[SOLVED] upgrade from 6.4 to 7 no webgui, ssh access

[tauceti]

After upgrading from 6.4 with opt in kernel 5.11 to 7 and rebooting I cannot reach/ping/access webgui and ssh. Also I cannot ping inside the machine via VNC.
ifupdown2 wasn't installed and the network names were not renamed.
my vmbr0 has the external IP set.
I also tried disabling the firewall and restart networking but it didn't work

It is also not a dns problem as I cannot ping ip addresses outside the internal network in VNC.
I also cannot ping my proxmox server from outside with its external ip address.

I don't get it

pve-firewall status says enabled/stopped

 

[tauceti]

omg I think I got it...
added hwaddress under the vmbr0 bridge with my physical mac address...
strangely it works with ifupdown 1 and 2 was not installed

is that all or do I have todo some other additional config?

 

[t.lamprecht]

added hwaddress under the vmbr0 bridge with my physical mac address...
strangely it works with ifupdown 1 and 2 was not installed

That's expected, it should work with either ifupdown.

is that all or do I have todo some other additional config?

If all is working again for you then that should be enough.

 

[Fischje]

For me, that does not helped me. I'm happy that i can access my machine now cause i commented everything in /etc/network/interfaces and add the public ip at my networkadapter itself.
if i add it to the vmbr0 and/or add the mac under hwadress the machine is not reachable. at kvm-console i see that i can't ping anythig from the machine itself.

i don't understand it.

/etc/network/interfaces

auto lo
iface lo inet loopback

auto eno1
#iface eno1 inet manual
iface eno1 inet static
address xx.xx.xx.xxx/32
gateway xx.xx.xx.xx

auto vmbr0
#iface vmbr0 inet static
#       address xx.xx.xx.xxx/32
#       hwadress YY:YY:YY:YY:YY:YY
#       gateway xx.xx.xx.xx
#       bridge_ports eno1
#       bridge_stp off
#       bridge_fd 0
#       pointopoint xx.xx.xx.xx

#       up sysctl -w net.ipv4.ip_forward=1
#      up sysctl -w net.ipv4.conf.eno1.send_redirects=0

auto vmbr1
iface vmbr1 inet manual
        bridge-ports none
        bridge-stp off
        bridge-fd 0

 

[tauceti]

For me, that does not helped me. I'm happy that i can access my machine now cause i commented everything in /etc/network/interfaces and add the public ip at my networkadapter itself.
if i add it to the vmbr0 and/or add the mac under hwadress the machine is not reachable. at kvm-console i see that i can't ping anythig from the machine itself.

i don't understand it.

/etc/network/interfaces

auto lo
iface lo inet loopback

auto eno1
#iface eno1 inet manual
iface eno1 inet static
address xx.xx.xx.xxx/32
gateway xx.xx.xx.xx

auto vmbr0
#iface vmbr0 inet static
#       address xx.xx.xx.xxx/32
#       hwadress YY:YY:YY:YY:YY:YY
#       gateway xx.xx.xx.xx
#       bridge_ports eno1
#       bridge_stp off
#       bridge_fd 0
#       pointopoint xx.xx.xx.xx

#       up sysctl -w net.ipv4.ip_forward=1
#      up sysctl -w net.ipv4.conf.eno1.send_redirects=0

auto vmbr1
iface vmbr1 inet manual
        bridge-ports none
        bridge-stp off
        bridge-fd 0

There is a typo: hwadress->hwaddress

 

[Fischje]

There is a typo: hwadress->hwaddress

Yes. That was the Problem. German "Adresse".... Thank you so much.

 

[tauceti]

Yes. That was the Problem. German "Adresse".... Thank you so much.

Kenne ich

PS: Hosting provider is netcup...don't know if they only allow network traffic through the defined physical mac address

Thanks for updating the doc with exact steps!!:
https://pve.proxmox.com/wiki/Upgrade_from_6.x_to_7.0#Check_Linux_Network_Bridge_MAC

 

[Josh Douglas]

Having very similar issues after upgrading from 6.4 to 7.0. Adding the "hwaddress" option in /etc/network/interfaces did *not* work for me.

Both ethernet devices show as "UP" in the output of 'ip -c link', however my bond0 (LACP) and vmbr0 devices show as "DOWN".

 

[Astraea]

After I upgraded my cluster from 6 to 7 I did not have any network connectivity. what ended up fixing the issues was logging into each node and running the following command

systemctl enable networking; reboot

After that, the node rebooted and came back online and everything was working again. Not sure why the networking stack got disabled during the update. This is on a 5 node cluster with multiple NICs with and without bonds that was using ifupdown2 before the upgrade.