev eth0 proto kernel metric 256 pref medium
fe80::/64 dev eth0 proto kernel metric 256 pref medium
default via fe80::1 dev eth0 metric 1024 pref medium

Surely hetzner has an FAQ on how to configure ipv6 to suite their network
if they dont auto configure it (which they should)