IPv6 ULA default gateway and post-up ignored?

QaMT4kJZP7

Member
Aug 23, 2023
7
0
6
Hello,

I was wandering if anyone could assist with an issue I'm currently facing.

I am trying to add IPv6 support to my PVE Cluster, basically I would like to add ULAs to each note, because I want them to be available via IPv6. For this I used my eno1 interface and added IP + Gateway, but the Gateway as a default route seems to be ignored, because it is always using a different interface Gateway (which I have not set), I also tried adding post-up to the interface with no success.

The Output of /etc/network/interfaces is:
Code:
# network interface settings; autogenerated
# Please do NOT modify this file directly, unless you know what
# you're doing.
#
# If you want to manage parts of the network configuration manually,
# please utilize the 'source' or 'source-directory' directives to do
# so.
# PVE will preserve these directives, but will NOT read its network
# configuration from sourced files, so do not attempt to move any of
# the PVE managed interfaces into external files!

auto lo
iface lo inet loopback

auto eno1
iface eno1 inet static
        address 10.10.5.21/24
        gateway 10.10.5.1
#Management - 1Gbit/s

iface eno1 inet6 static
        address fd00:dead:c0de:5:4e52:0000:0000:2411/64
        gateway fd00:dead:c0de:5::1
        post-up ip -6 route add default via fd00:dead:c0de:5::1 dev eno1

iface enp134s0f2 inet6 manual

iface enp134s0f3 inet6 manual

auto enp24s0f0
iface enp24s0f0 inet6 static
        address fd00:dead:c0de:23:4e52:0000:0000:2411/64
        mtu 9000
#PVECephPublic - 10Gbit/s

auto enp24s0f1
iface enp24s0f1 inet6 static
        address fd00:dead:c0de:25:4e52:0000:0000:2411/64
        mtu 9000
#PVECephCluster - 10Gbit/s

iface enp94s0f0 inet6 manual

iface enp94s0f1 inet6 manual

iface eno3np0 inet manual
#Network Infrastructure - 1Gbit/s

auto eno2
iface eno2 inet6 static
        address fd00:dead:c0de:21:4e52:0000:0000:2411/64
#PVECorosync - 1Gbit/s

iface eno4np1 inet manual
#DOCKER-PVE - 1Gbit/s

iface eno5np2 inet6 manual

iface enp134s0f0 inet6 manual

iface eno6np3 inet6 manual

iface enp134s0f1 inet6 manual

auto vmbr_infra
iface vmbr_infra inet manual
        bridge-ports eno3np0
        bridge-stp off
        bridge-fd 0
#Network Infrastructure - 1Gbit/s

auto vmbr_dock
iface vmbr_dock inet manual
        bridge-ports eno4np1
        bridge-stp off
        bridge-fd 0
#DOCKER-PVE - 1Gbit/s

source /etc/network/interfaces.d/*

In this case it is using vmbr_dock
Code:
traceroute to 2a01:4f8:c014:437c::1 (2a01:4f8:c014:437c::1), 30 hops max, 80 byte packets
 1  unifi.localdomain (fd00:dead:c0de:100::1)  0.301 ms  1.806 ms  1.798 ms
 2  2003:0:0000:b000::1 (2003:0:0000:b000::1)  2.099 ms  2.124 ms  2.106 ms
 3  * * *
 4  2003:0:0000::a1d (2003:0:0000::a1d)  6.924 ms  6.857 ms  6.840 ms
 5  core21.fsn1.hetzner.com (2a01:4f8:0:3::4d5)  11.449 ms core23.fsn1.hetzner.com (2a01:4f8:0:3::4b5)  11.738 ms  11.741 ms
 6  * * *
 7  spine3.cloud2.fsn1.hetzner.com (2a01:4f8:0:e172::a152)  13.368 ms spine3.cloud2.fsn1.hetzner.com (2a01:4f8:0:e172::a14e)  12.245 ms spine4.cloud2.fsn1.hetzner.com (2a01:4f8:0:e172::a15a)  12.155 ms
 8  * * *
 9  32218.your-cloud.host (2a01:4f8:0:e172::121f)  11.799 ms  11.439 ms  11.269 ms
10  srv01.vpn64.de (2a01:4f8:c014:437c::1)  14.086 ms  11.431 ms  11.155 ms

I guess it is picking up some DHCP or RA configurations, because DHCPv6 incl. RA is running on fd00:dead:c0de:100::1/64
Code:
# ip -6 route show
fd00:dead:c0de:5::/64 dev eno1 proto kernel metric 256 pref medium
fd00:dead:c0de:21::/64 dev eno2 proto kernel metric 256 pref medium
fd00:dead:c0de:23::/64 dev enp24s0f0 proto kernel metric 256 pref medium
fd00:dead:c0de:25::/64 dev enp24s0f1 proto kernel metric 256 pref medium
fd00:dead:c0de:100::/64 dev vmbr_dock proto kernel metric 256 expires 86013sec pref medium
fe80::/64 dev enp24s0f0 proto kernel metric 256 pref medium
fe80::/64 dev enp24s0f1 proto kernel metric 256 pref medium
fe80::/64 dev vmbr_dock proto kernel metric 256 pref medium
fe80::/64 dev eno1 proto kernel metric 256 pref medium
fe80::/64 dev eno2 proto kernel metric 256 pref medium
default via fd00:dead:c0de:5::1 dev eno1 proto kernel metric 1024 onlink pref medium
default via fe80::6222:0000:0000:b0ab dev vmbr_dock proto ra metric 1024 expires 1413sec hoplimit 64 pref high
default via fe80::6222:0000:0000:b0ab dev enp24s0f1 proto ra metric 1024 expires 1422sec mtu 1500 hoplimit 64 pref low
default via fe80::6222:0000:0000:b0ab dev enp24s0f0 proto ra metric 1024 expires 1414sec mtu 1500 hoplimit 64 pref low
default via fe80::6222:0000:0000:b0ab dev eno1 proto ra metric 1024 expires 1427sec hoplimit 64 pref high
default via fe80::6222:0000:0000:b0ab dev eno2 proto ra metric 1024 expires 1419sec hoplimit 64 pref high

Tracroute is currently working, because I allowed Internet Access to fd00:dead:c0de:100::1/64, but that was not the plan.
Any idea how to solve this?

Many Thanks,
Jonas