[SOLVED] VLAN issues after upgrading to PVE6

[Ernstl]

Hi all,

some hours ago I did upgrade my PVE5 to PVE6 following all steps in the official upgrade documentation. With PVE5 I used the following network configuration which was working flawlessly.

auto lo
iface lo inet loopback

iface eno1 inet manual

auto vmbr0
iface vmbr0 inet static
address 192.168.1.208
netmask 255.255.255.0
gateway 192.168.1.1
bridge_ports eno1
bridge_stp off
bridge_fd 0

iface vmbr0 inet6 static
address 2a02:2450:1021:9:96c6:91ff:fea9:62d7
netmask 64
gateway fe80::1

auto vmbr1
iface vmbr1 inet static
address 10.111.1.1
netmask 255.255.255.0
broadcast 10.111.1.255
bridge_ports none
bridge_stp off
bridge_fd 0

post-up echo 1 > /proc/sys/net/ipv4/ip_forward
post-up iptables -t nat -A POSTROUTING -s '10.111.1.0/24' -o vmbr0 -j MASQUERADE
post-down iptables -t nat -D POSTROUTING -s '10.111.1.0/24' -o vmbr0 -j MASQUERADE

vmbr0 connects to my "real" LAN while vmbr1 creates a virtual network between a few VMs. Everything was working like expected.

Since the upgrade I am facing the following issue with this configuration:

• the PVE server is not able to ping any other device in my 192.168.1.0 network, not even the gateway at 192.168.1.1
• none of my devices from the 192.168.1.0 network is able to ping the PVE server (so no access to SSH, Webfrontend etc.)
• the PVE server can ping itselfs at 192.168.1.208

After a lot of trial and error I noticed that the situation changes when simple modifying the

bridge_ports eno1

to

bridge_ports none

while leaving the rest of the configuration file unchanged. Now the situation is as follows:

• PVE server is able to access my 192.168.1.0 network
• All devices from the 192.168.1.0 network are able to access the PVE server
• The VMs in vmbr1 can ping each other but none of them can access the internet via the NAT rules
• I have one VM what is directly connected to vmbr0/192.168.1.0: This VM is not able to ping any other device from this network nor to access the internet

I am really running out of ideas how to solve these problems. Especially because everything was working without problems bevor the upgrade. Any help would be appreciated. Thanks in advance.

 

[spirit]

are you sure that eno1 interface name still exist ? (I have seen some bugreports with name rollback to ethX)

 

[Ernstl]

Hi,

yes eno1 is the right interface. After some more hours I finally found the issue: At some point I noticed that the Proxmox server has been reachable for a few seconds after each reboot. And then disappeared. Long story short:

dhcpcd got installed/activated during the upgrade process. And this one was reconfiguring my network interfaces! I did not notice that in the first place, because static IP and leased dhcp IP where identically 192.168.1.208.

systemctl disable dhcpcd

brought my network back to life

Anyway: Thanks for trying to help.